NCP Nutrition in Clinical Practice

Volume 33 • Number 6 • December 2018

IMPLICATIONS OF MEDICAL
TECHNOLOGY ON NUTRITION
SUPPORT
Nutrition Support in Adult Patients Receiving
Extracorporeal Membrane Oxygenation

Nutrition Support During Pediatric Extracorporeal

Membrane Oxygenation

Metabolic Support of the Patient on Continuous

Renal Replacement Therapy

Use of Intradialytic Parenteral Nutrition in Patients

Undergoing Hemodialysis

Exploring the Potential Effectiveness of Combining

Optimal Nutrition With Electrical Stimulation to Maintain
Muscle Health in Critical Illness: A Narrative Review

Technology in Parenteral Nutrition Compounding

Standards for Nutrition Support: Adult Hospitalized

Patients

Pediatric Nasogastric Tube Placement

and Verification: Best Practice Recommendations
From the NOVEL Project

wileyonlinelibrary.com/journal/ncp
Peer-reviewed, practical solutions in clinical nutrition ISSN: 0884-5336

NCP_33_6_cover.indd 1 31/10/18 3:45 PM

 Nutrition in Clinical Practice
EDITOR-IN-CHIEF
Jeanette M. Hasse, PhD, RD, LD, FADA, CNSC
Dallas, TX
ASSOCIATE EDITORS
Roland N. Dickerson, PharmD, BCNSP Mary S. McCarthy, PhD, RN, CNSC, FAAN
Memphis, TN Tacoma, WA
Mary Marian, DCN, RDN, CSO, FAND Jayshil Patel, MD
Tucson, AZ Milwaukee, WI

CONTRIBUTING EDITOR
Ryan T. Hurt, MD, PhD
Rochester, MN

ASPEN STAFF
Colleen K. Harper, CAE Catherine J. Klein, PhD
Interim Chief Executive Officer Publications Manager

EDITORIAL BOARD
Teruyoshi Amagai, MD, PhD Ronald L. Koretz, MD Carol J. Rollins, MS, RD, CNSC,
Nishinomiya, Japan Granada Hills, CA PharmD, BCNSP
Tucson, AZ
Deborah A. Andris, MSN, APNP Debra S. Kovacevich, RN, MPH
Milwaukee, WI Ann Arbor, MI Mary K. Russell, MS, RDN, LDN
David A. August, MD, FACS, CNSP Kenneth A. Kudsk, MD Chicago, IL
New Brunswick, NJ Madison, WI
Denise Baird Schwartz, MS, RD,
Albert Barrocas, MD, FACS Laura E. Matarese, PhD, RD, LDN, CNSC, FADA, FAND, FASPEN
East Point, GA FADA, FASPEN Studio City, CA
David L. Burns, MD, FACG Greenville, NC
David S. Seres, MD, ScM, PNS
Burlington, MA Carol McGinnis, DNP, RN, CNS, CNSC New York, NY
Yvon Carpentier, MD Sioux Falls, SD
Brussels, Belgium Pierre Singer, MD
Remy Meier, MD Tel Aviv, Israel
Isabel Correia, MD, PhD Liestal, Switzerland
Minas Gerais, Brazil Annalynn Skipper, PhD, RD, FADA
Sarah J. Miller, PharmD, BCNSP
Chicago, IL
Gail Cresci, PhD, RD, LD, CNSC Missoula, MT
Cleveland, OH Krishnan Sriram, MD, FRCS(C),
Charles M. Mueller, PhD, RD, CNSC, CDN
Robert S. DeChicco, MS, RD, New York, NY FACS, FCCM
LD, CNSC Chicago, IL
Cleveland, OH Gerard E. Mullin, MD, CNSC
Baltimore, MD Dan Waitzberg, MD, PhD, FASPEN
Rose Ann DiMaria-Ghalili, PhD, RN, CNSC São Paulo, Brazil
Philadelphia, PA Lynne M. Murphy, MSN, RN
Washington, DC Jacqueline Wessel, MEd, RDN, CNSC, CSP,
Beverly J. Holcombe, PharmD, BCNSP,
Luis Nin, MD CLE, LD
FASHP
Montevideo, Uruguay Cincinnati, OH
Ocean Isle Beach, NC
Caroline M. Kiss, RD, DCN Kim Robien, PhD, RD, CSO Patricia H. Worthington, RN, MSN, CNSC
Basel, Switzerland Washington, DC Philadelphia, PA
NUTRITION IN CLINICAL PRACTICE (Print ISSN: 0884-5336; Online ISSN: 1941-2452) is published bimonthly in
February, April, June, August, October, and December on behalf of the American Society for Parenteral and Enteral Nutrition
(ASPEN) by Wiley Subscription Services, Inc., a Wiley Company, 111 River Street, Hoboken, NJ 07030-5774 USA. Periodicals
postage paid at Hoboken, NJ and additional mailing offices. Postmaster: Send all address changes to Nutrition in Clinical Practice,
John Wiley & Sons Inc., C/O The Sheridan Press, PO Box 465, Hanover, PA 17331 USA.
COPYRIGHT  C 2018 by ASPEN. All rights reserved. No part of this publication may be reproduced, stored or transmitted

in any form or by any means without the prior permission in writing from the copyright holder. Authorization to copy
items for internal and personal use is granted by the copyright holder for libraries and other users registered with their local
Reproduction Rights Organization (RRO), e.g. Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923,
USA (www.copyright.com), provided the appropriate fee is paid directly to the RRO. This consent does not extend to other kinds
of copying such as copying for general distribution, for advertising or promotional purposes, for republication, for creating new
collective works or for resale. Permissions for such reuse can be obtained using the RightsLink “Request Permissions” link on
Wiley Online Library. Special requests should be addressed to: permissions@wiley.com
INFORMATION FOR SUBSCRIBERS Nutrition in Clinical Practice is published in 6 issues per year. Institutional subscription
prices for 2018 are:
Print & Online: US$607 (US and Rest of World), €563 (Europe), £470 (UK).
Prices are exclusive of tax. Asia-Pacific GST, Canadian GST/HST and European VAT will be applied at the appropriate rates. For
more information on current tax rates, please go to www.wileyonlinelibrary.com/tax-vat. The price includes online access to the
current and all online backfiles to January 1st 2014, where available. For other pricing options, including access information and
terms and conditions, please visit www.wileyonlinelibrary.com/access.
DELIVERY TERMS AND LEGAL TITLE Where the subscription price includes print issues and delivery is to the recipient’s
address, delivery terms are Delivered at Place (DAP); the recipient is responsible for paying any import duty or taxes. Title to
all issues transfers Free of Board (FOB) our shipping point, freight prepaid. We will endeavour to fulfil claims for missing or
damaged copies within six months of publication, within our reasonable discretion and subject to availability.
BACK ISSUES Single issues from current and recent volumes are available at the current single issue price from cs-
journals@wiley.com.
DISCLAIMER The Publisher, ASPEN and Editors cannot be held responsible for errors or any consequences arising from the
use of information contained in this journal; the views and opinions expressed do not necessarily reflect those of the Publisher,
ASPEN and Editors, neither does the publication of advertisements constitute any endorsement by the Publisher, ASPEN and
Editors of the products advertised.
PUBLISHER Nutrition in Clinical Practice is published by Wiley Periodicals, Inc., 350 Main St., Malden, MA 02148. E-mail:
cs-journals@wiley.com
JOURNAL CUSTOMER SERVICES: For ordering information, claims and any enquiry concerning your journal subscription
please go to www.wileycustomerhelp.com/ask or contact your nearest office.
Americas: Email: cs-journals@wiley.com; Tel: +1 781 388 8598 or +1 800 835 6770 (toll free in the USA & Canada).
Europe, Middle East and Africa: Email: cs-journals@wiley.com; Tel: +44 (0) 1865 778315. Asia Pacific: Email: cs-
journals@wiley.com; Tel: +65 6511 8000.
Japan: For Japanese speaking support, Email: cs-japan@wiley.com.
Visit our Online Customer Help available in 7 languages at www.wileycustomerhelp.com/ask
PRODUCTION EDITOR email: NCPprod@wiley.com.
Wiley’s Corporate Citizenship initiative seeks to address the environmental, social, economic, and ethical challenges faced in our
business and which are important to our diverse stakeholder groups. Since launching the initiative, we have focused on sharing
our content with those in need, enhancing community philanthropy, reducing our carbon impact, creating global guidelines and
best practices for paper use, establishing a vendor code of ethics, and engaging our colleagues and other stakeholders in our
efforts. Follow our progress at www.wiley.com/go/citizenship
Wiley is a founding member of the UN-backed HINARI, AGORA, and OARE initiatives. They are now collectively known as
Research4Life, making online scientific content available free or at nominal cost to researchers in developing countries. Please
visit Wiley’s Content Access – Corporate Citizenship site: http://www.wiley.com/WileyCDA/Section/id-390082.html
American Society for Parenteral and Enteral Nutrition (ASPEN), 8401 Colesville Road, Suite 510, Silver Spring, MD 20910.
Printed in the USA by The Sheridan Group.
Nutrition in Clinical Practice accepts articles for Open Access publication. Please visit http://olabout.wiley.com/WileyCDA/
Section/id-828081.html for further information about OnlineOpen.

This journal is a member of the Committee on Publication Ethics (COPE) (http://publicationethics.org/about)

 Nutrition in Clinical Practice
Volume 33 Issue 6 December 2018

Editorial
Editor’s Note 736
Jeanette M. Hasse, PhD, RD, LD, FADA, CNSC

Invited Reviews
Nutrition Support in Adult Patients Receiving Extracorporeal Membrane Oxygenation 738
Danielle E. Bear, RD, MRES; Elizabeth Smith, RD, MSc; and Nicholas A. Barrett, FCICM
Nutrition Support During Pediatric Extracorporeal Membrane Oxygenation 747
Bethany J. Farr, MD; Samuel E. Rice-Townsend, MD; and Nilesh M. Mehta, MD
Metabolic Support of the Patient on Continuous Renal Replacement Therapy 754
Erin M. Nystrom, PharmD, BCNSP and Andrea M. Nei, PharmD, BCPS, BCCCP
Use of Intradialytic Parenteral Nutrition in Patients Undergoing Hemodialysis 767
Menaka Sarav, MD and Allon N. Friedman, MD
Exploring the Potential Effectiveness of Combining Optimal Nutrition With Electrical Stimulation to Maintain 772
Muscle Health in Critical Illness: A Narrative Review
Selina M. Parry, PT, PhD; Lee-anne S. Chapple, PhD; and Marina Mourtzakis, PhD
Methods of Enteral Nutrition Administration in Critically Ill Patients: Continuous, Cyclic, Intermittent, 790
and Bolus Feeding
Satomi Ichimaru, PhD, RD, CNSC
Technology in Parenteral Nutrition Compounding 796
Caitlin Curtis, PharmD, BCNSP

Reviews
Effect of Early vs Late Start of Oral Intake on Anastomotic Leakage Following Elective Lower Intestinal Surgery: 803
A Systematic Review
Boudewijn J. J. Smeets, MD; Emmeline G. Peters, MD; Eelco C. J. Horsten, BSc; Teus J. Weijs, PhD;
Harm J. T. Rutten, PhD; Willem A. Buurman, PhD; Wouter J. de Jonge, PhD; and Misha D. P. Luyer, PhD
The Prognostic Role of Phase Angle in Advanced Cancer Patients: A Systematic Review 813
Mayane Marinho Esteves Pereira, BS; Mariana dos Santos Campello Queiroz, BS;
Nathália Masiero Cavalcanti de Albuquerque, BS; Juliana Rodrigues, PhD;
Emanuelly Varea Maria Wiegert, MD; Larissa Calixto-Lima, MD; and Livia Costa de Oliveira, PhD

Clinical Research
Effect of Vitamin D Level on Clinical Outcomes in Patients Undergoing Left Ventricular Assist Device Implantation 825
Fadi Abou Obeid, MD; Gardner Yost, MS; Geetha Bhat, PhD, MD; Erin Drever, MD; and Antone Tatooles, MD
Diagnostic Accuracy of Bioelectrical Impedance Analysis Parameters for the Evaluation of Malnutrition in Patients 831
Receiving Hemodialysis
Angela Teodósio da Silva, MSc; Daniela Barbieri Hauschild, MSc; Yara Maria Franco Moreno, PhD;
João Luiz Dornelles Bastos, PhD; and Elisabeth Wazlawik, PhD
Comparison of Two Methods for Estimating the Tip Position of a Nasogastric Feeding Tube: A Randomized 843
Controlled Trial
Tim Torsy, RN, MSc; Renée Saman, RN, CNS, MSc; Kurt Boeykens, RN, CNS, MSc; Ivo Duysburgh, MD;
Nele Van Damme, RN, MSc; and Dimitri Beeckman, RN, MSc, PhD
Long-Term Use of Mixed-Oil Lipid Emulsion in Adult Home Parenteral Nutrition Patients: A Case Series 851
Manpreet S. Mundi, MD; Megan T. McMahon, PA-C; Jennifer J. Carnell, PharmD; and Ryan T. Hurt, MD, PhD
Nutrition Status Among HIV-Positive and HIV-Negative Inpatients with Pulmonary Tuberculosis 858
Tássia Kirchmann Lazzari, MSc; Gabriele Carra Forte, PhD; and Denise Rossato Silva, MD, PhD
Associations Between Enteral Nutrition and Acute Respiratory Infection Among Patients in New York Metropolitan 865
Region Pediatric Long-Term Care Facilities
Marissa Burgermaster, PhD; Meghan Murray, MPH, RN; Lisa Saiman, MD, MPH;
David S. Seres, MD, ScM, PNS, FASPEN; and Elaine L. Larson, RN, PhD, CIC, FAAN
Reducing Blood Glucose Testing Is Safe in Patients Receiving Parenteral Nutrition 872
Ashley Ratliff, MS, RD, LD, CNSC; Amy Nishnick, RD, LD, CNSC; Robert DeChicco, MS, RD, LD, CNSC;
and Rocio Lopez, MS, MPH
Risk of Malnutrition Evaluated by Mini Nutritional Assessment and Sarcopenia in Noninstitutionalized Elderly People 879
Ilaria Liguori, MD; Francesco Curcio, MD; Gennaro Russo, MD; Michele Cellurale, MD; Luisa Aran, MD;
Giulia Bulli, MD; David Della-Morte, MD, PhD; Gaetano Gargiulo, MD; Gianluca Testa, MD, PhD;
Francesco Cacciatore, MD, PhD; Domenico Bonaduce, MD; and Pasquale Abete, MD, PhD
Height Prediction From Ulna Length of Critically Ill Patients 887
Micheli S. Tarnowski, RD; Estela I. Rabito, RD, PhD; Daieni Fernandes, RD, MSc; Mariane Rosa, RD;
Manoela L. Oliveira, RD; Vânia N. Hirakata, MSc; and Aline Marcadenti, RD, PhD

Clinical Observations
Venovenous Extracorporeal Membrane Oxygenation in an Adult Patient With Prader-Willi Syndrome: 893
A Nutrition Case Report
Stacy Pelekhaty, RD, LDN, CNSC and Jay Menaker, MD
Stoned—A Syndrome of D-Lactic Acidosis and Urolithiasis 897
Casey M. Berman, MD and Russell J. Merritt, MD, PhD
Avoidance of Overt Precipitation and Patient Harm Following Errant Y-Site Administration of Calcium Chloride 902
and Parenteral Nutrition Compounded With Sodium Glycerophosphate
Collin Anderson, PharmD, PhD, BCPS, BCPPS; Chanelle Stidham, PharmD, BCPS;
Sabrina Boehme, PharmD, BCPS, BCPPS; and Jared Cash, PharmD, BCPS

Standards of Practice
Standards for Nutrition Support: Adult Hospitalized Patients 906
Andrew Ukleja, MD, AGAF; Karen Gilbert, RN, MSN, CNSC, ACNP; Kris M. Mogensen, MS, RD-AP, LDN, CNSC;
Renee Walker, MS, RD, LD, CNSC, FAND; Ceressa T. Ward, PharmD, BCPS, BCNSP, BCCCP;
Joe Ybarra, PharmD, BCNSP; Beverly Holcombe, PharmD, BCNSP, FASHP, FASPEN; and Task Force on Standards for
Nutrition Support: Adult Hospitalized Patients, the American Society for Parenteral and Enteral Nutrition

Special Report
Pediatric Nasogastric Tube Placement and Verification: Best Practice Recommendations From the NOVEL Project 921
Sharon Y. Irving, PhD, CRNP, FCCM, FAAN; Gina Rempel, MD, FRCPC; Beth Lyman, RN, MSN, CNSC, FASPEN;
Wednesday Marie A. Sevilla, MD, MPH, CNSC; LaDonna Northington, DNS, RN, BC; Peggi Guenter, PhD, RN, FAAN,
FASPEN; and The American Society for Parenteral and Enteral Nutrition

Cover Art Note

This issue of Nutrition in Clinical Practice is devoted to medical technology that can be used to advance nutrition support or
other aspects of medical care thereby impacting nutrition therapies.
Cover art credit: everythingpossible 
C 123RF.COM

The ASPEN Rhoads Research Foundation depends on gifts from caring individuals to continue its work. Honor the memory of
a loved one or celebrate a special occasion by giving a gift to the ASPEN Rhoads Research Foundation.

We have now made it easier to support the foundation by providing a secure online donation form on our website:
www.nutritioncare.org. In addition to the online donation form, you can also charge your gift to a Visa, Mastercard, or American
Express by phoning our office, or you can mail a personal check directly to the ASPEN Rhoads Research Foundation.

Editor’s Note
Medical technology continues to be introduced into clin-
ical practice at a rapid pace. The use of some medical
technologies may affect when, what, and how we feed our
patients. Other technologies may be used to help nutrition
support practitioners individualize nutrition support for
their patients. This issue of Nutrition in Clinical Practice
features articles on advances in medical technology and
how they affect nutrient needs and the delivery of nutrition
support.
Recent advances in the use of extracorporeal membrane
oxygenation (ECMO) have expanded its use in intensive
care units (ICUs) for adults and children with severe
respiratory or circulatory failure. A review by Danielle
Bear and colleagues focuses on delivery of enteral nu-
trition support and overcoming barriers to feeding adult
patients undergoing ECMO. In a related article, Farr,
Rice-Townsend, and Mehta address special considerations
for feeding critically ill infants and children undergoing
ECMO and how to optimize their weight gain and growth
over time. For additional information on this topic, be
sure to listen to the podcasts with Danielle Bear and
Dr. Samuel Rice-Townsend discussing nutrition issues re-
lated to ECMO in adult and pediatric populations. In
addition, the article by Bear and colleagues is the CE article
for this issue.
Continuous renal replacement therapy (CRRT) is often
utilized in critically ill patients with renal failure. Infor-
mation on the CRRT prescription and implications for
nutrition support was reviewed by Erin Nystrom, who
delves into replacement and dialysate fluids and provides
a case example and an algorithm for estimating how
these solutions and anticoagulation treatment may impact
caloric delivery. Sarav and Friedman undertook a review
of the benefits and continuing controversies of providing
intravenous infusion of essential nutrients to malnour-
ished patients during hemodialysis treatments, known as
intradialytic parenteral nutrition (IDPN). Although its
long-term nutrition benefits need further study, IDPN is
one option for supplementing nutrition in patients re-
quiring hemodialysis who have poor nutrient intake. The
Nutrition in Clinical Practice
Volume 33 Number 6
December 2018 736

C 2018 American Society for
Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10220
wileyonlinelibrary.com
authors outline some of the benefits and risks of this
therapy.
Parry and colleagues introduce the concept of using
electrical muscle stimulation in critically ill patients. The
authors describe skeletal muscle atrophy that occurs in
critical and prolonged illness and then review the principles
and potential effects of electrical muscle stimulation in
preserving muscle health when combined with adequate
nutrition.
Other technological improvements are seen in
parenteral nutrition (PN) compounding. In a review
of PN compounding, Caitlin Curtis discusses how
technological improvements such as barcode-assisted
medication preparation systems and electronic health
record (EHR)-to-compounder interfaces provide better
safeguards and encourages planning for back-up systems
in the event of EHR failures or electric outages.
There are 2 American Society for Parenteral and Enteral
Nutrition (ASPEN) documents in this issue. Important
consensus recommendations from the ASPEN NOVEL
project are published in this issue; these recommendations
describe best practices for placing nasogastric tubes (NGT)
in pediatric patients and verifying location to avoid pa-
tient harm. Also in this issue are the updated ASPEN
“Standards for Nutrition Support for Adult Hospitalized
Patients”. These practice-based standards are intended for
use by healthcare professionals charged with the care of
patients in any hospital with or without a formal nutrition
support service or team. The Standards address professional
responsibilities as they relate to patient assessment, diag-
nosis, education, care plan development, implementation,
clinical monitoring, evaluation, and other issues of nutrition
support such as nutrition therapy at end-of-life.
Jeanette M. Hasse, PhD, RD, LD, FADA, CNSC
Editor-in-Chief, Nutrition in Clinical Practice
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Nutrition Support in Adult Patients Receiving Extracorporeal December 2018 738–746

C 2018 American Society for

Membrane Oxygenation Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10211
wileyonlinelibrary.com

Danielle E. Bear, RD, MRES1,2,3,4 ; Elizabeth Smith, RD, MSc1,2 ;

2,3,4
and Nicholas A. Barrett, FCICM

Abstract
The use of extracorporeal membrane oxygenation (ECMO) for both severe respiratory and cardiac failure is increasing. Because
these patients are some of the sickest in the intensive care unit, a multidisciplinary approach to their treatment, including appropriate
nutrition therapy, is warranted. Currently, limited data exist on the optimal timing, type, and amount of nutrition to be provided.
This review focuses on describing the current nutrition practices in patients receiving ECMO, details research that is currently being
undertaken, and lists important research questions that require exploration in this field. Observational data suggest that early enteral
nutrition is safe and that although nutrition targets can be met, underfeeding is still common. Until further research is available,
these patients should be fed according to guidelines for the general critically unwell population. (Nutr Clin Pract. 2018;33:738–746)

Keywords
critical illness; enteral nutrition; extracorporeal membrane oxygenation; nutrition assessment; nutrition requirements; nutrition
support

Introduction (CESAR) study.2 ECMO has 2 configurations: venovenous

Extracorporeal membrane oxygenation (ECMO) is a life-
saving supportive therapy used for patients with severe respi-
ratory and/or cardiac failure. ECMO for adult cardiorespi-
ratory failure has seen significant growth in the last decade,
following the influenza A (H1N1) pandemic of 2009–20111
and the conventional ventilatory support vs extracorporeal
membrane oxygenation for severe adult respiratory failure
(VV), which is used for respiratory support, and veno-
arterial (VA), which is used for cardiac support. To provide
ECMO, blood is drained from the great veins and passed
through a gas exchange membrane that adds oxygen and
removes carbon dioxide; the blood is then returned to the
circulation into either the right atrium (VV ECMO) or aorta
(VA ECMO).3,4

From the 1 Department of Nutrition and Dietetics, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; 2 Department of
Critical Care, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; 3 Lane Fox Research Clinical Respiratory Physiology
Research Centre, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; and 4 Centre for Human and Applied Physiological
Sciences, King’s College London, London, United Kingdom.
Financial disclosure: D.E.B. is funded by a Health Education England/National Institute for Health Research (NIHR) ICA Clinical Doctoral
Fellowship (ICA-CDRF-2015-01-047). The views expressed are those of the authors and are not necessarily those of the National Health Service
(NHS), the NIHR, or the Department of Health.
Conflicts of interest: None declared.
This article originally appeared online on October 15, 2018.

Podcast available
Listen to a discussion of this manuscript with NCP Editor-in-Chief Jeanette M. Hasse, PhD, RD, LD, FADA, CNSC, and author Danielle E.
Bear, RD, MRES. This and other NCP podcasts are available at: https://onlinelibrary.wiley.com/page/journal/19412452/homepage/podcasts

Corresponding Author:
Danielle E. Bear, RD, MRES, HEE/NIHR Clinical Doctoral Fellow and Critical Care Dietitian, St Thomas’ Hospital, Westminster Bridge Road,
London SE1 7EH, United Kingdom.
Email: danielle.bear@gstt.nhs.uk
Bear et al
Despite the fact that these patients are some of the sickest
and most resource-intensive patients in the intensive care
unit (ICU), there is a paucity of data regarding the optimum
timing, type, and amount of nutrition support that should
be provided. Indeed, the Extracorporeal Life Support Or-
ganization, the umbrella organization of ECMO providers,
includes only 1 line regarding nutrition support in their
Guidelines for Adult Respiratory Failure, which states “full
energy and protein support is essential.”5 This lack of
guidance is reflected in the significant variation in feed-
ing practices reported both in the United Kingdom and
internationally.6,7 However, it is acknowledged that investi-
gating nutrition support practices to improve outcome and
recovery in these patients is warranted.8
There is much debate surrounding optimum nutrition
support practices for the general critically ill patient; how-
ever, the importance of avoiding overfeeding in the early
stages of critical illness is agreed on.9 Given patients re-
ceiving ECMO are some of the most acutely unwell in the
ICU, this point may be even more relevant, and differences
in responses to feeding strategies between patients receiving
VV and VA ECMO may be present.
We aim to provide a comprehensive review of the
evidence surrounding the provision and management of
nutrition support in adult patients receiving ECMO, and
provide some questions to direct future research in this area.
Timing and Route of Nutrition Support
Concerns about the safety of early enteral nutrition (EN)
in critically ill patients exist, particularly in those patients
who are in significant states of shock (inadequate organ
perfusion) and receiving high-dose vasopressors (commonly
adrenergic drugs).9 For these reasons, historical practices
encouraged withholding, or at least delaying, EN during
ECMO support.10 However, recent European guidelines11
have recommended that early EN is safe and feasible in these
patients, albeit based on expert opinion.
Seven observational studies were found describing nu-
trition practices in patients receiving ECMO (both VV
and VA).10,12-17 In these studies (Table 1), EN was most
frequently commenced within 24 hours of starting ECMO,
with several studies reporting a mean or median time to
feeding of around 13 hours. In the largest study,16 96%
of patients were fed within 24 hours. The reporting of
adverse events is rare and likely to be impacted on by the
predominantly retrospective nature of the studies, but in a
prospective multicenter observational study of 107 patients,
5 patients (4.5%) were reported to have bowel ischemia.15
This was a population of 60% VA ECMO and 40% VV
ECMO patients, but it is not known which type of ECMO
those patients who experienced bowel ischemia were receiv-
ing. Thus, it cannot be determined whether this occurrence
was due solely to the underlying disease, due in part to
739
support with ECMO, or contributed to by the EN strategy.
However, a recent retrospective observational study of pa-
tients receiving VA ECMO for at least 2 days and receiving
early EN reported that early EN was associated with both
lower in-hospital (hazard ratio [HR] 0.78, 95% CI: 0.62–
0.98, P = .032) and 28-day mortality (HR 0.74, 95% CI:
0.56–0.97, P = .031).18 Despite this finding, prospective
randomized controlled trials are warranted because only
12% of patients received early EN, and reasons for feeding
decisions cannot be elucidated. Indeed, in a recent large
trial of EN vs early parenteral nutrition (PN) in critically ill
patients with shock who were receiving vasopressor support,
there was no difference in the primary outcome of 28-day
all-cause mortality.19 However, those in the EN group had
higher rates of adverse gastrointestinal events.19 Notably,
more patients in the EN group suffered bowel ischemia
(19 [2%] vs 5 [<1%], 3.84 [1.43–10.3], P = .007). Although
a secondary outcome, this may be worth considering in
patients receiving VA ECMO who are in severe shock
and receiving high-dose vasopressor support whereby it
may not be unreasonable to delay EN for a short time
with daily review to assess a more appropriate time to
commence.
Nasogastric feeding is the most common route of nutri-
tion support reported in all studies. Use of postpyloric feed-
ing appears to be infrequent overall, although rates of up to
33% have been reported.16 Postpyloric feeding tube place-
ment is more difficult and takes a longer time to achieve
than gastric placement; however, successful insertion rates
of >90% using a bedside placement technique with an
electromagnetic device have been reported.20 The relative
acceptability and perceived utility of gastric compared with
postpyloric feeding appears to vary between institutions
and countries, with an international survey of feeding
practices in adult patients receiving ECMO reporting that
postpyloric feeding is most commonly used in the United
States compared with other countries.7 It may also indicate
that this method of feeding may be more common in centers
who are experienced at bedside placement techniques in
their general ICU cohort.
The reported use of PN, either alone or in combination
with EN, ranges from 4% to 30% of patients receiving
ECMO support. The use of PN in patients receiving ECMO
remains controversial because of the possibility of lipid
infiltration into the oxygenator causing oxygenator failure.21
However, data supporting this are limited and although it
is clearly a possibility, it is not currently possible to make
evidence-based recommendations as to the appropriate use
of PN during ECMO.22 In addition, newer generation
membranes used in the circuit may now negate this risk.8
Given this, if PN is deemed to be in the best interest of the
patients on nutrition grounds, it is worth ensuring that close
monitoring of the ECMO circuit is undertaken when pa-
tients are receiving PN, especially if other lipid-containing
740
Table 1. All Observational Studies Describing Nutrition Practices in Patients Receiving ECMO.

Authors Lu et al MacGowan Ridley et al Ferrie et al Umezawa Lukas et al Scott et al

(2018)17 et al (2018)16 (2015)15 (2013)13 Makikado (2010)12 (2004)10
et al (2013)14
Patient population 102a 203 107 86 7 48 27
Study design Retrospective Retrospective Prospective, Retrospective Prospective Retrospective Retrospective
observational observational multicenter observational observational observational observational
observational
Age, y (median)b 55.4 (mean) 44 42 44.4 (mean) NR 44 NR
APACHE II score Survived: 22 18 20 26 (mean) NR NR NR
(median)b Deceased: 26
(mean)
Type of ECMO VA: 77 VV VA: 61% VA: 31 patients VA VA: 35 patients VV
VV: 25 VV: 39% VV: 55 patients VA VV: 13 patients
Provision of energy Survived: 97% Median energy target Median daily Mean daily nutrition Mean cumulated Mean nutrition NR
during ECMO Deceased: 88% delivered: 89.8% delivery: 20 delivery ࣙ80% in deficit by end of adequacy
Kcal goal achieved kcal/kg 52/86 patients study −3264.94 during
kcal/d ECMO: 55%
(VV: 67%, VA:
50%)
Provision of protein >100% protein Median target Median daily 73% of daily protein NR NR NR
during ECMO requirement delivered: 84.7% delivery: 0.9 g/kg goal tolerated in
the first 2 weeks
Definition and 73% of surviving Prokinetics 66% of patients ࣙ2 aspirates Day 1: GRVs 4-hourly GRVs ×2 consecutive
incidence of GI patients tolerated considered after 2 × received >200 mL and checked every >200 mL GRVs >150 mL
intolerance bolus feeding, GRVs >300 mL prokinetic abdominal 6 hours indicated erythromycin
27% tolerated Prokinetics used in medications distension/ Day 2: every intolerance started
continuous 52.2% of patients discomfort 12 hours GRV >150 mL 95% received
NG feedingc in 33 patients (38%) From day 3 daily prokinetics prokinetics by
No postpyloric started 48 hours
tubes or 71% of patients
prokinetics used received
prokinetics

(continued)
 Table 1. (continued)

Time taken to initiate Survived: 1.6 days 13.5 13 13.1 (mean) 2 patients within NR 25 patients within
nutrition support, Deceased: 0.8 days 24 hours 24 hours
hours (median)b (mean) 7 patients within 27 patients within
48 hours 36 hours
Method used to Energy: Energy: 25 kcal/kg/d Energy: Schofield Energy: Schofield Energy: 25 Energy: Schofield Energy: 25 kcal/kg
estimate nutrition Harris-Benedict Protein: 1.2 g/kg/d equation + SF equation + SF kcal/kg/d equation + SF Protein: 1.2–1.5
targets during equation +AF (minimum) (51% of patients) (1.1–1.2) Protein: NR (1.2–1.5) g/kg/d
ECMO (1.2) +SF Weight based Protein: 1.2 g/kg/d Protein:
(1.2–1.5) (33%) (minimum) 1.2–1.5 g/kg/d
Protein: 1.2 g/kg/d Other (16%) 1.5–2 g/kg (RRT)
(minimum) Protein: NR
Route of nutrition NG: 80% of patients Gastric: 60.6% 1602 study Gastric: all patients EN in all patients EN: 69% EN: 18
support Nil other routes Postpyloric: 34.5% days:EN: 1342 started gastric PN: 4% PN: 1
reported PN: 4.9% (84%)d feeds EN and PN: 25% EN and PN: 8
PN: 111 (7%)d Postpyloric: nil 1 patient received no
ON: 155 (10%)d PN: 18 patients nutrition
No nutrition: 83
(5%)d

AF, Activity Factor; APACHE II, Acute Physiology and Chronic Health Evaluation; ECMO, extracorporeal membrane oxygenation; EN, enteral nutrition; GI, gastrointestinal; GRV, gastric
residual volume; NG, nasogastric; NR, not reported; ON, oral nutrition; PN, parenteral nutrition; RRT, renal replacement therapy; SF, stress factor; VA, veno-arterial; VV, venovenous.
a Population divided into survived (n = 41) and deceased (n = 61).
b Median unless otherwise specified.
c Definition of tolerance not provided.
d Reported as number and percent of study days.

741
742
medications are used concomitantly. Regular monitoring
of triglycerides may also help to ascertain risk and the
point at which lipid-free PN may be warranted. In our
practice, for all patients, transoxygenator gas exchange with
regular serum lipids is monitored once daily if patients are
receiving PN. For patients with triglycerides level >3 g/L,
consideration is given to using lipid-free PN with weekly
lipids and daily fat-soluble vitamin administration.
Dose of Nutrition Support
The optimal dose of nutrition support in terms of energy
and protein in the critically ill population is unknown,9 and
this is also true for patients receiving ECMO. The American
Society for Parenteral and Enteral Nutrition guidelines23
recommend the use of a nutrition risk scoring tool in all
critically ill patients to determine the level of attention paid
to the dose of both energy and protein. However, given that
severity of illness scoring systems are reliant on physiology
and age, it is likely that they underestimate the severity
of illness on ECMO because physiology is manipulated,
patients are commonly younger than the average ICU
patient, patients may have been in the ICU for some days
before commencing ECMO, and patients tend to have a
longer ICU stay. Consequently, it is likely that current
nutrition risk scores may systematically underestimate the
actual risk, and individualized nutrition support regimens
should be provided to these patients.
Energy
Indirect calorimetry (IC) is considered the gold standard for
determining energy expenditure (EE) in critically ill adults
and is recommended as first-line choice in evidence-based
nutrition guidelines.23,24 IC determines EE by measuring
the amount of oxygen inhaled and carbon dioxide exhaled,
and relies on several assumptions including a stable dis-
tribution of carbon dioxide and knowledge of nitrogen
loss.25 Measuring IC in patients receiving ECMO needs to
take into account gas exchange via the native lung and via
the membrane lung, and also the impact of recirculation.
Recirculation adds an error into the calculation because
both the carbon dioxide and oxygen exchange across the
membrane lung can be underestimated. Recirculation also
varies based on the state of intravascular blood volume, as
well as thoraco-abdominal compliance and cardiac output.
Recirculation fraction can be measured using ultrasonic
techniques.
Two studies26,27 have demonstrated it is possible to mea-
sure EE in patients receiving ECMO in 2 ways. De Waele
et al27 described a method whereby EE is first measured in
the native lung via the ventilator; then the metabolic cart is
connected to the ECMO circuit to measure at the artificial
lung. Both numbers are then added together and inserted
into the Weir equation for calculation. Along with the time-
Nutrition in Clinical Practice 33(6)
consuming nature of this method, the authors report that
they used an in-house, custom-made adaptor to connect
the metabolic cart to the ECMO circuit, which is not avail-
able elsewhere. Wollersheim et al26 presented an alternative
method that still requires measurement of gas exchange
using IC at the native lung, but rather than connecting
the IC to the ECMO circuit, blood gas is measured before
and after the membrane oxygenator, with carbon dioxide
and oxygen content being subsequently calculated using
a referenced equation.28 The Weir equation is then used
to determine EE. Both protocols provide an opportunity
for EE to be measured in patients receiving ECMO, but
require prospective validation, particularly at different time
points during the ECMO run because EE is not static and
different blood flow rates may influence the measurement.
However, these methods will most certainly allow further
investigations into the optimum energy targets for patients
receiving both VA and VV ECMO.
As with general ICU patients, when IC is not available,
predictive or weight-based equations are the method of
choice for determining energy targets in patients receiving
ECMO,10,12-17 although these have not been derived from
patients receiving ECMO, and it is not known whether
they are truly appropriate. Studies of clinical practice de-
scribe the main equations used to be 25 kcal/kg12,16-18 or
Schofield with added stress factor12,13,15 ranging from 1.1 to
1.5.12,13 When compared with measured energy expenditure
(MEE), Wollersheim et al26 demonstrated that common
prediction equations significantly underfed or overfed pa-
tients, with 48% overestimated by >500 kcal, when com-
pared with MEE. Importantly, the weight-based equation,
25 kcal/kg,23,24 was shown to significantly underestimate
EE when compared with MEE. This finding contradicts de
Waele et al,29 who found that MEE in patients receiving
ECMO was on average 19 kcal/kg, albeit in a group of only
6 patients. The differences in these 2 studies may be because
of different methods used to measure EE or differences in
ECMO; one included only VV ECMO patients,26 whereas
the other included both VV and VA ECMO patients.29 This
underscores the importance of undertaking further research
in this area and using clinical judgment at the bedside when
estimating energy targets in these patients.
In the absence of nutrition guidelines specifically for
patients receiving ECMO, it appears reasonable to follow
currently available guidelines for general critically unwell
patients that recommend IC where available or weight-based
equations ranging from 20 to 25 kcal/kg.23,24 European So-
ciety for Parenteral and Enteral Nutrition guidelines24 go on
to further recommend 25–30 kcal/kg when the patient has
moved into the recovery phase of critical illness. With more
and more patients being liberated from the ventilator and
mobilizing out of bed while still receiving ECMO,30,31 it may
be suitable to increase energy provision to 30 kcal/kg when
a patient is in this phase of recovery despite still receiving
Bear et al
ECMO. However, currently, no studies have measured EE
in ambulatory patients receiving ECMO.
Improved outcomes have been demonstrated in general
ICU patients receiving around 80% of their estimated en-
ergy targets32 and 70% of measured energy targets,33 which
is likely to be comparable in patients receiving ECMO.
Although often the most unwell patients in an ICU, studies
have shown it is possible for patients receiving ECMO
to achieve and tolerate this target. Ferrie et al13 reported
that 60% of their patients receiving ECMO received >80%
of the estimated energy targets, and more recently Mac-
Gowan et al16 showed that 89.8% of target energy was
delivered to patients receiving VV ECMO. However, in the
latter study, underfeeding was still common with patients
receiving <80% of their energy targets on one-quarter of
feeding days. Interestingly, both studies found a difference
in the energy delivered in the ECMO and post-ECMO
periods, with those in the post-ECMO period receiving
more. It is thought that the reason for this relates to
both gastric/enteral feeding intolerance and to the greater
number of procedures requiring cessation of EN during the
ECMO run.16
When considering the relationship of the dose of energy
to clinical outcome in adult patients with ECMO, little data
exist. However, receiving adequate energy intake (>80%)
was associated with longer length of stay in 1 study.16 It is
not known whether this is simply due to the patients who
remained in the ICU longer being able to reach nutrition
targets, or whether there is a true burden associated with
nutrition that increases length of stay. In contrast, in a
retrospective study of Taiwanese patients receiving ECMO,
achieving >80% of nutrition targets was associated with
improved mortality.17 It is clear that prospective random-
ized controlled trials are required to further investigate these
findings.
Protein
Recommendations for protein intakes in the critically ill
range from 1.2 g/kg/d for the general ICU patient up to
2.5 g/kg/d for those who are obese or admitted with burns
or after trauma.23,24 However, these recommendations are
based on limited, mainly observational data. The assumed
benefit of providing adequate protein is a reduction in skele-
tal muscle wasting, which may lead to improved physical
function post-ICU.34 Providing adequate protein to patients
receiving ECMO may be particularly important because
of the significant amount of muscle wasting experienced.
Recently, the trajectory of muscle wasting, measured using
muscle ultrasound, in patients receiving both VV and VA
ECMO was reported. In line with critically ill, non-ECMO
patients, rectus femoris cross-sectional area had decreased
by 19% at day 10 and 30% at day 20.35 Other quadriceps
muscles also decreased in size to a similar degree. Smaller
743
muscle size (total muscle thickness) was associated with
lower strength and mobility score.36
Only 1 study aiming to investigate optimum protein in-
takes in patients receiving ECMO exists. This small study in-
vestigated the nitrogen balance in obese and nonobese adult
patients receiving VV ECMO.37 Protein targets were set
at 1.5–2.0 g/kg/d for nonobese patients and 2.0–2.5 g/kg/d
for obese patients, with around 85% of this target be-
ing delivered. Results were compared between obese and
nonobese individuals, and it was concluded that obese
compared with nonobese patients had a more negative
nitrogen balance (−1.7 ± 5.7 vs −11.5 ± 9.6). Despite
the fact that nitrogen balance has inherent limitations that
make interpretation in the critically ill population difficult,38
these data provide some initial insights into the nitrogen
balance and potential protein requirements in these patients,
and this topic should certainly be explored further.
From existing observational data undertaken in patients
receiving ECMO, a minimum 1.2 g/kg/d is targeted for pro-
tein in clinical practice. Although there is a discrepancy be-
tween prescribed and delivered protein intakes, it is possible
to meet >80% of the prescribed target,16,17 which has been
shown to lead to lower mortality in critically ill, non-ECMO
patients.33,39 Whether meeting current recommendations for
protein intake in this patient population improves outcomes
not related to muscle wasting is unknown, but higher protein
intakes were associated with a longer length of stay in
the largest observational study.16 However, as mentioned
previously, these data were not adjusted for length of stay.
Until further data are available, it would not seem unrea-
sonable to aim for protein intakes in line with the general
critically ill population accounting for clinical condition and
factors such as the use of continuous renal replacement
therapy.
Barriers to the Delivery of EN
Delayed gastric emptying is frequently cited as a
complication associated with enteral feeding in patients
receiving ECMO.6,12,15,16 This was confirmed in a United
Kingdom–wide survey of nutrition practices in patients
receiving VV ECMO, where 69% of responders reported
that ࣙ50% of their patients require prokinetics while
receiving VV ECMO.6
However, the definition of delayed gastric emptying
is based on gastric residual volumes (GRVs), and there
are a variety of thresholds ranging from 150 to 300 mL.
Furthermore, in the observational studies reported to date,
those with a lower acceptable GRV report a higher incidence
of medications used to promote gastric emptying than those
with a higher GRV threshold. Where a GRV of 150 mL is
accepted and postpyloric feeding not used, prokinetic use
is 71%–95%.10,12 Furthermore, there are other approaches,
including use of postpyloric feeding, on an ad hoc or
744
planned basis, which alter the utility of GRV measurements.
It is not currently known what an ideal GRV threshold is, or
indeed whether they should be measured at all.40
The largest observational study to date indicated that
prokinetics/postpyloric feeding was used in >50% of their
patients with the result being that >80% of the energy
and protein targets were delivered.16 This indicates that
with a flexible approach to nutrition support, underfeeding
can be avoided. Although patients in this study were all
receiving VV ECMO, Ferrie et al13 reported no difference
in the incidence of enteral feeding intolerance between
those receiving VA and VV ECMO, so the common be-
lief that VA ECMO reduces gut perfusion and therefore
reduces motility may not hold true in modern-day clinical
practice.
Delayed gastric emptying is not the only barrier to
delivering EN in these patients. Both MacGowan et al16
and Ridley et al15 reported bedside and operating room
procedures to be the most frequent reason for enteral
feeding interruptions, occurring in >30% of patients. This
is also a frequently cited reason for feeding interruptions
in the general critically ill population, and strategies such
as volume-based feeding have proven useful in prevent-
ing the associated poor delivery of nutrition.41 However,
volume-based feeding has not been studied specifically in
patients receiving ECMO, but should not be discounted as
an option, particularly in the more stable patient. Certainly
guidelines directing when EN should be stopped and started
for procedures have been shown to increase nutrition targets
in non-ECMO centers42 and should be considered as stan-
dard practice for all critically ill patients.
Studies Planned or in Progress
A search of clinical trials databases was undertaken to deter-
mine details of any studies that are in progress investigating
aspects of nutrition in patients receiving ECMO. Only 3
studies were found.
1. Measuring Energy Expenditure in ECMO
(Extracorporeal Membrane Oxygenation) Patients
(MEEP) (NCT01992237) aims to describe a
calculation to determine nutrition targets in patients
receiving ECMO. Secondary aims include describing
levels of nutrition needs for different mechanical
ventilation states, and to use oxygen and carbon
dioxide elimination to estimate cardiac output.
2. The use of Computed Tomography (CT) to Measure
Skeletal Muscle Quantity and Quality in Patients
Receiving ECMO (NCT03269825) aims to investi-
gate the clinical predictive value of and change in
muscle quantity and quality in patients receiving
VV ECMO. Secondary aims are to determine any
Nutrition in Clinical Practice 33(6)
relationship between total nutrition received and
change in muscle quantity and quality.
3. Characterizing changes in muscle quantity and
quality in patients requiring ECMO during criti-
cal illness (ECMO USS) (NCT02995811) aims to
identify changes in different muscle groups (trunk,
respiratory, and skeletal muscle) in patients receiv-
ing ECMO. The association between nutrition and
changes in muscle quality and quantity will also be
addressed.
4. Progressive Rehabilitation Therapy in Patients with
Advanced Lung Disease (NCT 03562728) aims to
investigate the impact of a multimodal rehabilitation
program in ameliorating the loss of muscle mass and
strength, and lower extremity balance, strength, and
coordination in patients requiring lung transplant or
ECMO as a bridge to transplant. The multimodal
rehabilitation program includes neuromuscular elec-
tric stimulation, strength and mobility training, and
nutrition supplementation with amino acids 3 times
daily.
Potential Research Questions
No prospective randomized controlled trials exist examin-
ing the optimum route, timing, or adequacy of nutrition
support in patients receiving ECMO, but it is clear that
these are urgently required. Because VA and VV patients
may be different in terms of their metabolic sequelae, we
recommend that VA and VV patients are recruited and
investigated separately, or at least subgroup analysis is
considered for each group where they have been recruited
together.
The Intensive Care Medicine research agenda for extra-
corporeal life support acknowledges the limited evidence
surrounding the optimal feeding strategy for these patients.8
Although nutrition did not feature in the top 10 recom-
mended research trials for these patients, early rehabili-
tation/mobilization did. We would propose that nutrition
and rehabilitation go hand in hand,34,43 and encourage
investigators to report appropriate nutrition parameters in
any of these future studies.
We have outlined 5 research questions that we feel are
priorities to answer:
1. What is the optimum energy intake in patients receiv-
ing VV and VA ECMO?
2. What is the optimum protein intake in patients
receiving VV and VA ECMO?
3. What is the impact of nutrition support on the phys-
ical and functional recovery of patients receiving VA
and VV ECMO?
4. What are the optimum energy and protein targets
during rehabilitation in patients receiving ECMO?
Bear et al
5. Are nutrient losses (micronutrients and macronutri-
ents) present across the ECMO membrane, and are
these losses clinically meaningful?
Conclusion
The use of ECMO as a lifesaving treatment for both
severe respiratory and cardiac failure is increasing. It is
clear that these patients are some of the sickest in the
ICU and require a multidisciplinary approach to treatment
that includes appropriate nutrition therapy. Currently, there
are limited data to guide nutrition therapy, but it appears
that early EN is safe. Until further data are available,
nutrition should be provided in line with current guidelines
for critically ill patients. Prospective randomized controlled
trials investigating optimal nutrition in these patients are
urgently required.
Statement of Authorship
D. E. Bear, N. A. Barrett, and E. Smith contributed to
conception/design of the manuscript; D. E. Bear, N. A.
Barrett, and E. Smith contributed to acquisition, analysis,
or interpretation of the data; D. E. Bear, N. A. Barrett,
and E. Smith drafted the manuscript; D. E. Bear, N. A.
Barrett, and E. Smith critically revised the manuscript; and
D. E. Bear, N. A. Barrett, and E. Smith agree to be fully
accountable for ensuring the integrity and accuracy of the
work. All authors read and approved the final manuscript.
Supplementary Information
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
References
1. Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracorporeal
membrane oxygenation center and mortality among patients with
severe 2009 influenza A(H1N1). JAMA. 2011;306(15):1659-1668.
2. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic
assessment of conventional ventilatory support versus extracorporeal
membrane oxygenation for severe adult respiratory failure (CESAR): a
multicentre randomised controlled trial. Lancet. 2009;374(9698):1351-
1363.
3. Brodie D. The evolution of extracorporeal membrane oxygenation for
adult respiratory failure. Ann Am Thorac Soc. 2018;15(suppl 1):S57-
S60.
4. Sherren PB, Shepherd SJ, Glover GW, et al. Capabilities of a mobile
extracorporeal membrane oxygenation service for severe respiratory
failure delivered by intensive care specialists. Anaesthesia. 2015;70(6):
707-714.
5. Extracorporeal Life Support Organization (ELSO). ELSO
Guidelines for Adult Respiratory Failure, Version 1.4. August 2017.
https://www.elso.org/Portals/0/ELSO%20Guidelines%20For%20Adult
%20Respiratory%20Failure%201_4.pdf
6. Bear DE, Halslam J, Camporota L, Shankar-Hari M, Barrett NA. An
international survey of nutrition practices in adult patients receiving
veno-venous ECMO. Intensive Care Med Exp. 2015;3(suppl 1):A295.
745
7. Nandwani V, Tauber A, McCarthy P, Herr D. 333. Nutrition practice
patterns in adult ECMO patients: results of an international survey.
Crit Care Med. 2014;42(12 suppl 1):A1440.
8. Combes A, Brodie D, Chen YS, et al. The ICM research agenda on
extracorporeal life support. Intensive Care Med. 2017;43(9):1306-1318.
9. Preiser JC, van Zanten AR, Berger MM, et al. Metabolic and nutri-
tional support of critically ill patients: consensus and controversies. Crit
Care. 2015;19:35. https://doi.org/10.1186/s13054-015-0737-8.
10. Scott LK, Boudreaux K, Thaljeh F, Grier LR, Conrad SA. Early en-
teral feedings in adults receiving venovenous extracorporeal membrane
oxygenation. JPEN J Parenter Enteral Nutr. 2004;28(5):295-300.
11. Reintam Blaser A, Starkopf J, Alhazzani W, et al. Early enteral
nutrition in critically ill patients: ESICM clinical practice guidelines.
Intensive Care Med. 2017;43(3):380-398.
12. Lukas G, Davies AR, Hilton AK, et al. Nutritional support in adult
patients receiving extracorporeal membrane oxygenation. Crit Care
Resusc. 2010;12(4):230-234.
13. Ferrie S, Herkes R, Forrest P. Nutrition support during extracorporeal
membrane oxygenation (ECMO) in adults: a retrospective audit of 86
patients. Intensive Care Med. 2013;39(11):1989-1994.
14. Umezawa Makikado LD, Flordelis Lasierra JL, Perez-Vela JL, et al.
Early enteral nutrition in adults receiving venoarterial extracorporeal
membrane oxygenation: an observational case series. JPEN J Parenter
Enteral Nutr. 2013;37(2):281-284.
15. Ridley EJ, Davies AR, Robins EJ, et al. Nutrition therapy in adult pa-
tients receiving extracorporeal membrane oxygenation: a prospective,
multicentre, observational study. Crit Care Resusc. 2015;17(3):183-189.
16. MacGowan L, Smith E, Elliott-Hammond C, et al. Adequacy
of nutrition support during extracorporeal membrane oxygenation
[published online ahead of print January 26, 2018]. Clin Nutr.
https://https://doi.org/10.1016/j.clnu.2018.01.012
17. Lu MC, Yang MD, Li PC, et al. Effects of nutritional intervention
on the survival of patients with cardiopulmonary failure undergoing
extracorporeal membrane oxygenation therapy. In Vivo. 2018;32(4):
829-834.
18. Ohbe H, Jo T, Yamana H, et al. Early enteral nutrition for cardiogenic
or obstructive shock requiring venoarterial extracorporeal membrane
oxygenation: a nationwide inpatient database study. Intensive Care
Med. 2018;44(8):1258-1265.
19. Reignier J, Boisrame-Helms J, Brisard L, et al. Enteral versus
parenteral early nutrition in ventilated adults with shock: a ran-
domised, controlled, multicentre, open-label, parallel-group study
(NUTRIREA-2). Lancet. 2018;391(10116):133-143.
20. Bear DE, Champion A, Lei K, Camporota L, Barrett NA. Electro-
magnetically guided bedside placement of post-pyloric feeding tubes
in critical care. Br J Nurs. 2017;26(18):1008-1015.
21. Buck ML, Ksenich RA, Wooldridge P. Effect of infusing fat emulsion
into extracorporeal membrane oxygenation circuits. Pharmacotherapy.
1997;17(6):1292-1295.
22. Lee HM, Archer JR, Dargan PI, Wood DM. What are the adverse
effects associated with the combined use of intravenous lipid emulsion
and extracorporeal membrane oxygenation in the poisoned patient?
Clin Toxicol (Phila). 2015;53(3):145-150.
23. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the
provision and assessment of nutrition support therapy in the adult
critically ill patient: Society of Critical Care Medicine (SCCM) and
American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.).
JPEN J Parenter Enteral Nutr. 2016;40(2):159-211.
24. Kreymann KG, Berger MM, Deutz NE, et al. ESPEN guide-
lines on enteral nutrition: intensive care. Clin Nutr. 2006;25(2):
210-223.
25. Brandi LS, Bertolini R, Calafa M. Indirect calorimetry in criti-
cally ill patients: clinical applications and practical advice. Nutrition.
1997;13(4):349-358.
746
26. Wollersheim T, Frank S, Muller MC, et al. Measuring energy ex-
penditure in extracorporeal lung support Patients (MEEP)—protocol,
feasibility and pilot trial. Clin Nutr. 2018;37(1):301-307.
27. De Waele E, van Zwam K, Mattens S, et al. Measuring resting
energy expenditure during extracorporeal membrane oxygenation: pre-
liminary clinical experience with a proposed theoretical model. Acta
Anaesthesiol Scand. 2015;59(10):1296-1302.
28. Dash RK, Bassingthwaighte JB. Erratum to: Blood HbO2 and HbCO2
dissociation curves at varied O2, CO2, pH, 2,3-DPG and temperature
levels. Ann Biomed Eng. 2010;38(4):1683-1701.
29. De Waele E, Staessens K, Demol J, La Meir M, Sapen HD. SUN-P027:
the energy expenditure of patients on ECMO is not elevated: beware
when you feed them! Clin Nutr. 2017;36:S62.
30. Fan E, Gattinoni L, Combes A, et al. Venovenous extracorporeal mem-
brane oxygenation for acute respiratory failure: a clinical review from
an international group of experts. Intensive Care Med. 2016;42(5):712-
724.
31. Marhong JD, DeBacker J, Viau-Lapointe J, et al. Sedation and mo-
bilization during venovenous extracorporeal membrane oxygenation
for acute respiratory failure: an international survey. Crit Care Med.
2017;45(11):1893-1899.
32. Heyland DK, Cahill N, Day AG. Optimal amount of calories for
critically ill patients: depends on how you slice the cake! Crit Care Med.
2011;39(12):2619-2626.
33. Zusman O, Theilla M, Cohen J, et al. Resting energy expenditure,
calorie and protein consumption in critically ill patients: a retrospective
cohort study. Crit Care. 2016;20(1):367.
34. Bear DE, Wandrag L, Merriweather JL, et al. The role of nutritional
support in the physical and functional recovery of critically ill patients:
a narrative review. Crit Care. 2017;21(1):226.
35. Hayes K, Holland AE, Pellegrino VA, Mathur S, Hodgson CL. Acute
skeletal muscle wasting and relation to physical function in patients
requiring extracorporeal membrane oxygenation (ECMO). J Crit Care.
2018;48:1-8.
Nutrition in Clinical Practice 33(6)
36. Puthucheary ZA, Rawal J, McPhail M, et al. Acute skeletal muscle
wasting in critical illness. JAMA. 2013;310(15):1591-1600.
37. Pelekhaty S, Galvagno SM Jr, Hochberg E, et al. Nitrogen balance
during venovenous extracorporeal membrane oxygenation support:
preliminary results of a prospective, observational study [published
online ahead of print May 25, 2018]. JPEN J Parenter Enteral Nutr.
https://https://doi.org/10.1002/jpen.1176
38. Kopple JD. Uses and limitations of the balance technique. JPEN J
Parenter Enteral Nutr. 1987;11(5 suppl):79s-85s.
39. Nicolo M, Heyland DK, Chittams J, Sammarco T, Compher C. Clin-
ical outcomes related to protein delivery in a critically ill population: a
multicenter, multinational observation study. JPEN J Parenter Enteral
Nutr. 2016;40(1):45-51.
40. Reignier J, Mercier E, Le Gouge A, et al. Effect of not mon-
itoring residual gastric volume on risk of ventilator-associated
pneumonia in adults receiving mechanical ventilation and early
enteral feeding: a randomized controlled trial. JAMA. 2013;309(3):249-
256.
41. Heyland DK, Murch L, Cahill N, et al. Enhanced protein-energy
provision via the enteral route feeding protocol in critically ill patients:
results of a cluster randomized trial. Crit Care Med. 2013;41(12):2743-
2753.
42. Segaran E, Barker I, Hartle A. Optimising enteral nutrition in crit-
ically ill patients by reducing fasting times. J Intensive Care Soc.
2016;17(1):38-43.
43. Heyland DK, Stapleton RD, Mourtzakis M, et al. Combining nutrition
and exercise to optimize survival and recovery from critical illness:
conceptual and methodological issues. Clin Nutr. 2016;35(5):1196-
1206.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Nutrition Support During Pediatric Extracorporeal December 2018 747–753

C 2018 American Society for

Membrane Oxygenation Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10212
wileyonlinelibrary.com

Bethany J. Farr, MD1,2 ; Samuel E. Rice-Townsend, MD1,2 ;

and Nilesh M. Mehta, MD2,3,4

Abstract
Providing adequate nutrition to critically ill pediatric patients is essential and positively impacts outcomes. Critically ill infants and
children receiving extracorporeal membrane oxygenation (ECMO) therapy are nutritionally vulnerable, yet there are challenges to
reliable assessment of nutrition requirements and to the delivery of optimal nutrition in this cohort. In this review of the relevant
literature, we present the current evidence and guidelines for the optimal prescription and delivery of nutrition for pediatric patients
receiving ECMO. We also discuss nutrient delivery considerations in ECMO survivors and identify areas where further study is
needed. (Nutr Clin Pract. 2018;33:747–753)

Keywords
critical illness; enteral nutrition; extracorporeal membrane oxygenation; nutrition requirements; nutrition assessment; nutrition
support; parenteral nutrition; pediatrics

Introduction population is heterogenous, encompassing neonates with

Critically ill patients in the pediatric intensive care unit
(PICU) require adequate nutrition support to achieve the
best outcomes. Patients with respiratory or cardiorespira-
tory failure who require support with extracorporeal mem-
brane oxygenation (ECMO) are, by definition, a severely ill
and vulnerable subset of the critically ill population who
would benefit from optimal nutrition. Yet, the assessment
of energy and protein requirements and the delivery of
these nutrients can be a particular challenge because of the
unclear metabolic demands and potential safety concerns
of enteral nutrition (EN) in these patients. Moreover, the
congenital heart disease who are receiving veno-arterial
(VA) ECMO to young adults with cystic fibrosis awaiting
lung transplant who are receiving venovenous (VV) ECMO,
with varied needs and considerations. However, as use of
ECMO has expanded over the last 2 decades, now with
>15,000 cases annually, our understanding of nutrition
requirements and optimal nutrient delivery practices for this
patient group needs to be reviewed.1 We aimed to examine
the existing literature on prescription and delivery of nutri-
tion therapy in these critically ill patients. In addition, we
sought to identify gaps in knowledge where future research
efforts are needed.

From the 1 Department of Surgery, Boston Children’s Hospital, Boston, Massachusetts, USA; 2 Harvard Medical School, Boston, Massachusetts,
USA; 3 Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s Hospital, Boston,
Massachusetts, USA; and 4 Center for Nutrition, Boston Children’s Hospital, Boston, Massachusetts, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on October 16, 2018.

Podcast available
Listen to a discussion of this manuscript with NCP Editor-in-Chief Jeanette M. Hasse, PhD, RD, LD, FADA, CNSC, and author Samuel E.
Rice-Townsend, MD. This and other NCP podcasts are available at: https://onlinelibrary.wiley.com/page/journal/19412452/homepage/podcasts

Corresponding Author:
Nilesh M. Mehta, MD, Division of Critical Care Medicine, Department of Anesthesiology, Critical Care and Pain Medicine, Boston Children’s
Hospital, 300 Longwood Avenue, Bader 634, Boston, MA 02115, USA.
Email: nilesh.mehta@childrens.harvard.edu
748
Nutrition Requirements During Pediatric
ECMO
Macronutrient Requirements
Our current understanding of and ability to accurately mea-
sure energy needs in patients receiving ECMO is limited.
Few studies have explored this question, and results have
been contradictory. In a study where energy expenditure in
neonates receiving ECMO was measured by stable isotope
tracer techniques, patients were hypermetabolic both during
ECMO and in the post-ECMO phase.2 Although patients
receiving ECMO are often presumed to be hypermetabolic
because of the inflammatory burden of an extracorporeal
circuit combined with underlying severe illness, this may not
always be true. Using respiratory mass spectroscopy, energy
expenditure in infants who underwent Norwood surgery
was reported to decline rapidly within 8 hours after surgery.3
In infants who underwent Fontan surgery with cardiopul-
monary bypass, the immediate postoperative metabolic
measurements revealed hypometabolism.4 In another study
using an intravenous isotope-labeled infusion technique to
estimate energy requirements in neonates supported with
ECMO, energy needs were not significantly greater than
age-matched counterparts.5
Due to the technical difficulties of traditional methods
to accurately measure gas exchange in a patient attached
to an ECMO circuit, data on energy expenditure are scant
in children receiving ECMO.6 Without indirect calorimetry,
intensivists must rely on standard equations to estimate
resting energy expenditure (REE). These equations, how-
ever, were developed using data from healthy children and
are therefore frequently inaccurate in estimating REE in
critically ill children.7 Therefore, the reliance on standard
equations to determine energy needs may result in un-
intended underestimation or overestimation of REE and
result in caloric overfeeding or underfeeding. A reliable
and practical method to measure energy expenditure in
patients receiving ECMO is needed. Until then, existing
best practices established in the general critically ill pediatric
and neonatal populations should be used to guide energy
delivery goals for patients supported with ECMO.
Neonatal and pediatric patients receiving ECMO
have increased protein requirements because of profound
catabolism. It is essential to offset the protein losses
from this catabolic state with provision of adequate
protein.2,8 In neonates, whole-body protein breakdown
was found to be 100% higher than in age-matched healthy
neonates.2 In neonates treated with ECMO, >1.5 g/kg/d
protein is needed to achieve a positive nitrogen balance.9
Nutrition guidelines of neonatal patients receiving ECMO
recommend provision of up to 3 g/kg/d protein to offset
catabolic losses.8 It has been suggested that administration
of insulin, as an anabolic hormone, may negate some of
Nutrition in Clinical Practice 33(6)
the protein breakdown seen in neonates receiving ECMO;
however, this practice has not been adopted because further
research is needed.10
Macronutrient Intake and Outcomes
Although data are limited to guide the prescription of
nutrition therapy, the importance of the nutrition state
at the onset of ECMO therapy and of the delivery of
adequate nutrition during therapy is clear. Anton et al11
reported in 2016 that being underweight at the time of
ECMO cannulation was an independent factor for mor-
tality. Of 491 patients, 24% were underweight at the time
of ECMO initiation and had an increased odds ratio of
1.99 for in-hospital mortality as compared with normal-
weight and obese children.11 The importance of adequate
nutrient delivery for these patients may be extrapolated
from studies performed on other critically ill patients, the
sickest of whom may ultimately require ECMO. For such
patients, effective delivery of nutrition while in the intensive
care unit (ICU) has been shown to impact outcomes, even
mortality. For mechanically ventilated patients in 31 PICUs
in 8 countries, receiving less than a third of the prescribed
energy during the first 10 days after admission to the PICU
was associated with significantly higher 60-day mortality.12
Another single-center retrospective study of PICU patients
with acute respiratory distress syndrome demonstrated that
adequate delivery of energy (>80% predicted REE) and
protein (>1.5 g/kg/d) was associated with improved survival
to ICU discharge.13 In patients with a >4-day PICU stay,
receiving at least 25% of goal calories via the enteral route
over the first 48 hours of admission resulted in a lower
mortality overall compared with those who did not meet
this goal.14 The role of nutrient delivery route in patients
receiving ECMO is discussed later in this review.
Adequate protein delivery was also associated with more
ventilator-free days and reduced mortality after adjusting
for illness severity.13 Similarly, in a multicenter study of
1245 mechanically ventilated children, delivery of ࣙ60% of
prescribed protein requirement was associated with lower
60-day mortality. Mortality for patients with <20% ade-
quate protein delivery was 9.3%, and decreased to 5.6% in
those receiving 20%–60% recommended protein, and 3.2%
in those receiving >60% recommended protein, demonstrat-
ing a dose-dependent relationship.15 The route of nutrient
delivery may be an important consideration, and early
enteral feeding is generally preferred in critically ill children.
Micronutrient Requirements
Children who are receiving extended ECMO may expe-
rience aggravated losses of micronutrients. In an ex vivo
model of ECMO, circuit losses of essential amino acid
isoleucine, vitamin A, and vitamin E were demonstrated.16
Farr et al
In an animal model of smoke-induced acute lung injury,
ECMO application was associated with exaggerated losses
of selenium.17 The mechanisms of these alterations in
micronutrients are not clear, but their impact on outcomes
in pediatric ECMO may be important and require fur-
ther investigation. In patients receiving ECMO who also
require dialysis, micronutrient losses may be further com-
pounded, because continuous renal replacement therapy
has been associated with loss of ionized calcium, inorganic
phosphorus, and selenium, among other trace elements.18,19
These observations may have implications for micronutrient
supplementation in the pediatric ECMO population, and at-
tempts to monitor micronutrients should be made. However,
the assessment of the micronutrient status of critically ill
patients is challenging because of the confounding effects of
inflammation on micronutrient levels in plasma. In patients
receiving ECMO, these challenges are further compounded
by the dilutional effects of the extracorporeal circuit blood
volume. Overall, micronutrient distribution, losses, and
requirements in pediatric ECMO remain important areas
for future investigations.
Timing and Route of Nutrient Delivery
Timing of Initiation
Current guidelines for critically ill pediatric patients recom-
mend initiation of EN within 24–48 hours of admission
whenever possible.7 In patients receiving ECMO, there are
multiple potential barriers to EN, and this route is often not
sufficient to meet daily nutrient delivery goals.20 The use of
parenteral nutrition (PN) to supplement insufficient EN is
therefore prevalent in this group. However, a recent study
has questioned the role of early PN in critically ill children,
although patients receiving ECMO were not included.21
Early discussions around the safe and efficient route of
nutrient delivery are crucial in children receiving ECMO,
particularly those with poor nutrition status at baseline or
with a protracted course of illness.
Safety of EN
The safety of enteral delivery of nutrition with ECMO has
long been a concern. It was initially argued that enteral
feeding in the setting of decreased intestinal perfusion
would increase the risk for gut ischemia, translocation of
bacteria, and sepsis.22,23 Thus, PN has been preferred in
many centers as a means to achieve nutrient delivery goals
during the acute phase. Using differential carbohydrate
absorption, gut permeability was shown to be increased
in neonatal patients receiving ECMO relative to healthy
infants. However, providing minimal enteral feeding to
this group did not negatively alter permeability or lead to
complications.22 It should be noted that this retrospective
749
study was not randomized. The first 9 patients were treated
with PN only, whereas the next 7 patients received EN,
initiated between days 3 and 9 of ECMO, introducing a
potential for bias and confounding.
Several retrospective studies have reported that enteral
feeding can be tolerated without excessive risk in patients
receiving VA and VV ECMO. One retrospective chart review
over a 5-year period reported success in feeding in 67 of 77
neonatal patients receiving ECMO.24 In patients who were
fed, there were no reports of bilious emesis, bloody stools, or
abdominal distention; however, feeds were frequently held
for a variety of reasons. Notably in this study, 80% of the
infants being enterally fed were receiving inotropic support.
Another cohort study of enterally fed pediatric patients
receiving VA and VV ECMO reported no increase in
sepsis, abdominal pathology, or respiratory infections, while
demonstrating the cost-effectiveness of this route relative to
PN.23 A retrospective single-institution study showed that
neonates receiving VA ECMO who were enterally fed did
not have any increase in bacteremia with gut pathogens as
compared with the infants receiving PN, concluding that
there was no increase in translocation associated with EN.25
Similarly, in another study including 49 infant and pediatric
patients receiving ECMO, there was no association between
EN and increased infection or abdominal pathology.26
Multiple retrospective studies including adults receiving
ECMO have also shown that enteral delivery is safe and well
tolerated, with no adverse complications.27-29
In light of these reports, a recent survey of 96 in-
stitutions characterizes the current institutional practice
patterns around nutrition in pediatric and neonatal patients
receiving ECMO.30 Eighty-four percent reported using EN
during ECMO, and 385 of these sites used an algorithm for
initiating and advancing EN. The use of EN was reported by
a higher proportion of respondents during VV (71%) vs VA
(55%). EN delivery was practiced by 23% during escalating
doses of vasopressor and 60% when vasopressors were being
weaned. It is evident that retrospective data have demon-
strated safety with the use of EN, and recommendations
favor its use. Current practice has also followed this trend
in increasing use of EN in patients with ECMO.
Timing of PN
Most studies of nutrition in patients receiving ECMO
demonstrate significant use of PN as a supplemental or
primary source of nutrition within the first week. There
are conflicting data on the appropriate timing of initiation
of PN, and none specific to the ECMO population. A
recent study in critically ill pediatric patients reported harm
from early initiation of PN (within 24 hours of admission)
compared with a strategy of withholding PN during the
first week in the ICU.21 A decrease in incidence of new
infections and decreased length of ICU and hospital stay
750
were reported in those who received late initiation of PN
(after 1 week). Infants and children receiving ECMO were
not included in this study. In addition, there were several
aspects of the study design that limit external validity.21,31
In the absence of data, a reasonable approach in infants
and children receiving ECMO may be to emphasize early
and cautious advancement of EN and to initiate PN within
3–5 days in children with malnutrition and within 5–7 days
in well-nourished children if nutrient delivery goal has not
been achieved by EN alone.32
In early years of ECMO, lipid emulsion for patients
receiving PN was delivered either via a separate intravenous
(IV) catheter or directly into the circuit. In an in vitro
study, layering and agglutination of lipid emulsion in the
circuit was noted by 30 minutes of the infusion. A survey
conducted at that time revealed that 73% of ECMO centers
preferentially used a separate IV access for lipid infusions,
and 18% used the ECMO circuit for infusion, with all cen-
ters reporting complications such as agglutination, cracking
of stopcocks, membrane malfunction, and clot formation.33
Based on such studies that suggested potential risk to
the circuit from direct infusion, lipid infusion through
a separate IV catheter was recommended.34 More recent
experience suggests that intralipid infusions are not associ-
ated with oxygenator failure with newer polymethylpentene
membranes.35
Practical Aspects of Nutrient Delivery:
Suggested Algorithm
In 2010, the American Society for Parenteral and Enteral
Nutrition published guidelines for delivery of nutrition to
neonates receiving ECMO.8 The guidelines recommended
protein requirements of up to 3 g/kg/d and energy require-
ments equivalent to healthy subjects. Early introduction
of enteral feeding was recommended once the patient had
stabilized. In practice, a majority (84.2%) of providers re-
ported use of EN during the ECMO course, and postpyloric
feeding was preferred over gastric. An institutional feeding
protocol for patients receiving ECMO is used in a minority
of centers.30 In general, the delivery of nutrients by EN
during the early ECMO course is limited, with a reliance
on PN.20,26 In a single-center report of 49 neonatal and
pediatric patients receiving ECMO, PN was used in 55%
of patients, any EN in 45% of patients, and an estimated
average of 67% of goal energy needs were supplied by day
5 of ECMO. For patients receiving some EN, it accounted
for 38% of total energy supplied, on average.26 In another
institutional report of 54 neonatal and pediatric noncardiac
patients receiving ECMO, energy and protein delivery was
adequate by the end of 1 week receiving ECMO. PN was
the major source of nutrition (94%); however, 85% of
patients were receiving a combination of PN and EN at
day 7, with 44% receiving greater than trophic EN volumes.
Nutrition in Clinical Practice 33(6)
Still, EN averaged only 10% of energy needs in the first
week of ECMO. Younger age, diagnosis of congenital
diaphragmatic hernia (CDH), and use of VA ECMO, but
not use of vasoactive agents, were factors associated with
low EN use.20 In a review of 491 pediatric patients receiving
ECMO, PN was initiated in 88% of patients at a median
time of day 1 from cannulation. EN was initiated in 30%
of patients, and 35% of those reached full enteral feeds. Of
note, 17% had feeds stopped for a concern of pneumatosis
intestinalis.11
We have described multiple reports of the safety and
tolerance of enteral feeds in patients receiving ECMO
after initial resuscitation and hemodynamic stability is
achieved, even while receiving vasoactive infusions.23,24,26
The benefits of initiating enteral feeds in an infant with
ongoing hemodynamic instability need to be balanced
with the potential for gut ischemia, bacterial translocation,
and sepsis secondary to poor perfusion and splanchnic
vasoconstriction. The clinical apprehension for gut-related
worsening after introduction of enteral feeding in infants
receiving ECMO is prevalent, resulting in a reliance on
PN during the acute phase.30 Critically ill children who
are not receiving ECMO support have tolerated EN while
receiving vasoactive medications with no adverse abdominal
complications, while adjusting for illness severity and degree
of vasoactive infusion use.36 These findings further support
that vasoactive use should not necessarily prohibit initiation
of EN in pediatric patients receiving ECMO.
We have included a suggested nutrient delivery algorithm
(Figure 1) for pediatric patients receiving ECMO. Because
there is wide variability in patient factors and underlying
disease processes, it is meant as a starting place as considera-
tion is given to the individual patient’s needs and condition.
Nutrition-Related Outcomes in Neonatal
ECMO Survivors
There are limited studies relevant to nutrition in patients
who survive ECMO, and most of this research, all on
neonates, has focused on the CDH population. There have
been reports of feeding difficulties in neonates after ECMO.
One study found that antroduodenal motility was impaired
in 10 neonatal ECMO survivors (excluding congenital heart
disease and CDH), and this was associated with failure to
achieve full feeds, longer hospitalization, and diagnosis of
reflux, and predicted the need for feeding tube placement.
Abnormalities were noted on head computed tomographic
scans in the majority of this cohort, which may suggest a
central mechanism for feeding difficulty in some ECMO
neonatal survivors.37
Among newborns who receive ECMO therapy, those
with CDH have longer duration of ECMO and more
complicated hospital courses with higher rates of bron-
chopulmonary dysplasia, gastroesophageal reflux, feeding
Farr et al 751

Detailed nutritional Enteral advancement plan:

• Start with continuous feeds at
assessment within 24h: 10 mL/kg/day and advance
• Calculate estimated energy every 6-12 h
requirement using • Goal delivery of 80%
At 24-48h assess hemodynamic stability: estimated energy requirement
Schofield or WHO
• Stable vasoactive medications by one week
equation.
• Fluid resuscitation complete • Consider PN if unlikely to
• Protein minimum 1.5
• No concern for NEC achieve 50% of EN goal by
g/kg/day day 3-5
• Stable lactate levels

Yes
Signs of EN intolerance:
• Increased girth, distension or
No emesis
• Hold feeds for 2h then
Contraindications for EN: Consider reassess, restart if
• Unrepaired CDH starting PN improvement
• Significant ileus • Consider use of pro-kinetic
• Other prohibitive agents and bowel regimen
• If gastric feeds not tolerated or
abdominal pathology
high risk of aspiration
consider post-pyloric feeds
Consider starting EN:
Yes No • Trophic feeds with slow
advancement
• Monitor for intolerance
Consider PN if unable to
tolerate goal EN:
Consider • By day 3-5 for neonates
starting PN and undernourished
• By day 5-7 for others

Figure 1. Our proposed algorithm: a rational approach to nutrition support during pediatric extracorporeal membrane
oxygenation. CDH, congenital diaphragmatic hernia; EN, enteral nutrition; NEC, necrotizing enterocolitis; PN, parenteral
nutrition; WHO, World Health Organization.

problems, and hypotonia. At 1 year of age, the CDH
population had adequate weight gain but had a feeding
tube requirement in 36%, as opposed to the patients with
meconium aspiration who had all transitioned to full oral
feeds.38 Recently, a prospective longitudinal study reported
growth outcomes up to 12 years in 172 patients with CDH,
43 of whom required ECMO. Growth was persistently
lagging in both groups, and the ECMO group had the
poorest growth. This trend continued into adolescence,
demonstrating stunting in both groups that was most severe
in the ECMO population. Notably, none of the patients
were feeding tube dependent by 8 years of age.39 This study
highlights the ongoing nutrition challenges of children with
CDH and the need for close follow-up and multidisciplinary
interventions.
needs requires further investigation. The optimal timing
of PN use in patients receiving ECMO needs to be
determined. In a subgroup of patients receiving chronic
ECMO support, such as those receiving VV ECMO as a
bridge to lung transplant, there is an increasing recognition
of the role of physical therapy along with nutrition to
optimize rehabilitation.40,41 Attention to optimal nutrient
delivery, preferably via the enteral route, along with active
rehabilitation has been safely achieved in these patients.42
Many centers have developed strategies to minimize
immobilization, provide passive physical therapy at the
bedside, and provide active therapy once feasible in patients
receiving ECMO. The safety and effect of exercise with
nutrition on muscle mass and strength in ECMO survivors
needs to be further studied.

Directions for Future Research Summary

The accurate assessment of energy and protein needs in Optimal nutrition is an important goal for infants and
children receiving ECMO is an important area for future children who are receiving ECMO therapy. Although
research. The impact of the mode of ECMO support specific studies in this vulnerable group of patients are
and underlying diagnoses on REE and macronutrient scant, there have been some important observations made.
752
There is potential for energy overfeeding from inaccurate
estimates of energy requirement and inadequate protein
delivery in the setting of higher requirements in this
group. The safety of EN has been demonstrated, and an
algorithmic approach with low-volume feeds early, followed
by cautious advancement as tolerated is prudent. Future
studies must explore the impact of bundled interventions,
including optimal nutrition and physical therapy, on clinical
and functional outcomes.
Statement of Authorship
B. J. Farr, S. E. Rice-Townsend, and N. M. Mehta equally
contributed to the conception and design of the manuscript;
B. J. Farr, S. E. Rice-Townsend, and N. M. Mehta contributed
to the acquisition and analysis of the data; B. J. Farr drafted the
manuscript. All authors critically revised the manuscript, agree
to be fully accountable for ensuring the integrity and accuracy
of the work, and read and approved the final manuscript.
References
1. Butt W, MacLaren G. Extracorporeal membrane oxygenation 2016: an
update. F1000Research. 2016;5:F1000 Faculty Rev-750.
2. Keshen TH, Miller RG, Jahoor F, Jaksic T. Stable isotopic quantitation
of protein metabolism and energy expenditure in neonates on- and
post-extracorporeal life support. J Pediatr Surg. 1997;32(7):958-963.
3. Li J, Zhang G, Herridge J, et al. Energy expediture and caloric and
protein intake in infants following the Norwood procedure. Pediatr Crit
Care Med. 2008;9(1):55-61.
4. Mehta NM, Costello JM, Bechard LJ, et al. Resting energy expenditure
after fontan surgery in children with single-ventricle heart defects.
JPEN J Parenter Enteral Nutr. 2012;36(6):685-692.
5. Jaksic T, Shew SB, Keshen TH, Dzakovic A, Jahoor F. Do critically
ill surgical neonates have increased energy expenditure? J Pediatr Surg.
2001;36(1):63-67.
6. Li X, Yu X, Cheypesh A, Li J. Non-invasive measurements of energy
expenditure and respiratory quotient by respiratory mass spectrometry
in children on extracorporeal membrane oxygenation—a pilot study.
Artif Organs. 2015;39(9):815-819.
7. Mehta NM, Skillman HE, Irving SY, et al. Guidelines for the provision
and assessment of nutrition support therapy in the pediatric critically
ill patient: Society of Critical Care Medicine and American Society
for Parenteral and Enteral Nutrition. JPEN J Parenter Enteral Nutr.
2017;41(5):706-742.
8. Jaksic T, Hull MA, Modi BP, Ching YA, George D, Compher C.
A.S.P.E.N. clinical guidelines: nutrition support of neonates supported
with extracorporeal membrane oxygenation. JPEN J Parenter Enteral
Nutr. 2010;34(3):247-253.
9. Weber TR, Shah M, Stephens C TT. Nitrogen balance in patients
treated with extracorporeal membrane oxygenation. J Pediatr Surg.
1993;28(7):906-908.
10. Agus MSD, Javid PJ, Ryan DP, Jaksic T. Intravenous insulin decreases
protein breakdown in infants on extracorporeal membrane oxygena-
tion. J Pediatr Surg. 2004;39(6):839-844.
11. Anton M, Papacostas M, Lee E, Nakonezn P, Green M. Underweight
status is an independent predictor of in-hospital mortality in pediatric
patients on extracorporeal membrane oxygenation. JPEN J Parenter
Enteral Nutr. 2016;201(1):1-8.
12. Mehta N, Bechard L, Cahill N, Wang M. Nutritional practices and
their relationship to clinical outcomes in critically ill children—an in-
ternational multicenter cohort study. Crit Care Med. 2012;40(7):2204-
2211.
Nutrition in Clinical Practice 33(6)
13. Wong JJ-M, Han WM, Sultana R, Loh TF, Lee JH. Nutrition delivery
affects outcomes in pediatric acute respiratory distress syndrome.
JPEN J Parenter Enteral Nutr. 2016;41(6):1007-1013.
14. Mikhailov TA, Kuhn EM, Manzi J, et al. Early enteral nutrition
is associated with lower mortality in critically ill children. JPEN
J Parenter Enteral Nutr. 2014;38(4):459-466.
15. Mehta NM, Bechard LJ, Zurakowski D, Duggan CP, Heyland DK.
Adequate enteral protein intake is inversely associated with 60-d
mortality in critically ill children: a multicenter, prospective, cohort
study. Am J Clin Nutr. 2015;102(1):199-206.
16. Estensen K, Shekar K, Robins E, McDonald C, Barnett AG, Fraser
JF. Macro- and micronutrient disposition in an ex vivo model of
extracorporeal membrane oxygenation. Theor Chem Acc. 2014;2(1):29.
https://icm-experimental.springeropen.com/articles/10.1186/s40635-
014-0029-7
17. McDonald CI, Fung YL, Shekar K, et al. The impact of acute lung
injury, ECMO and transfusion on oxidative stress and plasma selenium
levels in an ovine model. J Trace Elem Med Biol. 2015;30:4-10.
18. Datzmann T, Träger K, Reinelt H, Von Freyberg P. Elimination rates
of electrolytes, vitamins, and trace elements during continuous renal re-
placement therapy with citrate continuous veno-venous hemodialysis:
influence of filter lifetime. Blood Purif. 2017;44:210-216.
19. Broman M, Bryland A CO. Trace elements in patients on continuous
renal replacement therapy. Acta Anaesthesiol Scand. 2017;61(6):650-
659.
20. Armstrong LB, Ariagno K, Smallwood CD, Hong C, Arbuthnot
M, Mehta NM. Nutrition delivery during pediatric extracorporeal
membrane oxygenation therapy. JPEN J Parenter Enteral Nutr.
2018;42(7):1133-1138.
21. Fivez T, Kerklaan D, Mesotten D, et al. Early versus late parenteral
nutrition in critically ill children. N Engl J Med. 2016;374(12):1111-
1122.
22. Piena M, Albers MJIJ, Van Haard PMM, Gischler S, Tibboel D.
Introduction of enteral feeding in neonates on extracorporeal mem-
brane oxygenation after evaluation of intestinal permeability changes.
J Pediatr Surg. 1998;33(1):30-34.
23. Pettignano R, Heard M, Davis R, Labuz M, Hart M. Total enteral nu-
trition versus total parenteral nutrition during pediatric extracorporeal
membrane oxygenation. Crit Care Med. 1998;26(2):358-363.
24. Hanekamp MN, Spoel M, Sharman-Koendjbiharie I, Peters JWB,
Albers MJIJ, Tibboel D. Routine enteral nutrition in neonates
on extracorporeal membrane oxygenation. Pediatr Crit Care Med.
2005;6(3):275-279.
25. Wertheim HFL, Albers MJIJ, Piena-Spoel M, Tibboel D. The inci-
dence of septic complications in newborns on extracorporeal mem-
brane oxygenation is not affected by feeding route. J Pediatr Surg.
2001;36(10):1485-1489.
26. Greathouse KC, Sakellaris KT, Tumin D, et al. Impact of early ini-
tiation of enteral nutrition on survival during pediatric extracorporeal
membrane oxygenation. JPEN J Parenter Enteral Nutr. 2018;42(1):205-
211.
27. Scott LK, Boudreaux K, Thaljeh F, Grier LR, Conrad SA. Early en-
teral feedings in adults receiving venovenous extracorporeal membrane
oxygenation. JPEN J Parenter Enteral Nutr. 2004;28(5):295-300.
28. Umezawa Makikado LD, Flordelı́s Lasierra JL, Pérez-Vela JL, et al.
Early enteral nutrition in adults receiving venoarterial extracorporeal
membrane oxygenation. JPEN J Parenter Enteral Nutr. 2013;37(2):281-
284.
29. Ferrie S, Herkes R, Forrest P. Nutrition support during extracorporeal
membrane oxygenation (ECMO) in adults: a retrospective audit of
86 patients. Intensive Care Med. 2013;39(11):1989-1994.
30. Desmarais TJ, Yan Y, Keller MS, Vogel AM. Enteral nutrition in
neonatal and pediatric extracorporeal life support: a survey of current
practice. J Pediatr Surg. 2015;50(1):60-63.
Farr et al
31. Mehta NM. Parenteral nutrition in critically ill children. N Engl J Med.
2016;374(12):1190-1192.
32. Jimenez L, Mehta NM, Duggan CP. Timing of the initiation of
parenteral nutrition in critically ill children. Curr Opin Clin Nutr Metab
Care. 2017;20(3):227-231.
33. Buck ML, Ksenich RA, Wooldridge P. Effect of Infusing Fat Emulsion
into Extracorporeal Membrane Oxygenation Circuits. Pharmacother-
apy. 1997;17(6):1292-1295.
34. Buck ML, Wooldridge P, Ksenich RA. Comparison of methods for
intravenous infusion of fat emulsion during extracorporeal membrane
oxygenation. Pharmacotherapy. 2005;25(11):1536-1540.
35. Brogan TV, Lequier L, Lorusso R, MacLaren G, Peek G, eds. Extra-
corporeal Life Support: The ELSO Red Book. 5th ed. Ann Arbor, MI:
Extracorporeal Life Support Organization; 2017.
36. Panchal AK, Manzi J, Connolly S, et al. Safety of enteral feedings
in critically ill children receiving vasoactive agents. JPEN J Parenter
Enteral Nutr. 2016;40(2):236-241.
37. Jadcherla SR, Berseth CL. Antroduodenal motility and feeding out-
come among neonatal extracorporeal membrane oxygenation sur-
vivors. J Pediatr Gastroenterol Nutr. 2005;41(3):347-350.
753
38. Bernbaum J, Schwartz IP, Gerdes M, et al. Survivors of extracorporeal
membrane oxygenation at 1 year of age: the relationship of primary
diagnosis with health and neurodevelopmental sequelae. Pediatrics.
1995;96(5 Pt 1):907-913.
39. Leeuwen L, Mous DS, van Rosmalen J, et al. Congenital diaphragmatic
hernia and growth to 12 years. Pediatrics. 2017;140(2):e20163659.
https://https://doi.org/10.1542/peds.2016-3659
40. Ulerich L. Nutrition implications and challenges of the transplant
patient undergoing extracorporeal membrane oxygenation therapy.
Nutr Clin Pract. 2014;29(2):201-206.
41. Hayes D, Tobias JD, Galantowicz M, Preston TJ, Tzemos KK, Mc-
Connell PI. Video fluoroscopy swallow study and nutritional support
during ambulatory venovenous extracorporeal membrane oxygenation
as a bridge to lung transplantation. World J Pediatr Congenit Hear Surg.
2014;5(1):91-93.
42. Zebuhr C, Sinha A, Skillman H, Buckvold S. Active rehabilitation
in a pediatric extracorporeal membrane oxygenation patient. PM R.
2014;6(5):456-460.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Metabolic Support of the Patient on Continuous Renal December 2018 754–766

C 2018 American Society for

Replacement Therapy Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10208
wileyonlinelibrary.com

Erin M. Nystrom, PharmD, BCNSP; and Andrea M. Nei, PharmD, BCPS, BCCCP

Abstract
Continuous renal replacement therapy (CRRT) is the modality of choice in critically ill patients with hemodynamic instability
requiring renal replacement therapy. The goal of this review is to discuss an overview of CRRT types, components, and important
considerations for nutrition support provision. Evidence basis for guidelines and our recommendations are reviewed. Nutrition
support–related implications include the possibility of calorie gain with citrate-based anticoagulation, calorie loss with glucose-
free replacement fluids and dialysate, and significant amino acid losses in effluent. We challenge nutrition support clinicians to
develop a keen understanding of the specific CRRT modalities that are employed in their intensive care units and to be able to
determine how the CRRT prescription may impact a patient’s nutrition support prescription. (Nutr Clin Pract. 2018;33:754–766)

Keywords
critical care; dialysis; intensive care unit; nutritional support; renal replacement therapy

Introduction The CRRT circuit lends significant complexity given im-

Acute kidney injury (AKI) is a common complication in
critically ill patients, with a reported incidence of >30%,
depending on definitions used.1-3 Approximately 4%–6% of
patients with AKI in the intensive care unit (ICU) require a
form of renal replacement therapy (RRT).1,4-6 Continuous
renal replacement therapy (CRRT) offers multiple theoreti-
cal advantages over intermittent hemodialysis (IHD), but is
primarily used in the presence of hemodynamic instability
and/or need for aggressive fluid removal (ultrafiltration) in
the setting of massive hypervolemia.7
Malnutrition has been estimated to afflict up to 42% of
critically ill patients with AKI.8 Factors contributing to mal-
nutrition in critically ill individuals on RRT are multifacto-
rial. The acute stress response of critical illness leads to a
release of regulatory endocrine hormones (catecholamines,
cortisol, and glucagon) and proinflammatory cytokines that
contribute to net gluconeogenesis and glycogenolysis.9-11
This ultimately results in net skeletal muscle breakdown and
lipolysis to provide necessary substrates for cellular function
and contributes to tissue insulin resistance and inhibition
of lipoprotein lipase. So, although substrate plasma levels
are elevated, use by peripheral tissues may be reduced.11,12
Additional complicating factors affecting nutrition status in
patients with renal failure requiring RRT include decreased
intake prior to and during admission, concomitant organ
failure, and nutrient losses from sources such as RRT,
drains, and wounds.
portant macronutrient, micronutrient, and electrolyte losses
or gains that must be accounted for when formulating a nu-
trition support regimen. A general understanding of CRRT
types and prescription variables is thus essential for the
clinician responsible for the nutrition prescription. While
guidelines have attempted to provide a general framework
for practice, it is challenging to extrapolate dated studies of
macronutrient disposition in CRRT to contemporary clini-
cal practice given advancements in CRRT technology, dos-
ing, and types of fluids used. Whereas electrolytes and fluids
have more specific indicators to guide requirements via lab,
physical exam, or other clinical parameters, macronutrient
needs are more difficult to define in the absence of surrogate
markers that correlate with clinical outcomes.
The goal of this review is to discuss an overview of
CRRT types, components of the CRRT prescription,
and important considerations related to nutrition support
From the Department of Pharmacy, Mayo Clinic, Rochester,
Minnesota, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on October 15, 2018.
Corresponding Author:
Erin M. Nystrom, PharmD, BCNSP, Mayo Clinic, 200 First Street
SW, Rochester, MN 55902.
Email: nystrom.erin@mayo.edu
Nystrom and Nei 755

Figure 1. Diagram of diffusion and convection removal of solutes in hemofilter. Differences in mechanism of solute removal
comparing diffusion and convection across the hemofilter membrane are represented. The hemofilter has multiple hollow-fiber
tubules running blood through the device to allow maximal contact with hemofilter membrane. In diffusion, dialysate runs
countercurrent to the blood flow using a concentration gradient to allow removal of small solutes. In convection, the
transmembrane pressure, higher in the blood compartment and lower in the effluent/ultrafiltrate compartment, drives fluid across
the membrane, dragging small and medium solutes with it.

provision. We challenge nutrition support clinicians to un-
derstand the specific CRRT modalities and types of CRRT
anticoagulation, dialysate, and/or replacement fluids em-
ployed in their ICUs, as well as to determine how a patient’s
CRRT prescription may impact his or her nutrition support
regimen.
CRRT Overview
Multiple CRRT modalities are available, each designated
by location of vascular access and mechanism of solute
removal (Table 1). Vascular access is designated as either
veno-venous (VV) or arterio-venous (AV), indicating the lo-
cation of circuit entry and exit from the patient’s circulation.
AV CRRT, which uses a patient’s arterial blood pressure to
drive blood through the filter, has largely been abandoned
in favor of VV CRRT given the availability of CRRT
pumps.
The 2 mechanisms for solute removal in CRRT are
diffusion and convection (Figure 1). In diffusion, a dialysate
solution runs countercurrent to blood in the dialysis filter,
creating a concentration gradient across the dialysis mem-
brane and driving solute removal from the blood. Solute
removal is dependent on molecular weight, with smaller
molecules more readily removed, eg, creatinine, urea, and
electrolytes. Convection, also referred to as ultrafiltration,
uses a high transmembrane pressure gradient in the hemofil-
ter to force plasma water across a semipermeable membrane,
dragging solute with it (solvent drag). Convection is effec-
tive at removing both small-size and medium-size solutes
(Table S1). Contemporary practice uses high-flux mem-
branes, allowing better clearance of middle molecules. The
term sieving coefficient in convection is the ratio of solute
in effluent to blood, and is the primary determinant of how
readily a solute is cleared. Sieving coefficients range from 0
to 1, with a value of 1 indicating free passage of a molecule
through the membrane.
Continuous VV hemodialysis (CVVHD) is most sim-
ilar to IHD in that it uses diffusion to achieve solute
removal, but in a continuous manner, with dialysate run
countercurrent to blood flow. Continuous VV hemofiltra-
tion (CVVH) differs from CVVHD in its use of convec-
tion for fluid and solute removal. Due to the need for
a high ultrafiltration rate to achieve removal of solutes,
replacement fluids are required to replenish volume and
electrolytes. Continuous VV hemodiafiltration (CVVHDF)
uses both diffusion and convection for solute clearance and
thus requires use of both replacement and dialysate fluids.
No CRRT modality has been shown to be superior to
another.
Other modalities of RRT that are administered in a
prolonged (but not continuous) manner in comparison
with typical intermittent dialysis sessions include sustained
low-efficiency dialysis (SLED) and sustained continuous
ultrafiltration (SCUF). SLED uses diffusion to achieve
solute removal, but over a longer interval than IHD. SCUF
uses convection at a lower ultrafiltration rate than CVVH,
primarily to achieve volume removal. Selection of modality
is dependent on institution preference, fluid status, and
metabolic/electrolyte abnormalities.
756
Table 1. RRT Modalities.
Type Solute Clearance Use of RF
CVVH Convection Yes
CVVHD Diffusion No
CVVHDF Convection and Diffusion Yes
SLED Diffusion No
SCUF Convection No
CVVH, continuous veno-venous hemofiltration; CVVHD, continuous veno-venous hemodialysis; CVVHDF, continuous veno-venous
hemodiafiltration; RF, replacement fluids; RRT, renal replacement therapy; SCUF, slow continuous ultrafiltration; SLED, sustained
low-efficiency dialysis.
CRRT Prescription and Implications
for Nutrition Support
Since the early 1970s when CRRT was first administered,
significant advancements have been made in pump and filter
technology, replacement fluids, and anticoagulation. It is
vital to understand how CRRT has evolved to appropri-
ately evaluate available CRRT literature related to nutri-
tion support. Additionally, understanding the components
that make up a CRRT prescription and relevant solutes
removed or gained (glucose [dextrose], citrate, amino acids)
is fundamental to applying literature to clinical practice.
Recognizing advancements in CRRT technology in the last
decade,13 specific variables for consideration include the
following:
r Filter type: Contemporary high-flux, high-
permeability filters allow larger molecules to
pass through the membrane and also provide more
efficient solute removal.
r CRRT dose: CRRT dose is defined as the effluent
flow rate. The effluent volume is comprised of the
total ultrafiltration (convection) and dialysate (dif-
fusion) volume. For example, in CVVH the CRRT
dose is defined only by the ultrafiltration rate, while in
CVVHDF, it is a combination of ultrafiltration and
dialysate fluid rate. Guidelines generally recommend
a delivered rate of 20–25 mL/kg/hr.7 The higher the
CRRT dose, the higher the solute removal.
r Blood flow rate: Typically 100–300 mL/min, change
in blood flow has a relatively small impact on solute
clearance unless the effluent rate is increased pro-
portionately, as hemofiltration solution is otherwise
saturated.
r CRRT mode: Use of diffusion, convection, or both
determines the size and quantity of solutes removed
as well as the need for replacement fluids (con-
vection). Convective clearance allows removal of
larger solutes compared with diffusive clearance. The
glucose composition of replacement fluids required
to replace fluid removal with convection can signifi-
cantly affect energy gain or loss.
Nutrition in Clinical Practice 33(6)
Use of Dialysate Duration Anticoagulation
No 24 hours Yes
Yes 24 hours Yes
Yes 24 hours Yes
Yes 6–12 hours Usually not
No 6–12 hours Usually not
r Anticoagulation: Citrate anticoagulation can be a
significant source of carbohydrate calories from cit-
rate and dextrose. Citrate solutions available for
CRRT anticoagulation include acid citrate dextrose,
formula-A ([ACD-A], 2.45% dextrose and 2.2% cit-
rate) and 4% trisodium citrate (TSC). Delivery of
citrate and glucose is dependent on the rate of
infusion.
r Replacement or dialysate fluids: The dextrose (glu-
cose) composition of these fluids, ranging from 0 to
110 mg/dL, impacts overall net delivery of glucose.
Fluids without dextrose may contribute to net losses
of glucose in the effluent, while fluids that contain
dextrose will have minimal impact on net glucose gain
or loss in the effluent. Also, replacement fluids may
be administered prefilter, postfilter, or both, which
could impact overall glucose delivery; higher prefilter
glucose concentrations may lead to greater net loss of
glucose in the CRRT circuit.
Select variables with recognized impact on nutrient dis-
position will herein be reviewed.
Replacement and Dialysate Fluids
Replacement fluids are used with CVVH and CVVHDF to
replenish electrolytes and plasma volume removed during
convection. Replacement fluids are typically infused pre-
filter, postfilter, or both (Figure 2). Prefilter administration
of replacement fluids helps minimize circuit clotting, but
may decrease efficiency of solute removal from the blood
due to prefilter plasma dilution. Postfilter administration
allows direct delivery of desired replacement solute to
systemic circulation.14
Dialysate solutions are used in CVVHD and CVVHDF
to achieve diffusive clearance of solute. Dialysate is gener-
ally infused into the dialysis filter countercurrent to plasma
blood flow (Figures 1 and 2). Efficacy of solute removal,
in addition to solute characteristics, is dependent on both
the degree of concentration gradient across the membrane
and the dialysate rate, with an increased rate associated with
higher solute clearance.
Nystrom and Nei 757

Figure 2. CRRT modalities. (A) CVVH, (B) CVVHD, and (C) CVVHDF. In all 3 modalities, anticoagulation is infused early in
the venous side of the CRRT circuit. Replacement fluids are administered in CVVH and CVVHDF and may be given prefilter,
postfilter, or both. Total replacement fluid rate is approximate to effluent removal rate (CRRT dose). In CVVHD and CVVHDF,
dialysate fluids enter the hemofilter and run countercurrent to blood flow. In all 3 modalities the effluent bag contains fluid and
solutes removed from the CRRT circuit. Glucose losses in the effluent are variable and depend on glucose content of
replacement/dialysate fluids in addition to components of the CRRT prescription. *There are multiple types of anticoagulation; 1
example is ACD-A, and others include TSC, heparin, or no anticoagulation. Typical ACD-A rate is 100–300 mL/h, but may be
altered depending on institutional-specific protocols and alterations to the CRRT prescription. ACD-A, acid citrate dextrose,
formula-A; CRRT, continuous renal replacement therapy; CVVH, continuous veno-venous hemofiltration; CVVHD, continuous
veno-venous hemodialysis; CVVHDF, continuous veno-venous hemodiafiltration; TSC, trisodium citrate.
758 Nutrition in Clinical Practice 33(6)

Figure 2. Continued.

Table 2. Electrolyte Composition of Dialysate and ids with specific electrolyte content. Bicarbonate-based
Replacement Fluids. replacement fluids and dialysate solutions are now pre-
ferred over lactate-based solutions due to improved correc-
Component Concentration
tion of acidosis, lack of lactate accumulation, and better
Sodium 136–140 mmol/L hemodynamic tolerance.15,16 Lactate is a calorie source,
Potassium 0–4 mmol/L and, as such, studies demonstrating high caloric gain
Calcium 0–3.5 mEq/L with lactate-based replacement fluids must be taken into
Magnesium 0.5–1.5 mEq/L context.
Phosphate 0–1.25 mmol/L Note that the dextrose component of replacement and
Chloride 107.5–120.5 mmol/L
Bicarbonate 25–35 mmol/L
dialysate solutions is variable, ranging from 0 to 110 mg/dL,
Lactate 0–38 mmol/L which may result in net glucose (energy) loss due to unre-
Dextrose 0–110 mg/dL placed effluent losses of glucose if glucose-free fluids are
used.

Selection of replacement fluids and dialysate solu-

tions is based on a patient’s specific metabolic, acid–
base, and electrolyte profile. A variety of commercially
available replacement fluids and dialysate solutions, with
varying electrolyte and buffer components, are available
(Tables 2 and S2). Some institutions may compound flu-
Anticoagulation
Due to prolonged contact time between blood and the extra-
corporeal CRRT circuit, anticoagulation is recommended
to prevent filter clotting and optimize permeability and dial-
ysis efficiency. The type of anticoagulation used in CRRT
varies among institutions, with the most common options
Nystrom and Nei
being regional citrate anticoagulation (RCA), heparin, or
none.17 Current guidelines recommend RCA as first-line
therapy for anticoagulation in the absence of contraindica-
tions to citrate.7 RCA has been shown to prolong filter life,
decrease incidence of spontaneous filter failure, and reduce
bleeding compared with systemic heparin.18 Citrate achieves
local anticoagulation by chelating calcium within the circuit,
thus inactivating an essential component of the clotting
cascade. A postfilter infusion of calcium into the circuit is
necessary to regain a normal systemic calcium concentra-
tion. When blood from the CRRT circuit is infused back
into the systemic circulation, citrate is rapidly metabolized
by the liver, muscle, and kidney, producing bicarbonate.
Citrate is typically infused at a fixed rate with monitoring
of systemic total and ionized calcium levels for evidence of
citrate toxicity, suspected when the total calcium to ionized
calcium ratio is >2.5,19 or titrated to a low intracircuit
ionized calcium goal. Possible side effects of RCA include
metabolic alkalosis and hypocalcemia. Patients with liver
failure or profound shock with decreased muscle perfusion
are at highest risk of citrate accumulation and subsequent
side effects.19
Products available for RCA include ACD-A, which is
2.2% citrate formulated in a 2.45% dextrose solution, and
4% TSC. Given that citrate and dextrose are sources of
carbohydrate, energy contributed with use of RCA can be
clinically significant.
If citrate is not an option for anticoagulation, heparin
anticoagulation is recommended in patients who are not
at high risk for bleeding. In the setting of heparin-induced
thrombocytopenia, alternative agents such as direct throm-
bin inhibitors or fondaparinux may be considered after
weighing risks and benefits of therapy.7 If both citrate and
heparin are contraindicated, no anticoagulation may be
used with close monitoring for circuit and filter thrombosis.
Energy Requirements
Defining energy requirements in critical illness has been the
subject of much debate, lacking the input of high-quality
study data. Similarly, no randomized controlled studies have
defined optimal energy provision in CRRT patients, for
whom energy requirements have not been distinguished to
be different than patients without AKI. In a secondary
analysis of 1456 patients with AKI requiring CRRT, neither
higher daily caloric intake of at least 25 kcal/kg/d nor calorie
intake itself was associated with differences in RRT-free,
ICU-free, ventilation-free, or hospital-free days.20
Thus, we turn our attention to the avoidance of over
feeding, as CRRT fluids can be a significant source of
calories. Because metabolic alterations of critical illness and
AKI complicate the body’s ability to use non protein sources
of calories, it is especially important to avoid overfeeding,
which can result in accumulation of liver fat, increased
759
carbon dioxide production, and increased infection risk.21-23
Additionally, the energy prescription should be based on
a dry weight, which may be challenging to determine
in the hypervolemic patient. Hyperglycemia, high insulin
requirements, and hypertriglyceridemia may be indicators
of intolerance to the dextrose and/or lipid emulsion load
and warrant reassessment of the nutrition prescription with
an accounting of the possible contributions from CRRT
fluids.
Carbohydrate Gains and Losses With CRRT
When determining the amount of calories to provide via
enteral nutrition (EN) or parenteral nutrition (PN), clini-
cians should routinely assess for dextrose-containing fluids
administered with intravenous (IV) medications or as bolus
or continuous crystalloid infusions as well as the potential
for energy gain via fluids used in the CRRT circuit. Carbo-
hydrate sources in the CRRT circuit include:
r Citrate (3 kcal/g24,25 ) from RCA of ACD-A (2.2%
citrate) and TSC (4% trisodium citrate);
r Glucose (3.4 kcal/g) from ACD-A (2.45% dextrose)
and replacement and dialysate fluids (0–110 mg/dL);
r Lactate (3.62 kcal/g26 ), of lower importance in con-
temporary practice in replacement and dialysate
fluids.
Table 3 summarizes studies evaluating delivery of citrate,
lactate, and glucose associated with CRRT.27-29 Notice that
energy gain can be substantial, depending on the type and
rate of fluids used, with 1 study reporting up to 1300 kcal/d
with high-lactate replacement fluids and anticoagulation
with ACD-A.27 Variability in energy gain was noted in the
2 studies by Balik et al, depending on the lactate content of
replacement fluids and type of anticoagulation.27,28 Note-
worthy is the specific impact of replacement fluids, with
or without glucose, and citrate anticoagulation, on calorie
gain.
The most recent study published on this topic confirmed
the potential for significant caloric contribution, specifically
with CVVH. New et al measured energy uptake with
use of guideline-recommended citrate anticoagulation as
ACD-A and bicarbonate-based replacement fluids (glucose
110 mg/dL). Of the 10 patients studied, 8 received ACD-
A at a rate of 300 mL/h (1 patient received ACD-A
at 200 mL/h and 1 patient received both rates during the
study time period), with an average ultrafiltration rate of
30 mL/kg/h and blood flow rate of 200 mL/min. The study
reported an average daily delivery of 513 kcal/d (citrate
218 kcal/d, glucose 295 kcal/d).30 Investigators were unable
to determine differences in energy uptake comparing differ-
ent rates of ACD-A due to low numbers in the 200 mL/h
group. Additionally, the impact of glucose-free replacement
fluids on energy gain was not assessed. These factors limit
760 Nutrition in Clinical Practice 33(6)

Table 3. Carbohydrate Studies in CRRT.

Study CRRT Type CRRT Fluids Key Study Conclusions

Balik et al, CVVH ACD-A/lactate fluidsa r Average calorie gain 1318 kcal/d.
201227 (n = 18) r Net loss noted with bicarbonate fluidsa +
CVVHDF heparin (control).
(n = 23) r No difference in citrate, lactate, or glucose
delivery between CVVH and CVVHDF.
r Significant caloric gain noted; however, high-
lactate fluids are no longer recommended by
guidelines, and heparin was used for
anticoagulation in the control group, limiting
clinical applicability.
Balik et al, CVVHDF ACD-A/lactate fluidsa (n = 29) r Calorie gain (per day):
201328 (n = 81) vs TSC/bicarbonate fluidsa ◦ ACD-A/lactate fluids 1078 kcal
(n = 34) vs heparin/lactate ◦ TSC/bicarbonate fluids 115 kcal
fluidsa (n = 18) ◦ Heparin/lactate fluids 361 kcal
◦ Calculated ACD-A/bicarbonate fluids
457 kcal/d
r Kcal delivery in heparin group was solely
related to lactate fluids.
r Different blood flows used in each group
affected citrate rate.
r Net caloric gain noted in all groups, but
delivery was impacted by type of
anticoagulation and fluid used; ACD-A and
lactate-based fluids provided highest gain.
r Use of lactate fluids limits clinical application
in current practice.
Stevenson et al, CVVH Bicarbonate dialysate fluids r In vitro study comparing glucose kinetics
201329 CVVHD with and without glucose among CRRT types.
CVVHDF r Calculations in CVVH and CVVHDF
assumed replacement fluids without glucose.
r Net glucose losses noted with all CRRT
modalities when bicarbonate fluids were used
without glucose.
r Glucose loss decreased with inclusion of
glucose 100 mg/dL in dialysate fluids.
r ACD-A with CVVHD + bicarbonate dialysate
without glucose ˗120 to +470 kcal/d.
r Citrate delivery was not assessed (only
glucose).
r Glucose losses noted with non-glucose-
containing dialysate and replacement fluids;
glucose losses decreased with addition of
glucose to dialysate fluids and with citrate
anticoagulation.
New et al, CVVH ACD-A/bicarbonate r Average 513 kcal/d
201730 (n = 10) replacement fluidsa r Contemporary CVVH regimen using
guideline-recommended fluids.
r Notable caloric delivery from ACD-A
anticoagulation and bicarbonate replacement
fluids.a

ACD-A, acid citrate dextrose, formula-A; CRRT, continuous renal replacement therapy; CVVH, continuous veno-venous hemofiltration;
CVVHD, continuous veno-venous hemodialysis; CVVHDF, continuous veno-venous hemodiafiltration; TSC, trisodium citrate.
a Glucose-containing fluids (100–110 mg/dL). The term “fluids” in this table designates either replacement or dialysate fluids or both depending on

the type of CRRT studied.

Nystrom and Nei
extrapolation to clinical practice for CVVH when lower
ACD-A rates and/or glucose-free replacement fluids are
employed.
It is important to highlight how the glucose content of
replacement and dialysate solutions may impact net delivery
of carbohydrates with CRRT. Non-glucose-containing flu-
ids have the potential to contribute to glucose losses be-
cause effluent losses of glucose are left unreplaced with
use of glucose-free replacement fluids. Stevenson et al
confirmed this with in vitro models of CVVH, CVVHD,
and CVVHDF.29 The authors suggested an equation based
on logistic regression that could be used to calculate glucose
losses for any method of CRRT with any glucose concentra-
tion in dialysate, assuming use of non-dextrose-containing
replacement fluids in CVVH or CVVHDF:
y = 0.81 [QD × EC × (GPre − GD ) + QUf × EC × GPre ]
+ 4.16
y = total kcal/d removed; QD = dialysis flow rate (L/hr);
EC = extraction coefficient; GPre = glucose concentration
prefilter (mg/dL); GD = glucose concentration in dialysate
(mg/dL); QUf = ultrafiltration rate (L/hr)
Several considerations should be made if using this
equation in clinical practice. Noting glucose EC = Ef /Pre
(Ef = concentration in effluent leaving the dialysis filter,
Pre = concentration in the prefilter blood), this value (EC)
could be considered 1 for practical application given glucose
is freely filtered in both dialysis and ultrafiltration with
contemporary filters. Indeed, the value calculated in the
study was 1.04 (95% confidence interval [CI]: 1.03–1.05).
Additionally, it is not routine practice to sample immediate
prefilter glucose concentrations, which will likely be altered
compared with systemic glucose values if replacement fluids
are administered prefilter and if citrate anticoagulation
is administered. Third, this in vitro study used a fixed
blood flow rate of 200 mL/min. Finally, the equation does
not account for use of replacement fluids (in CVVH or
CVVHDF) that contain dextrose. Acknowledging these
limitations and in the absence of more specific data to guide
practice, this equation can provide a starting estimate for
bedside clinicians to estimate glucose losses in effluent.
Figure 3 illustrates a simple algorithm for estimating how
anticoagulation and replacement/dialysate solutions may
impact caloric delivery from carbohydrates in the CRRT
circuit, highlighting the prominent variables of citrate-
based anticoagulation and glucose content of replacement
fluids. Also for consideration is the CRRT dose (effluent
flow rate); the higher the dose, the higher the removal of
glucose and citrate. Additionally, replacement fluids may
be administered prefilter, postfilter, or both, which could
impact overall glucose delivery. Higher prefilter glucose
concentrations may lead to greater net loss of glucose in
761
the CRRT circuit. Lastly, the method of citrate titration is
variable among institutions, possibly leading to differences
in citrate rate, which will also impact overall delivery of
citrate and glucose.
Case Example
Patient A.B., a 70-year-old female weighing 66 kg, is ad-
mitted with septic shock secondary to bowel perforation,
complicated by AKI requiring CRRT. She is 7 days into
her ICU stay, and due to strict nothing by mouth status,
she is being considered for PN. CVVH has been prescribed
with an ultrafiltration rate of 30 mL/kg/h (2L/h), blood
flow rate 200 mL/min, and ACD-A anticoagulation at
300 mL/h. Replacement fluids are high-bicarbonate with
glucose concentration of 110 mg/dL. Based on recent study
data (New et al), we could expect ࣈ500 carbohydrate kcal/d
from her current convection-based CVVH regimen.30
A.B. later improves clinically, and her replacement fluid
is switched to a lower bicarbonate solution that does not
include glucose. As such, effluent losses of glucoses are no
longer being replaced via glucose-containing replacement
fluids, so we subtract estimated effluent glucose losses from
the 500 kcal estimated above, using the following method.
With an ultrafiltration rate of 2 L/h, we would expect
about 48 L removed over 24 hours. Effluent loss of 48 L with
an assumed prefilter glucose concentration of 100 mg/dL or
1g/L totals 48 g of estimated glucose losses or ࣈ160 kcal,
rounded to 150 kcal; 500 kcal (as previously demonstrated)
subtract 150 kcal = a rough estimate of ࣈ350 kcal/d net gain
with this CVVH program.
Alternatively, if A.B. had been receiving CVVHDF with
citrate anticoagulation (ACD-A at 300 mL/h), an ultrafiltra-
tion rate of 1 L/h, and dialysate rate of 1L/h with replace-
ment/dialysate fluids that did not contain dextrose, we could
estimate losses according to the Stevenson equation, where
EC is estimated to be 1, GPre is 100 mg/dL (highly variable),
and y is the number of kcal removed per day:
y = 0.81 [QD ×EC × (GPre − GD ) + QUf × EC × GPre ]
+ 4.16
166 kcal/d = 0.81[1 L/h × 1 × (100 mg/dL − 0)
+1 L/h × 1 × 100 mg/dL] + 4.16
Since we need to account for citrate anticoagulation as a
source for calories (not accounted for with the Stevenson
equation), we could extrapolate data from CVVH (New
et al) with total initial calories of 500 kcal and subtract
166 kcal (net loss calculated above) to total net gain of
334 kcal/d. This caloric gain is fairly similar to the estima-
tion above using only ultrafiltration (CVVH), but with a
similar total CRRT dose (ultrafiltration 1 L/h + dialysis flow
762 Nutrition in Clinical Practice 33(6)

Figure 3. Algorithm for energy gain and/or loss assessment. Flow chart depicts CRRT variables that can contribute to energy
gain or loss. *CRRT modalities of CVVHDF and CVVHD use diffusion and, thus, dialysate (not shown). Dialysate that is
dextrose free theoretically could result in calorie loss. ACD-A, acid citrate dextrose, formula-A; CRRT, continuous renal
replacement therapy; CVVH, continuous veno-venous hemofiltration; CVVHD, continuous veno-venous hemodialysis;
CVVHDF, continuous veno-venous hemodiafiltration; TSC, trisodium citrate.

rate 1 L/h = 2 L/h). It is important to note that this is an
estimation, given lack of in vivo data specific to CVVHDF
with citrate anticoagulation.
In conclusion, contemporary CRRT practices are likely
to contribute a net gain of calories from carbohydrates when
anticoagulation of ACD-A is used, noting this contribution
may vary depending on the citrate rate. This could also
be the case with other citrate-based anticoagulation, ie,
TSC. This energy gain is buffered with use of glucose-
free replacement fluids, as glucose losses in effluent are not
replaced. Notably, if citrate-based anticoagulation is not
used, an overall net energy loss may occur if glucose-free
replacement or dialysate fluids are employed. While we have
some literature to help guide estimation of energy gain
or loss with CRRT, it remains challenging to describe an
approach to incorporate all CRRT prescriptions given the
wide variability in practice. It is therefore critical to consider
individual institutional CRRT practices when determining
application of current study findings to clinical practice.
Protein (Amino Acids)
The catabolism that characterizes critical illness is com-
pounded in AKI by multiple factors, chief among them
amino acid losses in ultrafiltrate and/or dialysate with
CRRT.31-39 Quantifying absolute amino acid loss is com-
plicated by the multitude of variables that make up the
CRRT circuit and individual prescription, beyond the fact
that amino acids, with a sieving coefficient of around 1, are
readily filtered from the blood into effluent.33,35,40
CRRT-specific factors that determine amino acid clear-
ance, like that of other solutes, include use of convec-
tion (CVVH or CVVHDF) and/or diffusion (CVVHD or
CVVHDF), blood flow rate, dialysis flow rate, effluent rate,
and filter membrane properties.38,40 As such, these variables
are essential to consider when translating research into
recommendations for protein dosing in CRRT patients in
contemporary clinical practice. While core studies refer-
enced in guidelines41-43 have been limited by lack of ran-
domization and power and surrogate outcomes assessed (eg,
nitrogen balance), extrapolation is further constrained by
the evolution of CRRT technology since their publication.
Notably, newer CRRT technology allows higher blood flow
and effluent removal rates that easily double those studied
by Bellomo36,37 and Sheinkestel.38,44 Additionally 3 of these
4 studies evaluated CRRT modalities that used diffusion
(continuous arteriovenous hemodiafiltration [CAVHD] or
CVVHD), leaving only 7 adult patients studied on a CRRT
modality that employed convective clearance (CVVHDF).37
Maxvold and colleagues, in a study of 6 pediatric patients
randomized to either CVVH or CVVHD and crossed over
to the alternate modality after 24 hours, demonstrated
a 30%–40% higher amino acid clearance with convection
(CVVH) than with diffusion (CVVHD).35
Nystrom and Nei
To our knowledge, the most recent study evaluating
amino acid disposition in CRRT is that by Zappitelli et al,
who studied 15 critically ill children receiving CVVHD
and found a 10% loss of total amino acid intake and,
more specifically, a 17%–22% loss when amino acids were
provided by PN.39 This suggests the route of nutrition
may have implications for amino acid loss in dialysate. A
relationship between blood levels of amino acids, rather
than amino acid intake, and effluent losses has been demon-
strated by others, suggesting nutrition parameters, such as
rate of parenteral amino acid infusion, may be a factor in
losses with CRRT.33,34,38 The clinical implications of this are
unclear.
Research questions on protein provision in critical ill-
ness along with outcomes for evaluation have recently
been proposed.45 Additional questions related to protein
provision in patients receiving CRRT are in dire need of
investigation. Studies measuring amino acid loss in effluent
with contemporary CRRT modalities would be instructive
in the absence of clinical outcomes, although studies of
effluent losses should not be considered equal to those that
evaluate the effect of protein provision on clinical outcomes
in patients receiving CRRT. Specific queries evaluating
variables of convection (CVVH), diffusion (CVVHD), or
combination convection/diffusion (CVVHDF), amino acid
intake, rate of parenteral administration of amino acid, and
route of amino acid administration are warranted.
In the meantime, the published research and our clinical
experience is reassuring in that a high amino acid dose
(ie, 2.0–2.5 g/kg) as recommended41,42 can be adminis-
tered successfully without increasing blood urea nitrogen
levels or generating disproportionate amino acid loss (ie,
waste).33,34,36,37 Given the prevalent use of convection-
based modalities of CVVH and CVVHDF, prescription
of higher blood flow and effluent doses, and advances in
hemofilter technology since the publication of the above-
referenced studies, it is reasonable to suspect amino acid
losses in CRRT are being underestimated. Generous protein
provision of 2.5 g/kg with PN or via addition of protein
modules to high-protein tube feeding is warranted.
Fat
Alterations to IV lipid emulsion for CRRT patients are not
recommended beyond that which would be in response to
hypertriglyceridemia, a common complication in the ICU,
or for administration of non nutrition lipid emulsion with
clevidipine (2 kcal/mL) or propofol (1.1 kcal/mL).
Volume and Sodium Considerations
CRRT is a highly efficient means of fluid removal. Ac-
cordingly, it is not necessary to provide a concentrated
tube-feeding formula, which may limit optimal protein
provision. It is, however, our practice in the ICU to provide a
763
concentrated, minimum-sodium PN formula to allow fluids
to be managed primarily outside of PN. Significant sources
of sodium in the ICU include crystalloid administered with
medications and for resuscitation and as needed well beyond
the daily required sodium intake, which can exacerbate
hypervolemic states. Notably, dialysate and replacement
solutions contain sodium of 130–140 mEq/L. Hypona-
tremia that accompanies AKI is predominantly due to a
state of hypervolemia rather than a sodium deficit in the
absence of conditions of sodium loss, as with significant
gastrointestinal output. While significant fistula, ostomy, or
wound losses of sodium warrant sodium replacement, this
is to be considered separately from the hyponatremia due
to hypervolemia that is common in AKI. It is, therefore,
essential that volume status be assessed carefully whenever
there is hyponatremia. In hypervolemia, total sodium body-
sodium stores may be increased along with volume (edema),
thus requiring restriction of both fluid (assuming the ab-
sence of a fluid-responsive shock state) and sodium intake.23
Electrolytes and Minerals
Defining electrolyte needs is comparatively more straight-
forward than determining energy and protein needs. Crit-
ically ill patients on CRRT are closely monitored with
frequent laboratory tests to guide the CRRT prescrip-
tion, including dialysate and replacement-fluid electrolyte
composition; protocol-based or provider-specific orders for
potassium, magnesium, and phosphate supplementation;
and separate calcium infusion to replace calcium bound
to and lost with citrate when RCA is employed. Nutrition
support plays mostly a supporting role with regard to the
management of electrolytes, acknowledging the partnership
with primary ICU and nephrology teams when modifica-
tions to PN are needed.
Hypophosphatemia
Hypophosphatemia is an important complication of CRRT
and may have a negative impact on patient outcomes.46
CRRT continuously clears phosphate from the blood, while
replacement and dialysate fluids have historically excluded
phosphate. Close monitoring of phosphate levels is essen-
tial for the duration of CRRT. A variety of approaches
to prevent and treat hypophosphatemia that occurs with
CRRT are employed, including the more recent use of
commercial phosphate-containing CRRT solutions, Phox-
illum (Gambro Renal Products, Daytona Beach, FL) and
multiPlus (Fresenius Kabi, Bad Homberg, Germany), which
can reduce the need for supplemental phosphate. When
phosphate-free fluids are used, additional supplementation
is necessary, administered either by mouth or per tube
or via the IV route. Sodium phosphate may be infused
intermittently, continuously, or added to the PN. Selection
of route of phosphate supplementation should consider the
764
risk of diarrhea with enteral administration and availability
of parenteral products, which have recently been subject
to national shortages. Low-phosphate, renal-formula tube
feedings are not recommended for patients receiving CRRT.
Calcium
RCA works locally within the CRRT circuit to inactivate
the coagulation cascade and platelets by chelating calcium.
Since a portion of the citrate–calcium complex remains after
extracorporeal elimination, calcium must be infused on the
venous (return) side of the circuit prior to reentering the
patient’s circulation to achieve a normal systemic calcium
concentration.47 The amount of parenteral calcium neces-
sary to do this is exponentially higher than standard doses
that would be provided with PN. As such, a stable and
unchanged amount of calcium may be provided in the PN,
if any.
Potassium and Magnesium
Potassium and magnesium are provided in replacement and
dialysate fluids. Additional potassium and magnesium may
be prescribed as part of protocol orders or as needed. Rec-
ognizing wide variation among institutions, it is our practice
to zero the electrolytes in PN for patients receiving CRRT
and incrementally add back potassium or magnesium based
on recurrent and regular need for potassium or magnesium
outside of the RRT circuit, in close communication with
our nephrology colleagues. As with phosphorus, the risk of
diarrhea with enteral administration of potassium should
be weighed. Because enteral magnesium is poorly absorbed
and is dose-limited by diarrhea, IV replacement of magne-
sium sulfate to treat hypomagnesemia is favored.
In addition to understanding the CRRT modality used,
convection-based, diffusion-based, or both, nutrition sup-
port clinicians should be familiar with the electrolyte con-
centrations in replacement and dialysate fluids routinely
prescribed at their institutions. See Table 2 for ranges
of electrolyte concentrations used in available products.
The reader is also referred to product-specific labeling for
electrolyte content of commercial replacement and dialysate
fluids.
Vitamins and Trace Elements
Optimal dosing of vitamins and trace minerals in critical
illness is unknown.48 RRT adds a level of complexity,
potentially increasing the risk for deficiencies via filter
adsorption or effluent losses, and additional needs beyond
amounts administered with nutrition programs are gener-
ally not well understood.49,50 A review of the literature
suggests losses of selenium with CRRT to a degree that
may warrant additional selenium beyond that provided
by standard multiple trace element products in PN.39,51-54
Recognizing blood assay limitations, additional monitoring
Nutrition in Clinical Practice 33(6)
in patients on prolonged courses of CRRT is warranted.
The reader is referred to the excellent reviews cited in this
section for further information on vitamin and trace element
disposition in CRRT.
Conclusion
The patient receiving CRRT presents a unique nutrition
challenge for clinicians in the ICU. Current evidence and
practical experience suggests CRRT can be a significant
source of calories, in particular with the use of citrate-based
anticoagulation. Clinicians should observe patients for signs
of overfeeding, such as hyperglycemia, high insulin re-
quirements, and hypertriglyceridemia. Glucose losses from
blood into effluent are buffered by the use of dextrose-
containing replacement fluids (CVVH and CVVHDF) and
dialysate fluids (CVVHD and CVVHDF), but the clinician
should be aware that these losses can be significant with
the use of newer, dextrose-free dialysate and replacement
fluids. Amino acid losses are significant with CRRT, and
restriction of amino acids should be avoided in an attempt
to limit the need for RRT. Special considerations exist for
volume and electrolytes, the crux of which is managed
within the CRRT circuit and established institution-specific
protocols. We urge working collaboratively with nephrology
on fluid and electrolyte provision in the nutrition program,
recognizing the ability to tailor PN more readily than EN.
Transitions off CRRT should be anticipated with improved
hemodynamics, as discontinuing CRRT may warrant ad-
justments to calorie, protein, and electrolyte provision. We
recommend that nutrition support clinicians practicing in
the ICU work to understand the specific CRRT modalities
used in their practice, the specific components of the CRRT
prescription, and patient-specific variables that may affect
macronutrient and micronutrient disposition.
Statement of Authorship
E. M. Nystrom and A. M. Nei equally contributed to the
conception and design of the manuscript; E. M. Nystrom and
A. M. Nei equally contributed to the acquisition and analysis
of the data; E. M. Nystrom and A. M. Nei equally contributed
to the interpretation of the data; and E. M. Nystrom and A.
M. Nei drafted the manuscript. Both authors critically revised
the manuscript, agree to be fully accountable for ensuring the
integrity and accuracy of the work, and read and approved the
final manuscript.
Supplementary Information
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
References
1. Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute
kidney injury are associated with hospital mortality in critically ill
patients: a cohort analysis. Crit Care. 2006;10(3):R73.
Nystrom and Nei
2. Herrera-Gutierrez ME, Seller-Perez G, Sanchez-Izquierdo-Riera JA,
Maynar-Moliner J. Prevalence of acute kidney injury in intensive
care units: the “COrte de prevalencia de disFuncion RenAl y DEpu-
racion en criticos” point-prevalence multicenter study. J Crit Care.
2013;28(5):687-694.
3. Koeze J, Keus F, Dieperink W, van der Horst IC, Zijlstra JG, van
Meurs M. Incidence, timing and outcome of AKI in critically ill
patients varies with the definition used and the addition of urine output
criteria. BMC Nephrol. 2017;18(1):70.
4. Iwagami M, Yasunaga H, Noiri E, et al. Current state of continuous
renal replacement therapy for acute kidney injury in Japanese intensive
care units in 2011: analysis of a national administrative database.
Nephrol Dial Transplant. 2015;30(6):988-995.
5. Nash K, Hafeez A, Hou S. Hospital-acquired renal insufficiency. Am
J Kidney Dis. 2002;39(5):930-936.
6. Park S, Lee S, Jo HA, et al. Epidemiology of continuous renal replace-
ment therapy in Korea: results from the National Health Insurance
Service claims database from 2005 to 2016. Kidney Res Clin Pract.
2018;37(2):119-129.
7. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney
Injury Work Group. KDIGO clinical practice guideline for acute
kidney injury. Kidney Int. 2012;2(suppl):1-138.
8. Fiaccadori E, Lombardi M, Leonardi S, Rotelli CF, Tortorella G,
Borghetti A. Prevalence and clinical outcome associated with preex-
isting malnutrition in acute renal failure: a prospective cohort study.
J Am Soc Nephrol. 1999;10(3):581-593.
9. Weissman C. Nutrition in the intensive care unit. Crit Care.
1999;3(5):R67-R75.
10. Ilias I, Vassiliadi DA, Theodorakopoulou M, et al. Adipose tissue
lipolysis and circulating lipids in acute and subacute critical illness:
effects of shock and treatment. J Crit Care. 2014;29(6):1130.e1135-
e1139.
11. Carlson GL. Insulin resistance and glucose-induced thermogenesis in
critical illness. Proc Nutr Soc. 2001;60(3):381-388.
12. Patkova A, Joskova V, Havel E, et al. Energy, protein, carbohydrate,
and lipid intakes and their effects on morbidity and mortality in
critically ill adult patients: a systematic review. Adv Nutr. 2017;8(4):624-
634.
13. Ronco C. Continuous renal replacement therapy: forty-year anniver-
sary. Int J Artif Organs. 2017:40(6):257-264.
14. Murugan R, Hoste E, Mehta RL, et al. Precision fluid management
in continuous renal replacement therapy. Blood Purif. 2016;42(3):266-
278.
15. Barenbrock M, Hausberg M, Matzkies F, de la Motte S, Schaefer RM.
Effects of bicarbonate- and lactate-buffered replacement fluids on car-
diovascular outcome in CVVH patients. Kidney Int. 2000;58(4):1751-
1757.
16. Tian JH, Ma B, Yang K, Liu Y, Tan J, Liu TX. Bicarbonate- versus
lactate-buffered solutions for acute continuous haemodiafiltration or
haemofiltration. Cochrane Database Syst Rev. 2015(3):CD006819.
17. Uchino S, Bellomo R, Morimatsu H, et al. Continuous renal replace-
ment therapy: a worldwide practice survey. The beginning and ending
supportive therapy for the kidney (B.E.S.T. kidney) investigators.
Intensive Care Med. 2007;33(9):1563-1570.
18. Bai M, Zhou M, He L, et al. Citrate versus heparin anticoagulation
for continuous renal replacement therapy: an updated meta-analysis of
RCTs. Intensive Care Med. 2015;41(12):2098-2110.
19. Meier-Kriesche HU, Gitomer J, Finkel K, DuBose T. Increased total to
ionized calcium ratio during continuous venovenous hemodialysis with
regional citrate anticoagulation. Crit Care Med. 2001;29(4):748-752.
20. Bellomo R, Cass A, Cole L, et al. Calorie intake and patient outcomes
in severe acute kidney injury: findings from the Randomized Evaluation
of Normal vs. Augmented Level of Replacement Therapy (RENAL)
study trial. Crit Care. 2014;18(2):R45.
765
21. Wolfe RR. The 2017 Sir David P Cuthbertson lecture. Amino acids and
muscle protein metabolism in critical care. Clin Nutr. 2018;37(4):1093-
1100.
22. Butler SO, Btaiche IF, Alaniz C. Relationship between hyperglycemia
and infection in critically ill patients. Pharmacotherapy. 2005;25(7):963-
976.
23. Mundi MS, Nystrom EM, Hurley DL, McMahon MM. Management
of parenteral nutrition in hospitalized adult patients. JPEN J Parenter
Enteral Nutr. 2017;41(4):535-549.
24. Oudemans-van Straaten HM, Kellum JA, Bellomo R. Clinical review:
anticoagulation for continuous renal replacement therapy—heparin or
citrate? Crit Care. 2011;15(1):202.
25. Lanckohr C, Hahnenkamp K, Boschin M. Continuous renal replace-
ment therapy with regional citrate anticoagulation: do we really know
the details? Curr Opin Anaesthesiol. 2013;26(4):428-437.
26. Marino PL. The ICU Book. Baltimore, MD: Williams & Wilkins; 2014.
27. Balik M, Zakharchenko M, Otahal M, et al. Quantification of sys-
temic delivery of substrates for intermediate metabolism during citrate
anticoagulation of continuous renal replacement therapy. Blood Purif.
2012;33(1-3):80-87.
28. Balik M, Zakharchenko M, Leden P, et al. Bioenergetic gain of
citrate anticoagulated continuous hemodiafiltration—a comparison
between 2 citrate modalities and unfractionated heparin. J Crit Care.
2013;28(1):87-95.
29. Stevenson JM, Heung M, Vilay AM, Eyler RF, Patel C, Mueller BA.
In vitro glucose kinetics during continuous renal replacement therapy:
implications for caloric balance in critically ill patients. Int J Artif
Organs. 2013;36(12):861-868.
30. New AM, Nystrom EM, Frazee E, Dillon JJ, Kashani KB, Miles JM.
Continuous renal replacement therapy: a potential source of calories in
the critically ill. Am J Clin Nutr. 2017;105(6):1559-1563.
31. Fiaccadori E, Regolisti G, Maggiore U. Specialized nutritional support
interventions in critically ill patients on renal replacement therapy. Curr
Opin Clin Nutr Metab Care. 2013;16(2):217-224.
32. Davenport A, Roberts NB. Amino acid losses during continuous
high-flux hemofiltration in the critically ill patient. Crit Care Med.
1989;17(10):1010-1014.
33. Davies SP, Reaveley DA, Brown EA, Kox WJ. Amino acid clearances
and daily losses in patients with acute renal failure treated by con-
tinuous arteriovenous hemodialysis. Crit Care Med. 1991;19(12):1510-
1515.
34. Frankenfield DC, Badellino MM, Reynolds HN, Wiles CE 3rd,
Siegel JH, Goodarzi S. Amino acid loss and plasma concentration
during continuous hemodiafiltration. JPEN J Parenter Enteral Nutr.
1993;17(6):551-561.
35. Maxvold NJ, Smoyer WE, Custer JR, Bunchman TE. Amino acid
loss and nitrogen balance in critically ill children with acute renal
failure: a prospective comparison between classic hemofiltration and
hemofiltration with dialysis. Crit Care Med. 2000;28(4):1161-1165.
36. Bellomo R, Seacombe J, Daskalakis M, et al. A prospective com-
parative study of moderate versus high protein intake for criti-
cally ill patients with acute renal failure. Renal Failure. 1997;19(1):
111-120.
37. Bellomo R, Tan HK, Bhonagiri S, et al. High protein intake during
continuous hemodiafiltration: impact on amino acids and nitrogen
balance. Int J Artif Organs. 2002;25(4):261-268.
38. Scheinkestel CD, Adams F, Mahony L, et al. Impact of increasing
parenteral protein loads on amino acid levels and balance in critically
ill anuric patients on continuous renal replacement therapy. Nutrition.
2003;19(9):733-740.
39. Zappitelli M, Juarez M, Castillo L, Coss-Bu J, Goldstein SL. Con-
tinuous renal replacement therapy amino acid, trace metal and folate
clearance in critically ill children. Intensive Care Med. 2009;35(4):698-
706.
766
40. Btaiche IF, Mohammad RA, Alaniz C, Mueller BA. Amino acid
requirements in critically ill patients with acute kidney injury
treated with continuous renal replacement therapy. Pharmacotherapy.
2008;28(5):600-613.
41. Brown RO, Compher C. A.S.P.E.N. clinical guidelines: nutrition sup-
port in adult acute and chronic renal failure. JPEN J Parenter Enteral
Nutr. 2010;34(4):366-377.
42. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the
provision and assessment of nutrition support therapy in the adult
critically ill patient: Society of Critical Care Medicine (SCCM) and
American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.).
JPEN J Parenter Enteral Nutr. 2016;40(2):159-211.
43. Cano NJ, Aparicio M, Brunori G, et al. ESPEN guidelines on par-
enteral nutrition: adult renal failure. Clin Nutr. 2009;28(4):401-414.
44. Scheinkestel CD, Kar L, Marshall K, et al. Prospective randomized
trial to assess caloric and protein needs of critically Ill, anuric,
ventilated patients requiring continuous renal replacement therapy.
Nutrition. 2003;19(11-12):909-916.
45. Hurt RT, McClave SA, Martindale RG, et al. Summary points and
consensus recommendations from the International Protein Summit.
Nutr Clin Pract. 2017;32(1 suppl):142S-151S.
46. Heung M, Mueller BA. Prevention of hypophosphatemia during
continuous renal replacement therapy—an overlooked problem. Semin
Dial. 2018;31(3):213-218.
Nutrition in Clinical Practice 33(6)
47. Morabito S, Pistolesi V, Tritapepe L, Fiaccadori E. Regional citrate
anticoagulation for RRTs in critically ill patients with AKI. Clin J Am
Soc Nephrol. 2014;9(12):2173-2188.
48. Vanek VW, Borum P, Buchman A, et al. A.S.P.E.N. position paper:
recommendations for changes in commercially available parenteral
multivitamin and multi-trace element products. Nutr Clin Pract.
2012;27(4):440-491.
49. Oh WC, Gardner DS, Devonald MA. Micronutrient and amino acid
losses in acute renal replacement therapy. Curr Opin Clin Nutr Metab
Care. 2015;18(6):593-598.
50. Tucker BM, Safadi S, Friedman AN. Is routine multivitamin sup-
plementation necessary in US chronic adult hemodialysis patients? A
systematic review. J Renal Nutr. 2015;25(3):257-264.
51. Berger MM, Shenkin A, Revelly JP, et al. Copper, selenium, zinc, and
thiamine balances during continuous venovenous hemodiafiltration in
critically ill patients. Am J Clin Nutr. 2004;80(2):410-416.
52. Ben-Hamouda N, Charriere M, Voirol P, Berger MM. Massive copper
and selenium losses cause life-threatening deficiencies during prolonged
continuous renal replacement. Nutrition. 2017;34:71-75.
53. Gervasio JM, Garmon WP, Holowatyj M. Nutrition support in acute
kidney injury. Nutr Clin Pract. 2011;26(4):374-381.
54. Wooley JA, Btaiche IF, Good KL. Metabolic and nutritional aspects
of actue renal failure in critically ill patients requiring continuous renal
replacement therapy. Nutr Clin Pract. 2005;20(2):176-191.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Use of Intradialytic Parenteral Nutrition in Patients December 2018 767–771

C 2018 American Society for

Undergoing Hemodialysis Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10190
wileyonlinelibrary.com

Menaka Sarav, MD1 ; and Allon N. Friedman, MD2

Abstract
Intradialytic parenteral nutrition (IDPN) is a form of supplemental nutrition used to treat patients with malnutrition who receive
hemodialysis. Once the diagnosis of malnutrition is made in such patients, encouragement of oral intake is the first-line treatment. If
this fails, then enteral or parenteral nutrition may be needed. This review examines the literature on the use of IDPN and summarizes
the current recommendations. There is considerable controversy over indications and benefits of IDPN, and well-controlled, long-
term studies are needed to help tease out these issues. In the interim, clinical judgment should be used when considering IDPN for
individual patients. (Nutr Clin Pract. 2018;33:767–771)

Keywords
chronic kidney failure; hemodialysis; intradialytic parenteral nutrition; malnutrition; parenteral nutrition

Background Technical Aspects

The presence of protein energy malnutrition (PEM) can
be quite common1-3 among the hundreds of thousands of
patients in the United States who receive hemodialysis, and
it independently predicts higher hospitalization rates, lower
quality of life, and death.4 Causes of PEM in hemodialysis
patients include reduced food intake from uremia, concur-
rent illnesses, alterations in the balance between hunger and
satiety hormones, dietary restrictions, financial limitations,
depression/anxiety, and nutrient and energy losses that
occur during the hemodialysis session.3-5
Prevention of PEM requires a multidisciplinary effort
involving the entire healthcare team. Essential steps to im-
proving nutrition status include liberalizing any dietary re-
strictions, encouraging food intake, providing oral nutrition
supplements, and offering nutrition education. However,
these interventions are not always effective in helping the
patient achieve his or her nutrition goals. In such cases,
intradialytic parenteral nutrition (IDPN), which involves
an intravenous (IV) infusion of essential nutrients during
hemodialysis treatments, may be helpful as a component of
nutrition support.
This discussion will focus on the use of IDPN for patients
who are receiving in-center maintenance hemodialysis, who
comprise the great majority of dialysis patients in the
United States, and will review technical aspects and admin-
istration, the appropriate selection of patients, risks and
benefits, and evidence supporting its use. The use of IDPN
in patients receiving home and peritoneal dialysis will not
be covered in this report.
There are 2 main types of IDPN. In compounded
admixture-based IDPN, all-in-one IDPN bags are mixed by
a pharmacy based on individual patient needs. In commer-
cial admixture-based IDPN, premixed bags are provided for
generic use. Of the 2, commercial premixed bags are far
more popular because the process of compounding single
bags based on specific patient needs is time consuming and
adds cost.5 IDPN is administrated via IV infusion with
an infusion pump during hemodialysis. Typically the most
concentrated IDPN formula is used to reduce the risk for
volume overload and fit the treatment within the usual time
constraints of a standard hemodialysis session.
A typical IDPN infusion provides 800–1200 kcal in the
form of glucose, lipids, and amino acids (usually 30–60 g
of the latter).6 However, some amino acids can be lost
in the dialysate depending on the hemodialysis filter used
From the 1 Division of Nephrology and Hypertension, Department of
Medicine, NorthShore University HealthSystem, Evanston, Illinois,
USA; and the 2 Department of Medicine, Indiana University School
of Medicine, Indianapolis, Indiana, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on September 12, 2018.
Corresponding Author:
Menaka Sarav, MD, Division of Nephrology and Hypertension,
Department of Medicine, 2650 Ridge Avenue, Evanston,
IL 60201, USA.
Email: msarav@northshore.org
768 Nutrition in Clinical Practice 33(6)

Table 1. National Kidney Foundation–Recommended Measures for Monitoring Nutrition Status in Patients Receiving
Maintenance Dialysis.

Frequency of
Parameters Measurements Comments Limitations

Predialysis serum albumin level
Percentage change in
postdialysis body weight
Subjective global assessment
Dietary interview and/or dairy
nPNA
Monthly Target level is normal >4 g/dL
Monthly Trends from edema-free body
weight need to be monitored
Every 6 months A simple technique based on
subjective and objective aspects
of the medical history and
physical examination
Every 6 months 3-Day dietary records followed by
interviews and calculating
nutrient intake by an
experienced registered renal
dietitian
Monthly In the clinically stable patient,
nPNA can be used to estimate
protein intake
Affected by nonnutrition factors
such as inflammation,
dehydration, infection,
acid-base status
Getting a true dry weight may be
challenging
Does not include visceral protein
levels in the assessment; it is
subjective and has not been
studied well in hemodialysis
populations
Time-consuming and limited by
accuracy of documentation by
patient and/or interview and
calculation by dietitian
nPNA is not valid in catabolic or
anabolic state; nPNA may
fluctuate from day to day as a
function of protein intake, and a
single nPNA measurement may
not reflect usual protein intakes

nPNA, normalized protein nitrogen appearance.

and if they have been reused.7 Electrolyte-free admixtures
(ie, without sodium, potassium, and phosphorus) are also
available.5 When taking into account amino acids and en-
ergy that can be lost with each hemodialysis session (ie, 200–
480 kcal of energy and 10–12 g of amino acids),8 not >3000
kcal of energy and 150 g of amino acids will be provided by
IDPN with thrice weekly dialysis. Thus, in a 70-kg patient,
approximately 6 kcal/kg/d (target being 30–35 kcal/kg/d)
of energy9,10 and 0.30 g/kg/d (target being 1.2 g/kg/d) of
amino acids9,10 are provided by IDPN, highlighting the fact
that IDPN should be used only as supplemental nutrition in
patients with a spontaneous oral nutrition intake of at least
25 kcal/kg/d of energy and 0.9 g/kg/d of amino acids.5,9,11,12
Evidence Supporting the Use of IDPN
Several retrospective and prospective randomized studies
have examined the effects of IDPN on clinical end points.
Benefits noted included improvements in anthropometric
parameters and serum albumin and serum prealbumin albu-
min levels.12-15 One human study that infused stable isotopes
to assess muscle turnover during IDPN identified increases
in regional and total body protein synthesis.16 However, in
the largest randomized study of IDPN (n = 186), the addi-
tion of IDPN to oral nutrition supplementation over 1 year
did not improve 2-year mortality rates, hospitalization rates,
or general well-being as compared with oral supplements
alone.17 In addition, all of the randomized studies have
been open label in design, which increases the opportunity
for bias. Moreover, even though most studies used serum
albumin level as an inclusion criterion or surrogate outcome
representing nutrition improvement, serum albumin level
is primarily an indicator of inflammation or illness, thus
making interpretation of these studies more challenging.18
Clearly, longer-term prospective clinical trials with appro-
priate control groups are needed to identify which type of
patient is most likely to benefit from IDPN.12
Selection of Patients for IDPN
The process of selecting patients for IDPN first requires
identifying patients who are malnourished. Table 1 includes
a list of parameters from the National Kidney Foundation
Disease Outcome Quality Initiative (NKF KDOQI) that
can be used to identify malnutrition.11 As Table 1 notes,
each of the parameters have limitations, sometimes serious
ones, when applied to the hemodialysis population. There-
fore, identifying PEM in hemodialysis patients requires a
large dose of clinical judgment based on a careful history
and physical examination.
Once PEM is identified, IDPN should not be the initial
treatment choice. Rather, encouraging increased oral nu-
trition or supplements like protein drinks or bars through
dietary counseling and education is an important first step.
Sarav and Friedman 769

Table 2. Society Recommendations on Indications for IDPN.

Kidney Disease Outcomes Quality Initiative,
2000
European Society of Parenteral and Enteral
Nutrition, 2009
American Society of Parenteral and Enteral
Nutrition, 2010
1. Evidence of protein or energy malnutrition and inadequate dietary
protein and/or energy intake
2. Inability to administer or tolerate adequate oral nutrition, including food
supplements or tube feeding
3. Oral or enteral intake that, when combined with IDPN, meets the
recommended dietary intake
1. IDPN improves nutrition status in undernourished patients who are
receiving hemodialysis
2. In outpatients, if nutrition counseling and oral nutrition supplements are
unsuccessful, IDPN should be proposed
IDPN should not be used as a nutrition supplement in malnourished
chronic kidney disease stage 5 hemodialysis patients

IDPN, intradialytic parenteral nutrition.

Every effort should also be made to ease dietary restrictions
and financial constraints that may work to limit food intake.
Pharmacological agents such as megestrol and ghrelin may
stimulate appetite, and increase serum albumin level and
weight; however, these drugs have not been studied sys-
tematically, and large-scale prospective studies are needed
in patients who are receiving hemodialysis.19,20 Moreover,
the physician should also search for identifiable causes for
PEM. A complete gastrointestinal workup is needed to
make sure there are no physical or structural impediments
(eg, poor dentation, dysphagia, gastroparesis) to oral intake.
Other causes of anorexia such as depression, dementia, or
inadequate dialysis should also be excluded.
If despite the efforts described earlier the patient is
unable to meet the nutrition intake goals, then the option
of IDPN should be discussed.9 Notably, in patients who are
not meeting daily nutrition goals even with oral nutrition
supplements and IDPN, placement of a feeding tube for
enteral nutrition (EN) or daily parenteral nutrition (PN)
may be the appropriate next step. Another subgroup of
hemodialysis patients in whom EN or PN may be beneficial
are critically ill individuals. These persons should be treated
as intensive care patients.9
Several nephrology and nutrition societies have com-
mented on the management of PEM, including the use
of intensive nutrition counseling, dietary oral supplements,
and IDPN as shown in Table 2 and Figure 1. NKF
KDOQI guidelines recommend using IDPN in malnour-
ished patients who are receiving dialysis who are not able
to increase oral ingestion with dietary counseling, do not
tolerate oral supplementation or enteral tube feeding, and
have a sufficient spontaneous intake that combined with
the nutrients provided by IDPN can reach the nutrition
targets.11 European Society of Parenteral and Enteral Nu-
trition (ESPEN) guidelines suggest using IDPN in mal-
nourished dialysis patients who had unsuccessful responses
to nutrition counseling and oral nutrition supplements.9
Interestingly, NKF KDOQI recommends IDPN only after
EN fails, whereas ESPEN recommends IDPN as first-line
therapy when placement of a feeding tube is deemed to be
higher risk than IDPN. In contrast, the American Society
of Parenteral and Enteral Nutrition does not recommend
IDPN because of lack of adequate supporting data.10 The
differences in recommendations between societies is likely
due to a lack of high-grade clinical evidence, thus making
any judgment about the benefits and risks primarily opinion
based. These conflicting recommendations highlight the
need for clinical judgment in determining which patients
may benefit from IDPN.
When should IDPN be discontinued once initiated?
Stopping is reasonable if there is sustained improvement
in nutrition parameters, if adverse effects of IDPN are
observed (see the following section), or if there is sponta-
neous improvement of energy/protein intake, thus making
IDPN unnecessary. In contrast, lack of improvement after
3–6 months of IDPN should also lead to discontinuation
of IDPN.5 Ultimately the decision to discontinue IDPN,
like the decision to initiate it, is based primarily on clinical
judgment.
Risks and Advantages of IDPN
There are several benefits and drawbacks of IDPN, all of
which are listed in Table 3. Of particular interest in the
known drawbacks are associated metabolic and electrolyte
derangements. For example, IDPN is contraindicated if
triglyceride levels are >300 mg/dL because lipids present
in the IDPN could exacerbate hypertriglyceridemia. Hyper-
glycemia has also been identified as a possible complication
of IDPN. Because of the risk for electrolyte disturbances,
stringent monitoring of the electrolytes, especially during
the first weeks of IDPN support, is highly recommended.9
Depending on the compounding of the nutrient solution
blood glucose, triglycerides and liver function tests need to
770
Figure 1. Algorithm for intradialytic parenteral nutrition use in patients who are receiving hemodialysis. ASPEN, American
Society of Parenteral and Enteral Nutrition; ESPEN, European Society of Parenteral and Enteral Nutrition; KDOQI, Kidney
Disease Outcomes Quality Initiative.
Table 3. Advantages and Disadvantages of IDPN.
Advantages 1. No need for a dedicated enteral feeding
tube or parenteral nutrition vascular
access
2. Ultrafiltration during dialysis will help
minimize the risks for fluid overload
3. No demands on patient time or effort
Disadvantages
1. Provides insufficient energy and protein
to support long-term daily needs
2. Does not address the problem of
improving patient’s eating behavior
3. Side effects such as metabolic and
electrolyte abnormalities
4. Medicare reimbursement for IDPN is
complex and extremely time-consuming
IDPN, intradialytic parenteral nutrition.
be monitored on a routine basis.21 Overall, IDPN has been
demonstrated in hemodialysis patients to be safe with a low
complication rate. In the largest randomized study of IDPN
ever performed, the most common adverse effects identified
were digestive symptoms, hypotension, and muscle cramps
that occurred during treatment, although they occurred at
Nutrition in Clinical Practice 33(6)
a similar rate to the control group (which was consuming
oral supplementation).17 One potential nonmedical compli-
cation of IDPN involves its financial impact because cost
varies based on insurance coverage. Previously, Medicare
Part B required proof of gastrointestinal impairment before
coverage. However, because Medicare Part D has covered
IDPN under “drug therapies” since 2006, it is now easier to
obtain coverage once medical necessity is demonstrated.22
Conclusion
Patients requiring hemodialysis are at risk for PEM, which is
linked to a greater likelihood of adverse clinical outcomes.4
IDPN is a relatively safe and efficacious option for
delivering nutrients to hemodialysis patients with reduced
spontaneous oral intake. However, it is not designed to be
the only or even the major source of nutrition sustenance. It
is also not indicated for severely malnourished or critically
ill patients who are receiving hemodialysis. No studies have
demonstrated that IDPN improves major clinical outcomes,
although few randomized trials in this field exist.23 There-
fore, additional studies are needed to establish the benefits
of IDPN. In the meantime, clinical judgment and acumen
Sarav and Friedman
remain key to diagnosing PEM and determining the need for
IDPN.
Statement of Authorship
M. Sarav and A. Friedman contributed to the conception/
design of the manuscript; M. Sarav and A. Friedman
contributed to acquisition, analysis, or interpretation of the
data; M. Sarav and A. Friedman drafted the manuscript;
all authors critically revised the manuscript and agree to
be fully accountable for ensuring the integrity and accuracy
of the work; and all authors read and approved the final
manuscript.
References
1. Kopple JD, McCollum Award Lecture, 1996: protein-energy malnutri-
tion in maintenance dialysis patients. Am J Clin Nutr. 1997;65(5):1544-
1557.
2. Kalantar-Zadeh K, Block G, McAllister CJ, Humphreys MH, Kopple
JD. Appetite and inflammation, nutrition, anemia, and clinical out-
come in hemodialysis patients. Am J Clin Nutr. 2004;80(2):299-307.
3. Gracia-Iguacel C, Gonzalez-Parra E, Perez-Gomez MV, et al. Preva-
lence of protein-energy wasting syndrome and its association with
mortality in haemodialysis patients in a centre in Spain. Nefrologia.
2013;33(4):495-505.
4. Kalantar-Zadeh K, Rhee C, Sim JJ, Stenvinkel P, Anker SD, Kovesdy
CP. Why cachexia kills: examining the causality of poor outcomes in
wasting conditions. J Cachexia Sarcopenia Muscle. 2013;4(2):89-94.
5. Sabatino A, Regolisti G, Antonucci E, Cabassi A, Morabito S, Fiac-
cadori E. Intradialytic parenteral nutrition in end-stage renal disease:
practical aspects, indications and limits. J Nephrol. 2014;27(4):377-383.
6. Cano N. Nutritional supplementation in adult patients on hemodialy-
sis. J Ren Nutr. 2007;17(1):103-105.
7. Ikizler TA, Flakoll PJ, Parker RA, Hakim RM. Amino acid and
albumin losses during hemodialysis. Kidney Int. 1994;46(3):830-837.
8. Liu Y, Xiao X, Qin DP, et al. Comparison of intradialytic parenteral
nutrition with glucose or amino acid mixtures in maintenance hemo-
dialysis patients. Nutrients. 2016;8(6):e220.
9. Cano NJ, Aparicio M, Brunori G, et al. ESPEN Guidelines on
Parenteral Nutrition: adult renal failure. Clin Nutr. 2009;28(4):401-414.
10. Brown RO, Compher C. A.S.P.E.N. clinical guidelines: nutrition sup-
port in adult acute and chronic renal failure. JPEN J Parenter Enteral
Nutr. 2010;34(4):366-377.
771
11. Clinical practice guidelines for nutrition in chronic renal failure.
K/DOQI, National Kidney Foundation. American journal of kidney
diseases: the official journal of the National Kidney Foundation 2000;
35(6 suppl 2):S1-S140.
12. Dukkipati R, Kalantar-Zadeh K, Kopple JD. Is there a role for intradi-
alytic parenteral nutrition? A review of the evidence. Am J Kidney Dis.
2010;55(2):352-364.
13. Cano N, Labastie-Coeyrehourq J, Lacombe P, et al. Perdialytic
parenteral nutrition with lipids and amino acids in malnourished
hemodialysis patients. Am J Clin Nutr. 1990;52(4):726-730.
14. Marsen TA, Beer J, Mann H. Intradialytic parenteral nutrition in main-
tenance hemodialysis patients suffering from protein-energy wasting.
Results of a multicenter, open, prospective, randomized trial. Clin Nutr.
2017;36(1):107-117.
15. Chertow GM, Ling J, Lew NL, Lazarus JM, Lowrie EG. The
association of intradialytic parenteral nutrition administration with
survival in hemodialysis patients. Am J Kidney Dis. 1994;24(6):
912-920.
16. Pupim LB, Flakoll PJ, Ikizler TA. Nutritional supplementation acutely
increases albumin fractional synthetic rate in chronic hemodialysis
patients. J Am Soc Nephrol. 2004;15(7):1920-1926.
17. Cano NJ, Fouque D, Roth H, et al. Intradialytic parenteral nutrition
does not improve survival in malnourished hemodialysis patients: a 2-
year multicenter, prospective, randomized study. J Am Soc Nephrol.
2007;18(9):2583-2591.
18. Friedman AN, Fadem SZ. Reassessment of albumin as a nutritional
marker in kidney disease. J Am Soc Nephrol. 2010;21(2):223-230.
19. Yeh SS, Marandi M, Thode HC Jr, et al. Report of a pilot, double-
blind, placebo-controlled study of megestrol acetate in elderly dialysis
patients with cachexia. J Ren Nutr. 2010;20(1):52-62.
20. Wynne K, Giannitsopoulou K, Small CJ, et al. Subcutaneous ghrelin
enhances acute food intake in malnourished patients who receive
maintenance peritoneal dialysis: a randomized, placebo-controlled
trial. J Am Soc Nephrol. 2005;16(7):2111-2118.
21. Druml W, Kierdorf HP; Working group for developing the guidelines
for parenteral nutrition of The German Association for Nutritional
Medicine. Parenteral nutrition in patients with renal failure – Guide-
lines on Parenteral Nutrition, Chapter 17. Ger Med Sci. 2009;7:
Doc11.
22. Fuhrman MP. Intradialytic parenteral nutrition and intraperitoneal
nutrition. Nutr Clin Pract. 2009;24(4):470-480.
23. Pupim LB, Flakoll PJ, Brouillette JR, Levenhagen DK, Hakim RM,
Ikizler TA. Intradialytic parenteral nutrition improves protein and en-
ergy homeostasis in chronic hemodialysis patients. J Clin Invest. 2002;
110(4):483-492.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Exploring the Potential Effectiveness of Combining Optimal December 2018 772–789

C 2018 American Society for

Nutrition With Electrical Stimulation to Maintain Muscle Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10213
Health in Critical Illness: A Narrative Review wileyonlinelibrary.com

Selina M. Parry, PT, PhD1,∗ ; Lee-anne S. Chapple, PhD2,∗ ;

3,∗
and Marina Mourtzakis, PhD

Abstract
Muscle wasting occurs rapidly within days of an admission to the intensive care unit (ICU). Concomitant muscle weakness and
impaired physical functioning can ensue, with lasting effects well after hospital discharge. Early physical rehabilitation is a promising
intervention to minimize muscle weakness and physical dysfunction. However, there is an often a delay in commencing active
functional exercises (such as sitting on the edge of bed, standing and mobilizing) due to sedation, patient alertness, and impaired
ability to cooperate in the initial days of ICU admission. Therefore, there is high interest in being able to intervene early through
nonvolitional exercise strategies such as electrical muscle stimulation (EMS). Muscle health characterized as the composite of
muscle quantity, as well as functional and metabolic integrity, may be potentially maintained when optimal nutrition therapy is
provided in complement with early physical rehabilitation in critically ill patients; however, the type, dosage, and timing of these
interventions are unclear. This article explores the potential role of nutrition and EMS in maintaining muscle health in critical
illness. Within this article, we will evaluate fundamental concepts of muscle wasting and evaluate the effects of EMS, as well as the
effects of nutrition therapy on muscle health and the clinical and functional outcomes in critically ill patients. We will also highlight
current research gaps in order to advance the field forward in this important area. (Nutr Clin Pract. 2018;33:772–789)

Keywords
critical illness; nutrition support; intensive care unit; muscle stimulation; muscle wasting; protein intake

Introduction environments.”8 Traditionally, physical rehabilitation com-

There is a growing population of survivors of critical illness,
with almost 90% of patients surviving the initial insult of
critical illness.1 However, approximately 50% of survivors2
may develop intensive care unit–acquired weakness (ICU-
AW) and suffer ongoing significant morbidity in terms of
physical, cognitive, and psychological health.2-5 ICU-AW
refers to clinically detectable global muscle weakness that
develops as a result of no specific etiology other than being
critically unwell.6 Individuals may have evidence of my-
opathy, polyneuropathy, and/or significant muscle atrophy.
ICU-AW contributes to significant impairments in physical
functioning and decreased health-related quality of life
(HRQoL), which persist years after hospital discharge.2-5
Thus, it is essential to devote attention to the quality of
survivorship in terms of both clinical and physical function
outcomes.7
The World Health Organization defines rehabilitation
as “a set of measures that assist individuals who experi-
ence or are likely to experience disability, to achieve and
maintain optimum functioning in interaction with their
mences once the patient is awake and able to engage with
therapy. There is promising data to suggest that commenc-
ing within the first few days of an ICU admission may
be critical to minimizing the plethora of impairments that
can ensue.9,10 However, in the early stages of the ICU
admission there are numerous barriers, which can mean
From the 1 Department of Physiotherapy, The University of
Melbourne, Victoria, Australia; the 2 Intensive Care Research, Royal
Adelaide Hospital, South Australia, Australia; and the 3 Department
of Kinesiology, University of Waterloo, Ontario, Canada.
Financial Disclosure: Department of Health, Australian Government,
National Health and Medical Research Council (1111640).
Conflicts of Interest: None declared.
∗ All joint first author in this manuscript due to equal contribution.
This article originally appeared online on October 25, 2018.
Corresponding Author:
Selina M. Parry, PT, PhD, The University of Melbourne, Department
of Physiotherapy, School of Health Sciences, Level 7, Alan Gilbert
Building, 161 Barry St Carlton VIC 3010, Victoria, Australia.
Email: selina.parry@unimelb.edu.au
Parry et al
that it is challenging to provide “active” rehabilitation.11,12
We now look to technology that can be used to enable
patients to exercise without the need for volitional and direct
patient engagement.13 The evidence is emerging in terms
of the potential utility of assistive technologies such as
bedside cycle ergometry and electrical muscle stimulation
(EMS).10,14 Muscle stimulation in particular is an attractive
intervention because it enables targeted, artificial activation
and loading of individual muscles, with parameters being
adapted to provide strength or endurance load to the
muscle.
Muscle health can be characterized as a composite of
muscle quantity (measured as muscle mass, muscle cross-
sectional area, etc) as well as muscle quality (physical and
metabolic function of skeletal muscle).15 In critically ill
patients, muscle health may be supported when optimal
nutrition therapy is provided in addition to early physical re-
habilitation; however, the type, dosage, and timing of these
interventions are unclear.16-18 Nutrition support delivered
to ICU patients forms part of routine care for the majority
of ventilated patients. Marked catabolism occurs during
critical illness, and hence protein in particular may assist in
reducing at least some of the muscle wasting experienced
in the ICU setting. However, the delivery of nutrition does
not equate directly to utilization of energy in critical illness.
In other words, although one may be providing nutrition
support to ICU patients, patients’ tissues may develop
resistance to sufficiently taking up nutrients from enteral or
parenteral feeds to meet their energy demands. Critically ill
patients demonstrate a vast heterogeneity in energy needs
as well as magnitude in mitochondrial dysfunction and
dysregulated fat oxidation that unfavorably influence energy
generation in skeletal muscle.19 As such, we need to better
understand the metabolic energy demands of exercise in
critically ill patients to understand its bioenergetic impli-
cations and ultimately reduce muscle wasting and improve
functional recovery.20
Exercise is known to improve blood flow to skeletal
muscle in noncritically ill populations, which would enhance
nutrient and oxygen uptake by the muscle to maintain its
integrity and function. Optimal protein intake and exercise
could synergistically facilitate muscle protein deposition to
maintain muscle health. However, can these same concepts
be applied to a critically ill patient? Can muscle health be
preserved with EMS in an ICU patient? Can additional
protein be delivered and used by muscle in a critically
ill patient in the same manner as a healthy individual?
To address these questions, we will evaluate fundamen-
tal concepts of muscle wasting and evaluate the effects
of EMS, as well as the effects of nutrition therapy on
muscle health and the clinical and functional outcomes
in critically ill patients. We will also highlight current
research gaps in order to advance the field in this important
area.
773
Muscle Wasting in Critical Illness
Skeletal muscle is a highly plastic and adaptive tissue, which
responds to changes in mechanical loading and is 1 of
the largest tissue groups in the human body.21-23 Skeletal
muscle is an integral tissue in predicting clinical, physical,
and metabolic outcomes.17,24-26 To evaluate muscle health,
an aggregate of measures are essential, including muscle
mass and its integrity (ie, metabolic and physical features)
(Figure 1).15 Low muscle mass is reported in 20%–70%
of ICU patients at the time of ICU admission,25,26 and a
further ࣈ30% reduction in quadriceps muscle mass may
occur within 10 days of an ICU admission.27,28 Remarkably,
low muscle mass is also reported up to 6 and 12 months
following ICU discharge.24
Skeletal muscle atrophy is associated with morbidity,
increased hospital length of stay (LOS), and mortality.
Lower limb muscle mass is commonly associated with
strength, suggesting that reduction in muscle mass may
also contribute to poorer physical performance,28-32 and
these functional deficits may persist for years following
ICU discharge3 (Figure 1). Muscle atrophy has been
linked to elevated tumor necrosis factor-α and interleukin-
6 concentrations,33,34 and these have been further associ-
ated with increased risk of infection in older adults.35,36
Given that skeletal muscle comprises >75% of glucose
disposal,37,38 it is anticipated that glucose and fat dysregula-
tion may occur. For example, muscle atrophy as a result of
acute immobility has been associated with impaired glucose
tolerance,34 impaired glucose uptake,35 and reduced fat
oxidation.34 Although most of the weight lost during critical
illness is typically regained within 1 year following ICU
discharge,39 there is evidence of sustained muscle atrophy
with reduced muscle satellite cell content40 and substantial
fat gain24 up to 6 months and 1 year, respectively, following
ICU discharge. Thus, understanding the timeframe and
mechanisms of muscle atrophy will lead to developing meth-
ods that aim to preserve muscle mass and its physiological
functions.
There are multifaceted factors that lead to skeletal muscle
atrophy in critically ill patients, including prolonged bed
rest,21 catabolic signaling (including proinflammation and
insulin resistance),41 and undernourishment via low caloric
and protein intakes.42,43 Insufficient caloric intake generally
compromises protein intake, leading to reduced protein
synthesis. Amino acids provide the substrate for protein
synthesis44 ; thus, reduced untake may lead to reduced
protein synthesis.
However, in the presence of reduced activity and insulin
resistance, it is likely that critically ill patients also exhibit
anabolic resistance, which is the reduced ability of the mus-
cle to take up and use amino acids for protein synthesis.45
To compensate for the body’s high protein needs during
critical illness, skeletal muscle degradation is enhanced to
774
Figure 1. Factors affecting muscle health in pre-ICU, ICU trajectory, and ICU survivorship. ICU, intensive care unit.
provide amino acids for the synthesis of other proteins
that may be needed for immune function, for example,
resulting in muscle atrophy. On the other hand, bed rest,
independent of illness, can result in muscle loss.21 Reduced
blood flow is related to reduced delivery of amino acids to
skeletal muscle.46 Given the extent of bed rest combined
with accelerated catabolic processes in a critically ill patient,
reduced delivery of amino acids to skeletal muscle would
likely contribute extensively to muscle wasting.
It is unequivocally important to assess the implications
of morphological and metabolic deficits that evolve dur-
ing ICU stay. Loss of muscle quality may be evidenced
with infiltration of fat into skeletal muscle,47 which would
likely impair muscle glucose and fat metabolism. Muscle
characteristics and architecture including pennation angle,
echogenicity, and fascicle length are fundamental to force
production and consequently strength. During critical ill-
ness, these features deteriorate27,28 and result in functional
and strength deficits at ICU discharge.28 Skeletal muscle
relies on proper regulation of the electrical excitability
for adequate muscle contraction.48-50 Lower activation of
voltage-gated Na+ channels can reduce membrane ex-
citability, which leads to attenuated muscle contraction
and conseuqently muscle weakness.51 This acute neuropa-
thy is reported to be reversible—not degenerative—and
may be related to electrolyte abnormalities during critical
illness.48,49 Although this neuropathic mechanism may be
Nutrition in Clinical Practice 33(6)
acute, the consequential muscle atrophy that results may
be longer term. Targeted rehabilitative approaches that
require muscle contraction to preserve muscle mass or re-
covery from muscle atrophy would be essential to regaining
proper neural processes in rehabilitation of muscle strength.
However, given the unstable condition of many critically
ill patients, we need to consider a continuum of exercise
interventions that may begin with cost-effective and feasible
strategies in which participation is nonvolitional in nature
and then progresses to functional activities that require vol-
untary contraction as patient alertness and ability to engage
improve.
Principles of EMS
The quadriceps muscle is the largest muscle group in the
human body and integral to performance of activities such
as standing from a chair, stair climbing, and locomotion.
Given its size and importance in weight-bearing activities,
it is the most widely examined muscle within the muscle
stimulation literature.
An individual muscle consists of hundreds to thousands
of fibers arranged into functional groups called motor units.
The motor unit is the final common pathway in motor
processing, which would result in a physical action.52,53
Small motor units have small diameter axons and typically
innervate slow twitch (type I) muscle fibers, which are largely
Parry et al
Figure 2. Functional electrical stimulation-cycling machine
(RT-300 supine model and SAGE stimulator; Restorative
Therapies, Ltd, Baltimore, MD). Reprinted from Parry SM,
Berney S, Warrillow S. et al. Functional electrical stimulation
with cycling in the critically ill: A pilot case–matched control
study. J Crit Care. 2014:29:695.e1-e7, with permission from
Elsevier.
fatigue-resistant. Large motor units have large diameter
axons and typically innervate fast twitch (type II) muscle
fibers, which are less fatigue-resistant but enable greater
force production to be generated quickly.52,54 Physiologi-
cally, during volitional contraction muscle recruitment is
asynchronous and thought to follow the Hennemann’s size
principle with smaller units (ie, slow twitch, fatigue-resistant
fibers) recruited first. Because more force is needed, larger
units are recruited.23 The mechanism of recruitment with
EMS remains contentious.
Artificial muscle activation aims to preserve muscle mass
and/or recover/improve muscle function.54,55 Skeletal mus-
cle can be artificially stimulated to induce a visible and
palpable muscle contraction in the absence of volitional
control using 2 different methodologies:
1. Neuromuscular electrical stimulation (NMES),
which often involves stimulation of isolated muscle
groups in a nonfunctional manner (ie, stimulated in
supine)55
2. Functional electrical stimulation (FES), which gen-
erally involves stimulation of multiple muscle groups
while undertaking a combined functional activity
such as bedside cycling55,56 (Figure 2). The muscles
are artificially stimulated to contract at the point
in range at which volitional activation would occur
during the functional activity. For example, during
knee extension while cycling the quadriceps muscle
would be stimulated to contract, and on knee flexion
the hamstrings would be activated. This alternating
775
Table 1. Parameters of Relevance in Muscle Stimulation.14
Parameter Description
Amplitude Quantity of energy flowing per second measured in
(intensity) milliamplitude (mA)
Pulse width Duration of electrical impulse measured in
(duration) microseconds (usecs)
Frequency Number of electrical impulses per second (Hertz);
affected by twitch summation phenomenon,
commonly frequency is set between 35–50 Hz.
With higher frequencies >100 Hz tetany of the
muscle will occur and there is risk of faster
fatigability.
Ramp-up Current intensity will increase (ramp-up) to a preset
maximum over a defined period of time
Ramp-down Current intensity will decrease (ramp-down) to a
preset minimum over a defined period of time
On:Off Length of time over which each individual electrical
duration impulse is delivered versus no stimulation can be
preset with some muscle stimulators. For example
an On:Off duration of 5:1 means that the electrical
impulse is delivered for 5 seconds and off for
1 second.
muscle recruitment potentially may enable longer
time for contraction to be sustained prior to reaching
a point of fatigue.57
With both NMES and FES, there are a range of param-
eters that can be adapted to achieve muscle contraction.14
Table 1 provides an overview of these parameters. Sufficient
quadriceps tension may be achieved with frequencies of 5–
100 Hz and a pulse width of 100–400 usecs.55 With lower
pulse width, there is preferential activation of sensory fibers,
which can result in higher levels of sensory discomfort for
the patient. There is a direct inverse relationship between
pulse width and amplitude. Higher amplitudes are required
when using a lower pulse width in order to evoke muscle
contraction. With higher tissue impedance, as can occur
with presence of edema and adipose tissue, longer pulse
widths may be required. The size and location of surface
electrodes can influence the efficacy of muscle stimulation.
Ideally, electrodes should be placed close to the motor point
of a muscle, which is the point at which the motor units are
most superficial to the skin surface.
EMS in the ICU: What Do We Know?
There are a number of systematic and narrative reviews
published in the last 10 years on the topic of EMS within
the ICU setting13,14,58,59 in relation to muscle health.
During the writing of this article, we conducted a lit-
erature search and identified at least 23 distinct studies
published within the ICU literature specifically evaluating
the safety, feasibility, and/or efficacy of muscle stimulation
(NMES or FES) since 1987 (Table 2). More than half of the
literature (52%, n = 12/23) has been published since 2015,
highlighting the rapid and emerging interest in this area.
776
Table 2. Summary of Studies Examining EMS in Intensive Care.
Author Year, Time to First Muscles
Location N Population Session Stimulated EMS Parameters Main Findings

NMES
Bouletreau 1987, 10 Hospitalized for 8 or 12 days Gastroc Patient Position Significant reduction in excretion of
France110 at least 8 days (depending on Quads Not specified 3-MH in NMES
in the ICU allocation) Stimulation Parameters
0–120 V, 1.75 Hz, 3000 usecs
On:Off time: 5:5 sec
Training Parameters
30 min 2 × day for 4 days
Gerovasili 52 ICU patients 2 days Quads Patient Position Preservation of RF and VI muscle
2009,111 Routsi with APACHE peroneus Not specified thickness observed in NMES group
2010,61 II > 13 longus Stimulation Parameters (RF: −8%; VI: −13%) vs control
Karatzanos stratified based Mean mA for quads 38 (10) and 37 (11) mA for (RF: −14%; VI: −22%)111
2012112 Greece on age and peroneus longus No significant absolute or relative
gender 45 Hz, 400 usecs differences in handgrip strength112
On:Off time: 12 sec/6 sec Higher MRC scores and less ICU-AW
Ramp time: 0.8 sec with NMES group61,112
Training Parameters
55 min/day until ICU DC
Meesen 2010, 19 Hospitalised in Unclear Quads Patient Position Reduced thigh circumference loss in
Belgium113 ICU with MV Supine with half-roll under knee (to enable knee NMES vs control limb
> 1 day flexion)
Stimulation Parameters
0–5 min: 35–85 mA, 5 Hz, 250 usecs; Stimulation
intensity (mA) gradually increased in 2–10 mA
steps proportional to stimulation intensity in
warm-up trial
5–11 min: 60 Hz, 330 usecs
11–19 min: 100 Hz, 250 usecs
19–25 min: 80 Hz, 300 usecs
25–30 min: 2 Hz, 250 usecs
On:Off time: 90:30 sec (5 min): 10:20 sec (6 min):
10:20 sec (8 min): 7:14 sec (6 min): 90:30 sec
(5 min)
Training Parameters
Daily 30-min session as long as intubated/sedated
(NMES: right leg; sham: left leg)
Gruther 2010, 16 Short (< 7 days) ST group 3 (2) Quads Patient Position Significant increase in quadriceps
Austria62 and long-term days; Not specified muscle thickness (+4.9%) vs sham
ICU patients LT group 33 (15) Stimulation Parameters (−3.2%) for long-term group.
(> 14 days) days Max tolerated: mA not specified Significant loss of quadriceps muscle
50 Hz, 350 usecs thickness in both the short-term
On:Off time: 8 sec/24 sec NMES and sham groups
Training Parameters (ࣈ37%–39%).
30 min/day week 1
60 min/day week 2-week 4
5 sessions/week for 4 weeks
(continued)
 Table 2. (continued)
Author Year, Time to First Muscles
Location N Population Session Stimulated EMS Parameters Main Findings

Rodriguez 2012, 14 ICU patients 2 (1–2) days Biceps Patient Position No change in biceps thickness or
Argentina63 with sepsis Quads Not specified circumference with NMES
Stimulation Parameters Higher MRC scores for biceps and
100 Hz, 300 usecs quadriceps with NMES
On:Off time: 2 sec:4 sec
Training Parameters
30 min 2 × day continued until successful
extubation
Poulsen 2011, 8 ICU patients NR; baseline Quads Patient Position No preservation of quadriceps muscle
Denmark64 with septic measures Not specified volume observed with NMES. Both
shock assessed 26 Stimulation Parameters group had significant reduction
(16–52 hrs) 31 (23–48) mA for VM and 42 (37–54) mA for within 1 week (NMES 20%; control
VL 16%).
35 Hz, 300 usecs
On:Off time: 4 sec:6 sec
Ramp time: 0.5 secs
Training Parameters
60 min daily for 1 week
Angelopoulos 31 Acute group: Unclear Quads Patient Position Both medium and high-frequency
2013, presence of Peroneus Not reported stimulation induced
Greece114 SIRS criteria longus Stimulation Parameters microcirculatory changes to skeletal
or diagnosis of High-Frequency Protocol muscle
sepsis for 3–5 57–87 mA; 75 Hz, 400 usecs
days at time of On:Off time: 5 sec/21 sec
session Ramp time: 1.5 sec up/0.8 down
ICU-AW Group: Medium Frequency Protocol
individuals 45 Hz, 400 usecs
who had On:Off time: 5 sec/12 sec
confirmed Ramp time: 1.5 sec up/0.8 down
ICU-AW using Training Parameters
MRC-SS 30 min (with additional 5 min warm-up and 5
min recovery phase using 10 Hz current of 400
usecs), single session (pre-/postevaluation)
Hirose 2013, 25 Coma within first >7 days after Quads Patient position Preservation of leg muscle mass
Japan 115 24 hr of hospi- admission Tib ant Not specified observed with computed
talization; Hamstrings Stimulation Parameters tomography in EMS group
first-time Gastroc 40 mA, Hz and usecs not specified (compared with control)
stroke or TBI On:Off time 10:10 sec
injury; 16–75 Training Parameters
years of age 30 min daily starting 7 days after admission (3
with paralysis min warm-up; 25 min training; 2 min
of both or 1 cool-down)
limb Ceased after 14 days

(continued)

777
778
Table 2. (continued).

Author Year, Time to First Muscles

Location N Population Session Stimulated EMS Parameters Main Findings

Flavigna 2014, 11 Adult ICU Quads Patient Position Goniometry measurement of

Brazil116 patients MV Tib Anterior Not specified ankle dorsiflexion range was
for up to 48 Stimulation Parameters higher in the EMS limb
hours 18–60 mA, 50 Hz, 400 usecs compared with control
On:Off time: 9:9 sec No significant difference in
On time: 2 sec rise time, 5 sec MRC scores for muscle
contraction, 2 sec fall time strength
Training Parameters No significant difference in
20 min/daily until patient attained circumference measures
grade 4 strength for muscle except at 10-cm distance for
stimulated the thigh; lower scores in
control group
Segers 2014, 50 SICU/MICU Day 3 Quads Patient Position Nonresponders (unable to elicit
Belgium117 patients Supine, head-up 30° muscle contraction) were
enrolled on Legs in neutral: solid knee support more likely to have sepsis,
days 3–5 of roll placed under knees to achieve edema, and receipt of
admission approx. 15° hip flex and 30° knee vasopressors
whom had flex Inverse relationship between
an expected Stimulation Parameters edema quantity and type of
prolonged 0–80 mA, 50 Hz, 300–500 usecs muscle contraction
LOS for at Series time: 8 sec Series pause: 20 sec Could not predict adequate
least 3 more Ramp time: 2 sec contraction using NCS or
days Training Parameters EMG
5 times/week (Mon-Fri) With 50% of sessions erythema
25 min/day until ICU DC (5-min evident under the electrodes
warmup) but disappeared immediately
post sessions. No significant
cardiovascular or respiratory
response changes with
NMES

(continued)
 Table 2. (continued).

Author Year, Time to First Muscles

Location N Population Session Stimulated EMS Parameters Main Findings

Dirks 2015, 9 Expected sedation <4.5 days Quads Patient Position NMES leg showed no atrophy in
Netherlands65 time of >3 days Not specified type I or II muscle fibers
and admitted to Stimulation Parameters compared with control leg,
the ICU Intensity set to level at which full contractions were which had a significant decline
visible and palpable, intensity every 3 min of 16(9%) and 24(7%) in type I
Averaged 29.9 mA in first session and progressively and II muscle fiber CSA.
increased to 32.3 mA in final session
Warm-up 5 min (5 Hz, 250 usecs)
Stimulation period (30 min 100 Hz, 400 usecs, 5 sec
on (0.75 sec rise, 3.5 sec contraction, 0.75 sec fall)
and 10 sec off
Cool-down 5 mins (5 Hz, 250 usecs)
Training Parameters
2 sessions per day over period of 3–10 days (NMES
leg and control leg)
Akar 2015, 30 Diagnosis of Unspecified Deltoid Patient Position NMES and NMES + exercise
Turkey118 COPD and quads Not specified groups had significant
respiratory Stimulation Parameters improvements in lower
failure 20 mA-25 mA, 50 Hz, pulse width not reported, 6-sec extremity strength, whereas
treatment with duration, 1.5-sec increase and 0.75-sec decrease upper limb strength increased
IMV without Training Parameters across all groups
comorbidities 5 times/week for a total of 20 sessions duration not No difference for weaning time or
specified functional milestones
Kho 2015, 34 Adult ICU 4.6 (1.8) days Quads Patient Position 63% (n = 146/230) sessions
USA66 patients MV Tib ant Not specified completed
> 24 hrs and Gastroc Stimulation Parameters No significant difference in leg
expected to Ramp up/down: 2 sec: < 1 sec, 50 Hz muscle strength at hospital
require > 2 days Quads: 400 usecs, On:Off time: 5 sec:10 sec discharge (however, studies
more in the ICU Tib ant/gastroc: 250 usecs discontinued before reaching a
On:Off time: 5 sec:5 sec (alternating recruitment of priori sample size)
Tib ant and gastroc) Patients receiving NMES had
Training Parameters larger increase in leg strength
60 mins/day (either single or 2 × 30 mins sessions) between ICU awakening and
9.1 (8.7) days of NMES delivered ICU discharge, and between
ICU awakening and hospital
discharge. NMES group walked
>2 times further at hospital
discharge compared with
control group.

(continued)

779
 Table 2. (continued).

780
Author Year, Time to First Muscles
Location N Population Session Stimulated EMS Parameters Main Findings

Fischer 2016, 54 Adult ICU patients Day 1 Quads Patient position 94% sessions delivered
Austria67 undergoing Not specified (n = 136/145)
cardiothoracic Stimulation Parameters NMES no significant effect on
surgery and Median NMES intensities was 40–40.5 mA MLT quadriceps; NMES group
expected to remain range (2–120) mA regained strength 4.5 times
in ICU >48 hrs 0.4 ms, 66 Hz faster than patients in control
On:Off time 3.5 sec: 4.5 sec group
Ramp-up/-down: 0.5 sec
Training Parameters
2 × 30 mins, 7 days/week during entire ICU
stay (but not more than 14 days)
Dall’Acqua 25 Adults hospitalized < 48 hrs of Pec major Patient Position No complications or significant
2017, for no longer than IMV Rectus Supine changes in vital signs
Brazil60 15 days and > 24 abdominas Stimulation Parameters Significant difference with NMES
hrs of invasive MV Pec major: 53 (15) mA in terms of muscle thickness
Rec abdominas: 68 (18) mA and shorter ICU LOS
50 Hz, 300 usecs Preservation of muscle mass in
Ramp-up/-down: 1 sec NMES group (compared with
On:Off time: 3 sec:10 sec control group with significant
Training Parameters decrease in muscle mass across
NMES 30 min daily, ceased on day 7 or time)
when patient extubated or if deceased
(whichever occurred first)
Patsaki 2017, 128 ICU patients MV > ICU discharge Quads Patient Position No difference in MRC scores,
Greece 119 72 hrs and level of Peroneus Not specified handgrip strength, functional
consciousness longus Stimulation Parameters status, and hospital LOS
adequate to 45 Hz, 400 usecs between NMES and control
respond to at least On:Off time: 12 sec:6 sec groups
3 of 5 commands Rise/Fall time: 0.8 sec:0.8 sec In patients with ICU-AW the
stratified based on Training Parameters NMES group had significantly
age and diagnosis 55 mins daily 7 days/wk until hospital higher MRC scores at 2 weeks
of ICU-AW (MRC discharge compared with control
sum score ࣘ 48 or
> 48)
Shen 2017, 25 ICU patients MV > Enrolled on Biceps Patient position 68% of patients unable to
Taiwan 72 hrs ICU Day 3 Quadriceps Not specified complete handgrip strength
Stimulation Parameters No significant difference in
0–75 mA, 15,000 Hz, pulse width NR handgrip strength and MV time
Training Parameters between groups
32 mins daily 5 days/ week

(continued)
 Table 2. (continued).

Author Year, Time to First Muscles

Location N Population Session Stimulated EMS Parameters Main Findings

Silva 2017, 11 Adult APACHE II ICU Day 1 Glut max 56 (15) mA for glut max Feasibility and Safety:
Brazil120 score > 13, MV Gastroc 55 (13) mA quads N = 83 (85%) of sessions
24–48 hr with Tib ant 52 (13) mA hamstrings completed
prediction to Hamstrings 57 (13) mA tib ant N = 99 (100%) of sessions a
remain MV for Quads 57 (11) mA gastroc visible contraction present
at least 3 days During the 3 days pulse width No skin burns
by senior ICU 550 (170) usecs glut max Setup time: 107 (24) min
physician on 450 (150) hamstrings Other:
admission 500 (200) usecs quads NS change in CPK levels
550 (150) usecs tib ant
450 (150) usecs gastroc
Daily for 3 days
Once a day for 15 mins (90 contractions)
pulse width equal to chronaxie, 100 Hz
On:Off time: 5 sec:5 sec
No rise time or decay
No warm-up or cool- down period
3 times total of 45 minuntes
Woo 2017, 10 Patients aged 20 Unclear Quads Patient Position No difference in circumference
Korea121 years or older Supine knee extended and quads CSA between legs
admitted to Stimulation Parameters (no additive benefit of NMES
ICU and Lowest intensity with visible muscle with cycling)
required MV for contraction
at least 24 hrs 35 Hz, 250 sec
ON:off time: 10 sec:12 sec
Training Parameters
Cycling provided prior and separately to
isolated muscle stimulation. 10-min rest
prior to NMES, which was applied for 20
min on left thigh. Single session
FES
Parry 2015, 16 Adults admitted Median of Quads Patient Position 73% of sessions delivered. One
Australia56 with a diagnosis 15.3 Hamstrings Supine head-up 30° transient episode of
of sepsis or (12-31.5) hr Gluteals Stimulation Parameters desaturation; nil other safety
severe sepsis between Gastroc 0–140 mA, 300–400 usecs, 30–50 Hz events.
and predicted to recruit to Training Parameters Trend toward earlier and faster
be MV > 48 hrs first 20–60 min/day, 5 × week until ICU recovery of functional
and remain in treatment discharge milestones in intervention group
ICU > 4 days compared with control

(continued)

781
 Table 2. (continued).

Author Year, Time to First Muscles

782
Location N Population Session Stimulated EMS Parameters Main Findings

Fewer individuals required inpatient

rehabilitation in the intervention
group (43%) vs control (86%).
Lower frequency of delirium vs
control (25:87%) although not
significant. Duration of
delirium was significantly
shorter in intervention group
(0 [0–3] days) vs control
(6 [3.3–13.3])
Medrinal 19 Patients > 18 Unclear Quadriceps Patient Position Safety:
2018, years MV at During FES cycling, NMES synchronized with Only FES cycling increased CO
France68 least 24 hrs and knee extension, otherwise supine and produced sufficient
ventilated with Stimulation Parameters intensity of muscle work
pressure Stimulation intensity aimed to obtain a palpable Increase in CO by 1 L/min after 9
support contraction min of FES cycling
300 usecs, 35 Hz, No change in CO over time with
no ramp time NMES
Training Parameters Significant increase in HR,
4 consecutive 10-min sessions of bed exercise (10 TAPSE, and MAP during FES
mins PROM, 10 mins NMES, 10 mins cycle cycling. FES cycling increased
ergometry, 10 mins FES cycling), order of CO and physiological
intervention/sec Randomized. 30-min rest cardiorespiratory response and
between each intervention. Single session. reduced muscle HbO2
Fossat 2018, Patients > 18 31 (19-46) hr Quads Patient Position Safety: 1 skin allergy to electrode
France69 years, admitted Supine pads for NMES
to ICU < 72 hrs Stimulation Parameters No difference in combined cycling
prior to Based on Routsi61 + NMES in addition to usual
randomization Training Parameters care in terms of global muscle
and expected to 15 mins cycling + 50 mins NMES 5 days/week strength ICU discharge
need another 48 until ICU discharge No significant difference in
hrs in the ICU, ventilator-free days or
independent self-reported health-related
mobility quality-of-life at 6 months

APACHE II, Acute Physiologic Assessment and Chronic Health Evaluation II; CO, cardiac output; COPD, chronic obstructive pulmonary disease; CPK, creatine phosphokinase test; CSA,
cross sectional area; EMG, electromyography; EMS, electrical muscle stimulation; FES, functional electrical stimulation; Gastroc, Gastrocnemius; Glut max, gluteus maximus; HR, heart rate;
hr, hour(s); Hz, hertz; ICU, intensive care unit; ICU-AW, intensive care unit–acquired weakness; ICU DC, ICU discharge; IMV, invasive mechanical ventilation; LOS, length of stay; LT, long
term; mA, milliamplitude; MAP, mean arterial pressure; min, minute(s); MICU, medical ICU; MLT, muscle length thickness; MRC, Medical Research Council; MRC-SS, Medical Research
Council sum-score; MV, mechanical ventilation; NCS, nerve conduction study; NMES, neuromuscular electrical stimulation; n, number; NR, not reported; NS, not significant; Pec major,
pectoralis major; Rec abdominas, Rectus abdominas; RF, rectus femoris; Rx, treatment; ST, short term; Tib ant, tibialis anterior; TBI, traumatic brain injury; Quads, Quadriceps; sec,
second(s); V, voltage; VI, vastus intermedius; VL, vastus lateralis; VM, vastus medialis; SICU, surgical ICU; SIRS, systemic inflammatory response syndrome; usecs, microseconds; 3-MH,
3-methylhistidine; %, percentage; >, greater than; <, less than.
Parry et al
Whereas the majority of the research has been conducted in
Europe (61%, n = 13/23), publications from South America
(n = 3/23), Asia (n = 3/23), North America (n = 1/23), and
Australia (n = 1/23) are also evident.
Although the quadriceps muscle is the most widely
evaluated muscle group, there was 1 study that reported on
the preservation of muscle mass for pectoralis major and
rectus abdominus muscles after NMES, as well as reduced
ICU LOS.60 There is significant heterogeneity within the
studies in terms of stimulation parameters, intervention
duration, and timing of delivery relative to time in the
ICU (Table 2). The evidence is currently inconclusive in
terms of providing definitive guidance on the efficacy of
muscle stimulation within the ICU setting. In the studies,
which reported the time to first intervention session, the
majority commenced within the first 5 days of the admission
period (65%, n = 15 of 23) (Table 2). There is conflicting
evidence for the preservation of muscle mass using NMES,
particularly in individuals with higher severity of illness.60-64
Muscle biopsy data in 1 study demonstrated absence of
atrophy of type I and II muscle fibers.65 Data suggesting
a potential for accelerated gains in strength and physical
function have been reported at the time of ICU and hospital
discharge.66,67 However, there is limited evaluation of the
potential longer-term benefits of muscle stimulation beyond
hospital discharge, which is an important area for further
evaluation.
FES literature first emerged in 2015 with a pilot case
control FES cycling study in the ICU that primarily focused
on determining the safety and feasibility of early interven-
tion (ie, within a couple of days of ICU admission).56 The
study demonstrated safety and feasibility with 1 transient
adverse event (desaturation) and delivery of three-quarters
of potential intervention sessions.56 There was a trend
toward faster recovery of functional milestones, reduced
delirium, and reduced need for rehabilitation posthospital
discharge.56 Medrinal et al68 explored the impact of 4
different assistive rehabilitation strategies (passive range of
motion exercises with therapist, NMES, cycle ergometry,
and FES) on cardiorespiratory response to exercise. The
study demonstrated that FES cycling was the main in-
tervention, which increased cardiac output and produced
sufficient intensity of muscle work; however, a recent study
that combined NMES and cycling reported no difference
in outcomes in terms of muscle strength and ventilator-free
days or HRQoL at 6 months.69
Beyond the heterogeneity in muscle stimulation param-
eters, training duration, and timing to intervene, there
are also potential confounding factors unique to the
ICU population, which may contribute to the conflicting
evidence regarding the efficacy of muscle stimulation.
Tissue impedance will impact the intensity and delivery
of current flow toward the underlying motor units.54,55,70
Increased edema and high adipose tissue, 2 factors that
783
can be present in ICU patients, will negatively affect the
ability to deliver the intended stimulation intensity. In these
circumstances, a higher pulse width and/or intensity will be
required in order to induce a muscle contraction. However,
this often results in poorer tolerance due to increased
discomfort with sensory-evoked pain. There is emerging
interest in the role of thermally enhanced muscle stimu-
lation, which may assist with some of the issues around
edema and sensory discomfort while optimizing stimulation
parameters.71 Importantly, given that optimal muscle health
may be achieved with combined muscle contraction and
enhanced nutrition deliver, there is a need to investigate
the potential synergism between NMES and optimized
nutrition intervention in future research.
Although there are several merits to the use of EMS,
various issues need to be addressed, which include:
r Developing optimal stimulation parameters (inten-
sity, pulse width, frequency) without compromising
muscle force production due to fatigability
r Identifying those who respond versus those who do
not respond to muscle stimulation; does this modality
work for all patients?
r Developing an objective means of quantifying the
quality of muscle contraction
r Understanding the longer-term impact of muscle
stimulation on patient-oriented outcomes
r Further evaluate the safety and feasibility of muscle
stimulation
Is There a Role for Protein to Improve Muscle
Health?
Protein delivery is considered vital for muscle maintenance
in both health and disease. Because critical care manage-
ment shifts to focus on strategies to maintain muscle mass
and function, attention has been placed on the role of
protein in maintaining muscle health in this population.72
Accordingly, several narrative reviews advocate increased
protein delivery to critically ill patients,73-76 most notably
by Hoffer and Bistrian, which strongly suggested that
providing 2–2.5 g/kg/d is safe and could be optimal during
critical illness.77 Delivery of greater protein doses is also
recommended in a number of international guidelines: the
2016 update of the American Society for Parenteral and En-
teral Nutrition critical care nutrition guidelines recommend
delivery of protein doses of 1.2–2.0 g/kg/day or higher,78
and the European Society of Parenteral and Enteral Nu-
trition Parenteral Nutrition Guidelines for Critical Care
recommend providing 1.3–1.5 g protein/kg/day.79 These
recommendations are, however, based on very low quality
of evidence.
Early physiological studies show nitrogen balance is
negative during states of inflammation.80 More recently,
784
Figure 3. Potential effects of combined EMS and increased protein intake on muscle health during the ICU trajectory and
survivorship. (Text in light gray font indicates factors that are likely to be unaffected by EMS.) EMS, electrical muscle stimulation;
ICU, intensive care unit.
stable isotope-labeled amino acid tracer techniques have
been used to provide an estimate of protein metabolism
(synthesis, breakdown, and oxidation) at the whole-body
level. Rooyackers et al used this technique to demonstrate
that critically ill patients with multiorgan failure have an
increase in both protein synthesis and breakdown.81 Further
work from Sweden has shown that the addition of par-
enteral nutrition (PN) or enteral nutrition (EN) results in an
overall small improvement in whole-body protein balance.
Whereas nutrient provision appears to stimulate protein
breakdown, it also stimulates protein synthesis to a greater
degree, resulting in an overall positive protein balance.82-84
However, these studies only measure incorporation of la-
beled amino acids into the plasma, and hence it is unknown
whether additional amino acids are delivered, taken up,
and utilized by the muscle during critical illness. An expert
review published in 2017 thoroughly discusses the complex-
ity of protein kinetics in critical illness, demonstrating that
a greater understanding of the physiological response to
protein supplementation during different clinical conditions
is required.85
A number of observational studies show an association
between greater protein delivery and improved clinical
outcomes, including mortality and time to discharge from
ICU alive.86-88 However, few studies have shown a re-
Nutrition in Clinical Practice 33(6)
lationship between protein delivery and muscle health.
Puthucheary et al conducted an observational study in 63
critically ill adults and reported an association between
greater protein delivery and increased loss in ultrasound-
derived quadriceps muscle.27 Two RCTs of differing protein
doses have been conducted thus far, which both were
single-centered and evaluated the effect of protein dose
on ultrasound-derived muscle size. The first, by Ferrie
et al, used PN to deliver augmented protein (1.1 g/kg/day)
compared with standard care (0.9 g/kg/day) in 119 critically
ill adults. Greater protein was associated with an increased
forearm muscle thickness, with no difference in measures of
strength.89 Another Australian group showed that greater
enterally delivered protein to 60 critically ill patients was
associated with attenuation of quadricep muscle loss at ICU
discharge.90 It is important to consider that there is much to
be learned by the distinct study designs used in these studies,
which may have led to discrepant results. Future work
exploring protein intervention on specific muscle outcomes
is necessary to better understand the role of protein delivery
during critical illness (Figure 3).
There is also work being conducted on other nutrition
strategies that may influence muscle health during critical
illness. Bear et al are exploring the role of intermittent versus
continuous EN in critical illness on ultrasound-derived
Parry et al
rectus femoris cross-sectional area, for which results are
expected this year (ClinicalTrials.gov NCT02358512).
This same group is exploring the role of β-hydroxy-
β-methylbutyrate, a derivative of leucine, on the same
outcome (ClinicalTrials.gov NCT03464708). Although
physiological studies demonstrate improved protein balance
with nutrition support during critical illness, and pilot
studies show potential attenuation of muscle wasting with
protein delivery, it is unknown whether nutrition therapy
alone is able to reduce the significant muscle atrophy that
occurs during critical illness, particularly early after injury
for which protein catabolism is at its greatest.91 It also needs
to be considered whether early physical rehabilitation dur-
ing critical illness places a subsequent demand on increased
nutrient availability to facilitate muscle synthesis. Therefore,
it is thought that the combination of nutrition support and
exercise therapy during critical illness may provide the great-
est benefit on muscle attenuation and functional recovery.
How Could We Combine Exercise and Protein
for Best Patient Outcomes?
In health, the combination of exercise and amino acid
delivery has been shown to enhance the capacity of the
skeletal muscle to promote protein synthesis.92 In non-
ICU studies, combining protein and exercise demonstrate
far greater benefits in terms of muscle mass, strength, and
physical functioning compared with either therapy on their
own.92,93 Pennings et al showed low-intensity cycling and
resistance-type tasks prior to amino acid consumption in-
creased muscle protein synthesis rates when compared with
rest.94 Similarly, Biolo et al showed amino acid transport
after exercise is 30%–100% greater than following rest,95 and
this may be related to exercise-induced nutrient-stimulated
vasodilation.96 Bed-rest studies demonstrate the benefit of
multimodal interventions. Trappe et al97 conducted a 3-arm
study in 24 healthy women. During 60 days of bed rest,
participants underwent bed rest only, bed rest plus exercise
regime, or bed rest plus a leucine-enhanced high-protein
diet to 1.6 g/kg/day. Exercise therapy, which included a
combination of aerobic and resistance exercises, was shown
to have a positive impact on muscle attenuation, whereas it
was reported that the greatest muscle loss occurred in those
patients who received the high-protein diet.97
In critical illness, despite the coupling of protein and
exercise being identified as a key priority area during the
International Summit of Critical Care experts,98 there are
few studies with observational or controlled interventions,
that report data on both physical function and nutrition
intake. In a poster abstract, Wappel et al presented a
subanalysis of a pilot study that compared exercise alone
with an exercise-nutrition intervention that also included
NMES on the outcome of cross-sectional area of the
quadriceps.99 Although not powered to show an effect, the
785
group receiving NMES and nutrition support had a 2-fold
reduction in loss of muscle CSA. In addition, there are 2
registered trials currently recruiting that combine exercise
and nutrition interventions: 1) NEXIS (ClinicalTrials.org
NCT03021902), combining cycle ergometry and amino
acid supplementation versus standard care; and 2) ExPrEs
(ClinicalTrials.org NCT02509520), combining NMES and
leucine supplementation in elderly ICU patients.
In designing studies that provide both exercise and
nutrition early in critical illness, a number of factors need
to be carefully considered:
1. The route of nutrition support is important. We know
that in critical illness delayed gastric motility100
and impaired nutrient absorption of both fat101
and carbohydrate102 affect nutrient uptake from
enterally delivered nutrition. Whether enterally de-
livered protein can be adequately absorbed and
incorporated into muscle during critical illness is
yet to be determined (ANZCTR Trials Registration;
ACTRN12616001652460). Whereas EN may be the
preferred and commonly used route when compared
with PN, it is important to consider how much of
the nutrition delivered is available for muscle uptake.
In the elderly, postprandial splanchnic extraction
(uptake of amino acids for the splanchnic organs)
is increased, reducing the availability of amino acids
for muscle synthesis.103,104 Whether an exacerbation
of splanchnic sequestration during critical illness
occurs requires quantification.
2. The duration of intervention should be contemplated.
The majority of large randomized controlled trials
of nutrition interventions during critical illness are
only delivered for a short period of time early during
the ICU admission. This is despite studies report-
ing reduced nutrition intake in survivors of critical
illness,105,106 which is likely to influence functional
recovery.
3. Timing of both nutrient and exercise therapy should be
considered. In critical illness, it is common practice
to provide nutrition continuously, whereas exercise
interventions are likely to be administered at set
timepoints. As discussed by Bear et al in their review
on the pros and cons of feeding modalities, there
may be cause to believe that bolus or intermittent
delivery of EN will have a greater influence on
muscle protein synthesis.107 Based on this premise,
the effect of intermittent versus continuous EN on
ultrasound-derived muscle wasting is being explored
(ClinicalTrials.gov NCT02358512). Timing relative
to the exercise (ie, pre- vs postexercise nutrition)
may also have implications on the extent of protein
synthesis and degradation.17,94
786
4. Bioenergetics of critically ill patients need to be
comprehensively investigated. We need to understand
the bioenergetics and physiological ability of the
muscle to respond to muscle stimulation, particu-
larly in high severity of illness in which there may be
compromised muscle membrane function and mito-
chondrial dysfunction.20 The metabolic demands of
exercise are poorly understood within the ICU
setting.20 Pilot data demonstrates that energy re-
quirements are likely to be higher in individu-
als with critical illness compared with the healthy
population.108 We postulate that energy demands
will be lower with muscle stimulation compared with
use of technology such as cycle ergometry or func-
tional rehabilitation, but this needs to be explored.
We also need to better understand whether amino
acid uptake and protein turnover maintain their
capacity during critical illness to preserve muscle
integrity.
5. The patient population needs to be clearly defined.
ICU studies involve a highly heterogenous group
of individuals. However it is important to recognize
that the response to nutrition and exercise interven-
tions is likely to be influenced by a number of patient
characteristics, including sarcopenia, age, comor-
bidities, and obesity. Recent work has also proposed
that genetic predisposition may predict the extent
of muscle wasting.109 Therefore, consideration and
categorization of the patient population are vital.
Conclusions
Muscle dysfunction is common for survivors of critical
illness and remains a significant issue; it may influence
strength and physical performance. Current research sug-
gests that early intervention within the first few days may
be the critical window for improvement of strength and
functional recovery outcomes. However, our understanding
of the physiological changes that occur in muscle during
critical illness needs to be further explored. This work would
inform the potential effectiveness of combining optimal
protein nutrition with electrical stimulation. To preserve
muscle health and potentially improve clinical and patient-
oriented outcomes, there is a continual and significant need
to understand the mechanistic effects of combined therapies
in conjunction with consideration of the dosage, timing, and
delivery methods.
Statement of Authorship
S. M. Parry, L-A. S. Chapple, and M. Mourtzakis equally
contributed to the conception and design of the research;
S. M. Parry, L-A. S. Chapple, and M. Mourtzakis contributed
to the acquisition and analysis of the data; S. M. Parry, L.-A. S.
Chapple, and M. Mourtzakis contributed to the interpretation
Nutrition in Clinical Practice 33(6)
of the data; and S. M. Parry, L-A. S. Chapple, and M. Mourtza-
kis drafted the manuscript. All authors critically revised the
manuscript, agree to be fully accountable for ensuring the
integrity and accuracy of the work, and read and approved
the final manuscript.
References
1. Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality
related to severe sepsis and septic shock among critically ill patients in
Australia and New Zealand, 2000–2012. JAMA. 2014;311(13):1308-
1316.
2. Appleton RTD, Kinsella J, Quasim T. The incidence of intensive care
unit-acquired weakness syndromes: A systematic review. J Intensive
Care Soc. 2015;16(2):126-136.
3. Herridge MS, Tansey CM, Matte A, et al. Functional disability
5 years after acute respiratory distress syndrome. N Engl J Med.
2011;364(14):1293-1304.
4. Hermans G, Van Mechelen H, Clerckx B, et al. Acute outcomes and
1-year mortality of intensive care unit-acquired weakness. A cohort
study and propensity-matched analysis. Am J Respir Crit Care Med.
2014;190(4):410-420.
5. Fan E, Dowdy DW, Colantuoni E, et al. Physical complications in
acute lung injury survivors: a two-year longitudinal prospective study.
Crit Care Med. 2014;42(4):849-859.
6. Stevens R, Marshall S, Cornblath D, et al. A framework for diagnos-
ing and classifying intensive care unit-acquired weakness. Crit Care
Med. 2009;37(10):S299-308.
7. Needham DM, Davidson J, Cohen H, et al. Improving long-
term outcomes after discharge from intensive care unit: report
from a stakeholders’ conference. Crit Care Med. 2012;40(2):
502-509.
8. (WHO), W.H.O. Concept Paper – WHO Guidelines on Health-
Related Rehabilitation (Rehabilitation Guidelines), 2011.
9. Tipping CJ, Harrold M, Holland A, Romero L, Nisbet T, Hodgson
CL. The effects of active mobilisation and rehabilitation in ICU
on mortality and function: a systematic review. Intensive Care Med.
2017;43(2):171-183.
10. Hodgson CL, Tipping CJ. Physiotherapy management of intensive care
unit-acquired weakness. J Physiother. 2017;63(1):4-10.
11. Parry SM, Nydahl P, Needham DM. Implementing early physi-
cal rehabilitation and mobilisation in the ICU: institutional, clin-
ician, and patient considerations. Intensive Care Med. 2018;44(4):
470-473.
12. Parry SM, Remedios L, Denehy L, et al. What factors affect imple-
mentation of early rehabilitation into intensive care unit practice? A
qualitative study with clinicians. J Crit Care. 2017;38:137-143.
13. Needham DM, Truong AD, Fan E. Technology to enhance physical
rehabilitation of critically ill patients. Crit Care Med. 2009;37(10
Suppl):S436-S441.
14. Parry SM, Berney S, Granger CL, Koopman R, El-Ansary D,
Denehy L. Electrical muscle stimulation in the intensive care setting:
a systematic review. Crit Care Med. 2013;41(10):2406-2418.
15. Mourtzakis M, Parry S, Connolly B, Puthucheary Z. Skeletal Muscle
Ultrasound in Critical Care: A Tool in Need of Translation. Ann Am
Thorac Soc. 2017;14(10):1495-1503.
16. Bear DE, Wandrag L, Merriweather JL, Connolly B, Hart N, Grocott
MPW. The role of nutritional support in the physical and func-
tional recovery of critically ill patients: a narrative review. Crit Care.
2017;21(1):226.
17. Heyland DK, Stapleton RD, Mourtzakis M, et al. Combining nutri-
tion and exercise to optimize survival and recovery from critical illness:
Conceptual and methodological issues. Clin Nutr. 2016;35(5):1196-
1206.
Parry et al
18. Denehy L, Granger CL, El-Ansary D, Parry SM. Advances in
cardiorespiratory physiotherapy and their clinical impact. Expert Rev
Respir Med. 2018;12(3):203-215.
19. Puthucheary ZA, Astin R, McPhail MJW, et al. Metabolic phenotype
of skeletal muscle in early critical illness. Thorax. 2018.
20. Bear DE, Parry SM, Puthucheary ZA. Can the critically ill patient
generate sufficient energy to facilitate exercise in the ICU? Curr Opin
Clin Nutr Metab Care. 2018;21(2):110-115.
21. Parry SM, Puthucheary ZA. The impact of extended bed rest on
the musculoskeletal system in the critical care environment. Extrem
Physiol Med. 2015;4:16.
22. de Boer MD, Maganaris CN, Seynnes OR, Rennie MJ, Narici MV.
Time course of muscular, neural and tendinous adaptations to 23 day
unilateral lower-limb suspension in young men. J Physiol. 2007;583
(Pt 3):1079-1091.
23. Sherwood L. Human Physiology: From Cells to Systems, ed. CL Inc.
Vol. 9th Ed. 2016.
24. Chan KS, Mourtzakis M, Aronson Friedman L, et al. Evaluat-
ing Muscle Mass in Survivors of Acute Respiratory Distress Syn-
drome: A 1-Year Multicenter Longitudinal Study. Crit Care Med.
2018;46(8):1238-1246.
25. Moisey LL, Mourtzakis M, Cotton BA, et al. Skeletal muscle predicts
ventilator-free days, ICU-free days, and mortality in elderly ICU
patients. Crit Care. 2013;17(5):R206.
26. Weijs PJ, Looijaard WG, Dekker IM, et al. Low skeletal muscle area
is a risk factor for mortality in mechanically ventilated critically ill
patients. Crit Care. 2014;18(2):R12.
27. Puthucheary ZA, Rawal J, McPhail M, et al. Acute skeletal muscle
wasting in critical illness. JAMA. 2013;310(15):1591-1600.
28. Parry SM, El-Ansary D, Cartwright MS, et al. Ultrasonography in
the intensive care setting can be used to detect changes in the quality
and quantity of muscle and is related to muscle strength and function.
J Crit Care. 2015;30(5):1151.e9-14.
29. de Bruin PF, Ueki J, Watson A, Pride NB. Size and strength of the
respiratory and quadriceps muscles in patients with chronic asthma.
Eur Respir J. 1997;10(1):59-64.
30. Seymour JM, Ward K, Sidhu PS, et al. Ultrasound measurement
of rectus femoris cross-sectional area and the relationship with
quadriceps strength in COPD. Thorax. 2009;64(5):418-423.
31. Young A, Stokes M, Crowe M. Size and strength of the quadriceps
muscles of old and young women. Eur J Clin Invest. 1984;14(4):282-
287.
32. Hayashida I, Tanimoto Y, Takahashi Y, Kusabiraki T, Tamaki J.
Correlation between muscle strength and muscle mass, and their as-
sociation with walking speed, in community-dwelling elderly Japanese
individuals. PLoS One. 2014;9(11):e111810.
33. Visser M, Pahor M, Taaffe DR, et al. Relationship of interleukin-
6 and tumor necrosis factor-alpha with muscle mass and muscle
strength in elderly men and women: the Health ABC Study. J Gerontol
A Biol Sci Med Sci. 2002;57(5):M326-M332.
34. Schaap LA, Pluijm SM, Deeg DJ, et al. Higher inflammatory marker
levels in older persons: associations with 5-year change in muscle mass
and muscle strength. J Gerontol A Biol Sci Med Sci. 2009;64(11):1183-
1189.
35. Cosqueric G, Sebag A, Ducolombier C, Thomas C, Piette F, Weill-
Engerer S. Sarcopenia is predictive of nosocomial infection in care of
the elderly. Br J Nutr. 2006;96(5):895-901.
36. Brandt C, Pedersen BK. The role of exercise-induced myokines in
muscle homeostasis and the defense against chronic diseases. J Biomed
Biotechnol. 2010;2010:520258.
37. Shulman GI, Rothman DL, Jue T, Stein P, DeFronzo RA, Shulman
RG. Quantitation of muscle glycogen synthesis in normal subjects and
subjects with non-insulin-dependent diabetes by 13C nuclear magnetic
resonance spectroscopy. N Engl J Med. 1990;322(4):223-228.
787
38. DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J, Felber
JP. The effect of insulin on the disposal of intravenous glucose.
Results from indirect calorimetry and hepatic and femoral venous
catheterization. Diabetes. 1981;30(12):1000-1007.
39. Herridge MS, Cheung AM, Tansey CM, et al. One-year outcomes in
survivors of the acute respiratory distress syndrome. N Engl J Med.
2003;348(8):683-693.
40. Dos Santos C, Hussain SN, Mathur S, et al. Mechanisms of Chronic
Muscle Wasting and Dysfunction after an Intensive Care Unit Stay. A
Pilot Study. Am J Respir Crit Care Med. 2016;194(7):821-830.
41. Glass DJ. Signaling pathways perturbing muscle mass. Curr Opin Clin
Nutr Metab Care. 2010;13(3):225-229.
42. Rubinson L, Diette GB, Song X, Brower RG, Krishnan JA. Low
caloric intake is associated with nosocomial bloodstream infections
in patients in the medical intensive care unit. Crit Care Med.
2004;32(2):350-357.
43. Heyland DK, Schroter-Noppe D, Drover JW, et al. Nutrition
support in the critical care setting: current practice in canadian
ICUs–opportunities for improvement? JPEN J Parenter Enteral Nutr.
2003;27(1):74-83.
44. Drummond MJ, Bell JA, Fujita S, et al. Amino Acids are Necessary
for the Insulin-Induced Activation of mTOR/S6K1 Signaling and
Protein Synthesis in Healthy and Insulin Resistant Human Skeletal
Muscle. Clin Nutr. 2008;27(3):447-456.
45. Morton RW, Traylor DA, Weijs PJM, Phillips SM. Defining anabolic
resistance: implications for delivery of clinical care nutrition. Curr
Opin Crit Care. 2018;24(2):124-130.
46. Rasmussen BB, Fujita S, Wolfe RR, et al. Insulin resistance of muscle
protein metabolism in aging. Faseb j. 2006;20(6):768-769.
47. Looijaard WG, Dekker IM, Stapel SN, et al. Skeletal muscle quality
as assessed by CT-derived skeletal muscle density is associated with
6-month mortality in mechanically ventilated critically ill patients.
Crit Care. 2016;20(1):386.
48. Z’Graggen WJ, Lin CS, Howard RS, Beale RJ, Bostock H. Nerve
excitability changes in critical illness polyneuropathy. Brain. 2006;129
(Pt 9):2461-2470.
49. Teener JW, Rich MM. Dysregulation of sodium channel gating in
critical illness myopathy. J Muscle Res Cell Motil. 2006;27(5-7):291-
296.
50. Novak KR, Nardelli P, Cope TC, et al. Inactivation of sodium
channels underlies reversible neuropathy during critical illness in rats.
J Clin Invest. 2009;119(5):1150-1158.
51. Koch S, Bierbrauer J, Haas K, et al. Critical illness polyneuropathy in
ICU patients is related to reduced motor nerve excitability caused by
reduced sodium permeability. Intensive Care Medicine Experimental.
2016;4:10.
52. Sheffler LR, Chae J. Neuromuscular electrical stimulation in neurore-
habilitation. Muscle Nerve. 2007;35(5):562-590.
53. Edstrom L, Grimby L. Effect of exercise on the motor unit. Muscle
Nerve. 1986;9(2):104-126.
54. Maffiuletti NA. Physiological and methodological considerations for
the use of neuromuscular electrical stimulation. Eur J Appl Physiol.
2010;110(2):223-234.
55. Maffiuletti NA, Gondin J, Place N, Stevens-Lapsley J, Vivodtzev I,
Minetto MA. Clinical Use of Neuromuscular Electrical Stimulation
for Neuromuscular Rehabilitation: What Are We Overlooking? Arch
Phys Med Rehabil. 2018;99(4):806-812.
56. Parry SM, Berney S, Warrillow S, et al. Functional electrical stimula-
tion with cycling in the critically ill: a pilot case-matched control study.
J Crit Care. 2014;29(4):695.e1-7.
57. Decker MJ, Griffin L, Abraham LD, Brandt L. Alternating stim-
ulation of synergistic muscles during functional electrical stimula-
tion cycling improves endurance in persons with spinal cord injury.
J Electromyogr Kinesiol. 2010;20(6):1163-1169.
788
58. Maffiuletti NA, Roig M, Karatzanos E, Nanas S. Neuromuscular
electrical stimulation for preventing skeletal-muscle weakness and
wasting in critically ill patients: a systematic review. BMC Med.
2013;11:137.
59. Williams N, Flynn M. A review of the efficacy of neuromuscular
electrical stimulation in critically ill patients. Physiother Theory Pract.
2014;30(1):6-11.
60. Dall’ Acqua AM, Sachetti A, Santos LJ, et al. Use of neuromuscular
electrical stimulation to preserve the thickness of abdominal and chest
muscles of critically ill patients: A randomized clinical trial. J Rehabil
Med. 2017;49(1):40-48.
61. Routsi C, Gerovasili V, Vasileiadis I, et al. Electrical muscle stim-
ulation prevents critical illness polyneuromyopathy: a randomized
parallel intervention trial. Crit Care. 2010;14(2):R74.
62. Gruther W, Kainberger F, Fialka-Moser V, et al. Effects of neu-
romuscular electrical stimulation on muscle layer thickness of knee
extensor muscles in intensive care unit patients: a pilot study. J Rehabil
Med. 2010;42(6):593-597.
63. Rodriguez PO, Setten M, Maskin LP, et al. Muscle weakness in
septic patients requiring mechanical ventilation: protective effect of
transcutaneous neuromuscular electrical stimulation. J Crit Care.
2012;27(3):319.e1-8.
64. Poulsen JB, Moller K, Jensen CV, Weisdorf S, Kehlet H, Perner
A. Effect of transcutaneous electrical muscle stimulation on muscle
volume in patients with septic shock. Crit Care Med. 2011;39(3):456-
461.
65. Dirks ML, Hansen D, Van Assche A, Dendale P, Van Loon LJ.
Neuromuscular electrical stimulation prevents muscle wasting in
critically ill comatose patients. Clin Sci (Lond). 2015;128(6):357-365.
66. Kho ME, Truong AD, Zanni JM, et al. Neuromuscular electrical
stimulation in mechanically ventilated patients: a randomized, sham-
controlled pilot trial with blinded outcome assessment. J Crit Care.
2015;30(1):32-39.
67. Fischer A, Spiegl M, Altmann K, et al. Muscle mass, strength
and functional outcomes in critically ill patients after cardiothoracic
surgery: does neuromuscular electrical stimulation help? The Catastim
2 randomized controlled trial. Crit Care. 2016;20:30.
68. Medrinal C, Combret Y, Prieur G, et al. Comparison of exercise
intensity during four early rehabilitation techniques in sedated and
ventilated patients in ICU: a randomised cross-over trial. Crit Care.
2018;22(1):110.
69. Fossat G, Baudin F, Courtes L, et al. Effect of In-Bed Leg Cycling
and Electrical Stimulation of the Quadriceps on Global Muscle
Strength in Critically Ill Adults: A Randomized Clinical Trial. JAMA.
2018;320(4):368-378.
70. Quittan M. Aspects of physical medicine and rehabilitation in the
treatment of deconditioned patients in the acute care setting: the role
of skeletal muscle. Wien Med Wochenschr. 2016;166(1-2):28-38.
71. Callaghan M, Nouri B. Performance Of Thermal-Enhanced Electrical
Muscle Stimulation In Challenging Subjects: A Randomized Controlled
Trial, in C29. MUSCLES, EXERCISE ASSESSMENT, AND RE-
HABILITATION. p. A6664-A6664.
72. Hurt RT, McClave SA, Martindale RG, et al. Summary Points
and Consensus Recommendations From the International Protein
Summit. Nutr Clin Pract. 2017;32(1 suppl):142s-151s.
73. Heyland DK, Weijs PJ, Coss-Bu JA, et al. Protein Delivery in
the Intensive Care Unit: Optimal or Suboptimal? Nutr Clin Pract.
2017;31(1):58S-71S.
74. Hoffer LJ. Protein requirement in critical illness. Appl Physiol Nutr
Metab. 2016;41(5):573-576.
75. Gunst J, Vanhorebeek I, Thiessen SE, Van den Berghe G.
Amino acid supplements in critically ill patients. 2018;130:127-131.
https://www.ncbi.nlm.nih.gov/pubmed/29223645.
Nutrition in Clinical Practice 33(6)
76. Weijs PJ. Protein delivery in critical illness. Curr Opin Crit Care.
2016;22(4):299-302.
77. Hoffer LJ, Bistrian BR. Appropriate protein provision in criti-
cal illness: a systematic and narrative review. Am J Clin Nutr.
2012;96(3):591-600.
78. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the
Provision and Assessment of Nutrition Support Therapy in the Adult
Critically Ill Patient: Society of Critical Care Medicine (SCCM) and
American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.).
JPEN J Parenter Enteral Nutr. 2016;40(2):159-211.
79. Singer P, Berger MM, Van den Berghe G, et al. ESPEN Guidelines on
Parenteral Nutrition: intensive care. Clin Nutr. 2009;28(4):387-400.
80. Plank LD, Hill GL. Sequential metabolic changes following induction
of systemic inflammatory response in patients with severe sepsis or
major blunt trauma. World J Surg. 2000;24(6):630-638.
81. Rooyackers O, Kouchek-Zadeh R, Tjader I, Norberg A, Klaude M,
Wernerman J. Whole body protein turnover in critically ill patients
with multiple organ failure. Clin Nutr. 2015;34(1):95-100.
82. Liebau F, Sundstrom M, van Loon LJ, Wernerman J, Rooyackers O.
Short-term amino acid infusion improves protein balance in critically
ill patients. Crit Care. 2015;19:106.
83. Liebau F, Wernerman J, van Loon LJ, Rooyackers O. Effect of
initiating enteral protein feeding on whole-body protein turnover in
critically ill patients. Am J Clin Nutr. 2015;101(3):549-557.
84. Sundstrom Rehal M, Liebau F, Tjader I, Norberg A, Rooyackers
O, Wernerman J. A supplemental intravenous amino acid infusion
sustains a positive protein balance for 24 hours in critically ill patients.
Crit Care. 2017;21(1):298.
85. Martindale RG, Heyland DK, Rugeles SJ, et al. Protein Kinetics and
Metabolic Effects Related to Disease States in the Intensive Care Unit.
Nutr Clin Pract. 2017;32(1 suppl):21s-29s.
86. Nicolo M, Heyland DK, Chittams J, Sammarco T, Compher C.
Clinical Outcomes Related to Protein Delivery in a Critically Ill
Population: A Multicenter, Multinational Observation Study. JPEN
J Parenter Enteral Nutr. 2016;40(1):45-51.
87. Allingstrup MJ, Esmailzadeh N, Wilkens Knudsen A, et al. Provision
of protein and energy in relation to measured requirements in inten-
sive care patients. Clin Nutr. 2012;31(4):462-468.
88. Weijs PJ, Looijaard WG, Beishuizen A, Girbes AR, Oudemans-van
Straaten HM. Early high protein intake is associated with low mortal-
ity and energy overfeeding with high mortality in non-septic mechan-
ically ventilated critically ill patients. Crit Care. 2014;18(6):701.
89. Ferrie S, Allman-Farinelli M, Daley M, Smith K. Protein Require-
ments in the Critically Ill: A Randomized Controlled Trial Using
Parenteral Nutrition. JPEN J Parenter Enteral Nutr. 2016;40(6):795-
805.
90. Fetterplace K, Deane AM, Tierney A, et al. Targeted Full Energy
and Protein Delivery in Critically Ill Patients: A Pilot Randomized
Controlled Trial (FEED Trial). JPEN J Parenter Enteral Nutr. 2018.
91. Gamrin-Gripenberg L, Sundstrom-Rehal M, Olsson D, Grip J, Wern-
erman J, Rooyackers O. An attenuated rate of leg muscle protein
depletion and leg free amino acid efflux over time is seen in ICU long-
stayers. Crit Care. 2018;22(1):13.
92. Morton RW, Murphy KT, McKellar SR, et al. A systematic review,
meta-analysis and meta-regression of the effect of protein supple-
mentation on resistance training-induced gains in muscle mass and
strength in healthy adults. Br J Sports Med. 2018;52(6):376-384.
93. Heyland DK, Stapleton R, Compher C. Should We Prescribe More
Protein to Critically Ill Patients? Nutrients. 2018;10(4).
94. Pennings B, Koopman R, Beelen M, Senden JM, Saris WH, van Loon
LJ. Exercising before protein intake allows for greater use of dietary
protein-derived amino acids for de novo muscle protein synthesis in
both young and elderly men. Am J Clin Nutr. 2011;93(2):322-331.
Parry et al
95. Biolo G, Tipton KD, Klein S, Wolfe RR. An abundant supply of
amino acids enhances the metabolic effect of exercise on muscle
protein. Am J Physiol. 1997;273(1 Pt 1):E122-E129.
96. Timmerman KL, Dhanani S, Glynn EL, et al. A moderate acute
increase in physical activity enhances nutritive flow and the muscle
protein anabolic response to mixed nutrient intake in older adults. Am
J Clin Nutr. 2012;95(6):1403-1412.
97. Trappe TA, Burd NA, Louis ES, Lee GA, Trappe SW. Influence of
concurrent exercise or nutrition countermeasures on thigh and calf
muscle size and function during 60 days of bed rest in women. Acta
Physiol (Oxf). 2007;191(2):147-159.
98. Arabi YM, Casaer MP, Chapman M, et al. The intensive care
medicine research agenda in nutrition and metabolism. Intensive Care
Med. 2017;43(9):1239-1256.
99. Wappel S, Ali O, Serra M, et al. The effect of an exercise, nutrition
and neuromuscular electrical stimulation intervention on acute muscle
wasting in critically ill patients using mechanical ventilation. AJR-
CCM. 2017;195(A2747).
100. Chapman M, Fraser R, Vozzo R, et al. Antro-pyloro-duodenal motor
responses to gastric and duodenal nutrient in critically ill patients. Gut.
2005;54(10):1384-1390.
101. Ali Abdelhamid Y, Cousins CE, Sim JA, et al. Effect of Critical
Illness on Triglyceride Absorption. JPEN J Parenter Enteral Nutr.
2015;39(8):966-972.
102. Chapman MJ, Fraser RJ, Matthews G, et al. Glucose absorption and
gastric emptying in critical illness. Crit Care. 2009;13(4):R140.
103. Boirie Y, Gachon P, Beaufrere B. Splanchnic and whole-body leucine
kinetics in young and elderly men. Am J Clin Nutr. 1997;65(2):489-
495.
104. Jonker R, Engelen MP, Deutz NE. Role of specific dietary amino acids
in clinical conditions. Br J Nutr. 2012;108 (Suppl 2):S139-S148.
105. Chapple LS, Deane AM, Heyland DK, et al. Energy and protein
deficits throughout hospitalization in patients admitted with a trau-
matic brain injury. Clin Nutr. 2016;35(6):1315-1322.
106. Ridley EJ, Parke RL, Davies AR, et al. What Happens to Nutrition
Intake in the Post-Intensive Care Unit Hospitalization Period? An
Observational Cohort Study in Critically Ill Adults. JPEN J Parenter
Enteral Nutr. 2018.
107. Bear DE, Hart N, Puthucheary Z. Continuous or intermittent feeding:
pros and cons. Curr Opin Crit Care. 2018;24(4):256-261.
108. Black C, Singer M, Grocott M. The oxygen cost of rehabilitation in
mechanically ventilated patients. AJRCCM. 2017;195(A2742).
109. Garros RF, Paul R, Connolly M, et al. MicroRNA-542 Promotes
Mitochondrial Dysfunction and SMAD Activity and Is Elevated in
Intensive Care Unit-acquired Weakness. Am J Respir Crit Care Med.
2017;196(11):1422-1433.
789
110. Bouletreau P, Patricot MC, Saudin F, Guiraud M, Mathian B.
Effects of intermittent electrical stimulations on muscle catabolism in
intensive care patients. JPEN J Parenter Enteral Nutr. 1987;11(6):552-
555.
111. Gerovasili V, Stefanidis K, Vitzilaios K, et al. Electrical muscle
stimulation preserves the muscle mass of critically ill patients: a
randomized study. Crit Care. 2009;13(5):R161.
112. Karatzanos E, Gerovasili V, Zervakis D, et al. Electrical Muscle
Stimulation: An Effective Form of Exercise and Early Mobilization
to Preserve Muscle Strength in Critically Ill Patients. Critical Care
Research and Practice. 2012;2012:432752.
113. Meesen RL, Dendale P, Cuypers K, et al. Neuromuscular electrical
stimulation as a possible means to prevent muscle tissue wasting
in artificially ventilated and sedated patients in the intensive care
unit: A pilot study. Neuromodulation. 2010;13(4):315-320; discussion
321.
114. Angelopoulos E, Karatzanos E, Dimopoulos S, et al. Acute mi-
crocirculatory effects of medium frequency versus high frequency
neuromuscular electrical stimulation in critically ill patients – a pilot
study. Ann Intensive Care. 2013;3(1):39.
115. Hirose T, Shiozaki T, Shimizu K, et al. The effect of electrical muscle
stimulation on the prevention of disuse muscle atrophy in patients
with consciousness disturbance in the intensive care unit. J Crit Care.
2013;28(4):536.e1-7.
116. Falavigna L, Silva MG, AL F, et al. Effects of electrical muscle
stimulation early in the quadriceps and tibialis anterior muscle
of critically ill patients. Phyiother Theory Pract. 2014;30(4):223-
228.
117. Segers J, Hermans G, Bruyninckx F, Meyfroidt G, Langer D,
Gosselink R. Feasibility of neuromuscular electrical stimulation in
critically ill patients. J Crit Care. 2014;29(6):1082-1088.
118. Akar O, Gunay E, Sarinc Ulasli S, et al. Efficacy of neuromuscular
electrical stimulation in patients with COPD followed in intensive care
unit. Clin Respir J. 2017;11(6):743-750.
119. Patsaki I, Gerovasili V, Sidiras G, et al. Effect of neuromuscular
stimulation and individualized rehabilitation on muscle strength in
Intensive Care Unit survivors: A randomized trial. J Crit Care.
2017;40:76-82.
120. Silva PE, Babault N, Mazullo JB, et al. Safety and feasibility of
a neuromuscular electrical stimulation chronaxie-based protocol in
critical ill patients: A prospective observational study. J Crit Care.
2017;37:141-148.
121. Woo K, Kim J, Kim HB, et al. The Effect of Electrical Muscle
Stimulation and In-bed Cycling on Muscle Strength and Mass of
Mechanically Ventilated Patients: A Pilot Study. Acute Crit Care.
2018;33(1):16-22.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Methods of Enteral Nutrition Administration in Critically Ill December 2018 790–795

C 2018 American Society for

Patients: Continuous, Cyclic, Intermittent, and Bolus Feeding Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10105
wileyonlinelibrary.com

Satomi Ichimaru, PhD, RD, CNSC

Abstract
There are several methods of enteral nutrition (EN) administration, including continuous, cyclic, intermittent, and bolus techniques,
which can be used either alone or in combination. Continuous feeding involves hourly administration of EN over 24 hours assisted
by a feeding pump; cyclic feeding involves administration of EN over a time period of <24 hours generally assisted by a feeding
pump; intermittent feeding involves administration of EN over 20–60 minutes every 4–6 hours via pump assist or gravity assist;
and bolus feeding involves administration of EN over a 4- to 10-minute period using a syringe or gravity drip. In practice, pump-
assisted continuous feeding is generally acceptable for critically ill patients to prevent EN-related complications. However, a limited
number of studies have been conducted to support this practice. In addition, regarding muscle protein synthesis and gastrointestinal
hormone secretion, intermittent or bolus feeding may be more beneficial than continuous EN feeding for critically ill patients.
For medically stable patients with feeding tubes terminating in the stomach, bolus feeding is favored with respect to practical
factors, such as cost, convenience, and patient mobility. However, few studies have shown whether intermittent or bolus feeding is
beneficial in a critical care setting at present. Additional randomized controlled studies comparing intermittent with bolus feeding
are required. (Nutr Clin Pract. 2018;33:790–795)

Keywords
cost and cost analysis; critical illness; enteral nutrition; nutrition support; patient safety; protein synthesis; respiratory aspiration;
tube feeding

Introduction Continuous Feeding

Enteral nutrition (EN) is defined as nutrition provided
through the gastrointestinal (GI) tract via a tube, catheter,
or stoma that delivers nutrients distal to the oral cavity.1
The modes of EN administration include continuous, cyclic,
intermittent, and bolus techniques, which can be used either
alone or in combination. To determine the most appropriate
method of administration, the clinician should consider
numerous factors such as patient’s age, preexisting and
current medical condition, nutrition status and requirement,
enteral route for feeding (gastric vs small bowel), GI tol-
erance, formula type used, patient mobility, feeding pump
availability, and cost. Continuous feeding seems to be the
standard in the intensive care unit (ICU)2 ; however, there
is currently insufficient data to choose the best method
to improve patient outcomes of critically ill patients. This
narrative review summarizes EN administration methods,
their advantages and disadvantages, and the indications of
intermittent and/or bolus feeding in critically ill patients.
Methods of EN Administration
The 4 methods of EN administration include continu-
ous, cyclic, intermittent, and bolus feedings as shown in
Figure 1.3
Continuous feeding provides EN by electric enteral feeding
pump over 24 hours, which is generally initiated at a rate of
20–50 mL/h and advanced to goal rate by 10–25 mL/h every
4–24 hours.4 This method is selected for patients who are
critically ill, who have been intubated for respiratory failure,
who are fed through a postpyloric tube, or who cannot
tolerate intermittent or bolus feedings.5
Although continuous feeding is preferred by most ICUs,
it is supported by only a few relatively outdated studies.
A study on 76 adult burn patients reported that patients
who received continuous feeding had lower stool frequency
From the Department of Nutrition Management, Osaka Saiseikai
Nakatsu Hospital, Osaka, Japan.
This article was modified on 2018-06-25 after initial online publication
to correct the article title and to make some other editing changes.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on June 20, 2018.
Corresponding Author:
Satomi Ichimaru, PhD, RD, CNSC, Department of Nutrition
Management, Osaka Saiseikai Nakatsu Hospital, 2-10-39 Shibata
Kitaku, Osaka 530-0012, Japan.
Email: satomi.ichimaru@gmail.com
Ichimaru
Figure 1. Methods of delivering enteral tube feeding. (Reprinted by permission from Springer Nature: Diet and Nutrition in
Critical Care, Intermittent and Bolus Methods of Feeding in Critical Care, Ichimaru S and Amagai T, Springer Science+Business
Media New York 2015)
and reached their nutrition goals sooner than those who
received bolus feeding.6 In adult patients with neurological
impairment who received continuous feeding, aspiration
was observed less frequently (1/17) than in those who
received intermittent feeding (3/17).7 In a study of 18
postoperative patients, continuous feeding groups showed
improved cumulative nitrogen balance for 5 days compared
with cyclic feeding groups.8 Whether continuous feeding
provides better glycemic control than intermittent/bolus
feeding remains controversial.9-11
Continuous feeding is frequently interrupted in the
ICU.12 The most common reasons for discontinuation of
feeding include surgery (27%) and diagnostic procedures
(15%), and minor reasons for EN interruptions include
mechanical feeding tube problems (8%), pharmacy delivery
delay (4%), and miscellaneous factors (3%). Because of the
frequency with which enteral feeds are interrupted, resulting
in delivery of only 50%–60% of prescribed EN volume on a
daily basis,13 some institutions have implemented a volume-
based feeding protocol to ensure that the volume of EN
prescribed to their patients is actually provided.14-16
Cyclic Feeding
Cyclic feeding involves feeding by electric enteral feeding
pump over a period of <24 hours, in which the goal infusion
rate is determined by dividing the desired formula volume
by the number of hours of administration. The infusion
time may vary between 24 and 8 h/day depending on the
patient’s volume tolerance. This method can be used for
patients with feeding tubes terminating in the stomach or
small bowel. During the course of recovery, patients may
transition from the continuous feeding to nocturnal cyclic
feeding to stimulate patient’s appetite during the day. It also
increases patients’ mobility by freeing them from a feeding
set or pump.5
791
Bonten et al17 and Tamowicz et al18 hypothesized
that cyclic feeding beneficially affected gastric acidity and
bacterial colonization of the stomach and respiratory tract,
thereby helping to prevent ventilator-associated pneumonia
(VAP). In these studies, a significant decrease in gastric
pH was observed when EN was discontinued in the cyclic
feeding groups. However, only the study by Tamowicz
et al18 showed reduced rate of gastric colonization, and the
incidence of VAP was relatively similar between the cyclic
and continuous feeding groups in both studies.
With regard to mortality in the ICU, the cyclic and con-
tinuous feeding groups showed no significant differences.17
In 1 study, the length of hospital stay was found to be
significantly shorter in patients receiving cyclic feeding than
in those receiving continuous feeding.19
Intermittent Feeding
Intermittent EN feeding is usually delivered over 20–60
minutes by infusion pump or by the gravity drip method. In
the gravity drip method, formula flows out from a feeding
bag and into a feeding tube by gravity. The rate of infusion
is regulated by adjusting a roller clamp.5 Usually, the gravity
drip feeding is tolerated when infused into the stomach.1 In
intermittent feeding, a volume of 240–720 mL of feeding is
administered 4–6 times per day depending on the patient’s
nutrition needs.5 This feeding method is more physiological
than continuous/cyclic feeding because it permits greater
patient mobility between feedings. If tolerated, the volume
of each feeding can be increased and the total number of
feeds can be decreased to improve quality of life.5 According
to old studies, intermittent feeding has been believed to
have some disadvantages, such as risk for aspiration7 and
diarrhea20 ; however, in a recent study of ICU patients
receiving intermittent or continuous feeding, there was no
difference in outcomes, including the rate of aspiration
792
and diarrhea.21 In 1 study, intermittent and continuous
feeding groups of critically ill trauma patients showed no
significant differences in mortality in the ICU or incidence
of pneumonia.22 In 2 studies that compared intermittent
and 16-hour cyclic feeding in elderly patients, no significant
differences were noted between the groups in mortality,
incidence of diarrhea, or pneumonia.23,24
Bolus Feeding
Bolus feeding is administered via syringe or gravity drip
over a short period, usually 4–10 minutes. Generally, the
patient is fed a volume of 240 mL of feeding 3–6 times
daily.5 Feeding provided by this rapid infusion method may
result in diarrhea and/or aspiration.5,6,25 Therefore, bolus
feeding is usually reserved for the medically stable patients
with feeding tubes ending in the stomach.1 For patients with
a gastrostomy tube, not only commercial liquid formulas
but also blended food or viscosity-thickened formula can
be administered in the bolus feeding.26,27 One advantage
of bolus feeding is that medication can be separately ad-
ministered from feeding. Furthermore, this method closely
resembles normal eating patterns, increases the time away
from feeding, and provides freedom of movement and a
more normal life.
To the best of our knowledge, no study has evaluated
the outcomes of mortality or length of hospital stay be-
tween patients receiving bolus and continuous feeding. The
bolus feeding group showed improved nitrogen balance in
1 study,28 whereas the continuous feeding group showed
significantly improved body weight and arm circumference
in another study.29 Respiratory quotient, resting energy
expenditure, and blood sugar were comparable between
cyclic and bolus feedings in head-injured patients with
mechanical ventilation.30
Table 1 shows the advantages, disadvantages, and indica-
tions for each method.3
Potential Advantages of Intermittent/
Bolus Feeding: Muscle Protein Synthesis
and GI Hormone Secretion
In ICU patients, muscle breakdown often exceeds muscle
synthesis, and skeletal muscle wasting leads to functional
impairment in most survivors of critical illness.31-33 Insulin
and leucine enhance muscle protein synthesis (MPS) by
activating AKT/protein kinase B and the mammalian target
of rapamycin (mTOR).34-36 The continuous infusion of
amino acids for 30–60 minutes has been found to lead
to stimulated MPS; however, after 120 minutes, MPS was
found to have declined to baseline despite continuation
of the amino acid infusion.37 An oral whey protein bolus
has been found to cause a pulsatile release of insulin, a
pulsatile increase in intramuscular leucine concentration,
Nutrition in Clinical Practice 33(6)
and an increase in the rate of MPS. MPS has been found
to peak at 90 minutes, decrease thereafter, and return to
baseline at 180 minutes.38 In a study using neonatal pigs,
significant increases in intramuscular AKT and mTOR, as
well as increased MPS, were observed with bolus compared
with continuous feeding.39
After the intake of a meal, several hormones, such as
glucagon-like peptide-1, gastric inhibitory peptide, chole-
cystokinin, ghrelin, and peptide YY, are secreted from the
enteroendocrine cells lining the lumen of the GI tract.
These hormones regulate GI motility, gallbladder con-
traction, pancreatic function, and nutrient absorption40 ;
most of these hormones are secreted within minutes of
feeding, rise transiently, and decline to basal levels after
feeding. In continuous tube feeding, this enterohormonal
response to nutrition is almost completely absent.41-44
Furthermore, bolus feeding may stimulate small-intestinal
growth. In a study using neonatal pigs, greater small-
intestinal mucosal weight and ileal protein mass have been
observed fed via bolus feeding compared with continuous
feeding.45
Although the studies introduced earlier are conducted
in non-ICU settings, we cannot deny the possibility that
those results may be applicable to ICU patients. Thus,
regarding MPS and GI hormone secretion, intermittent
or bolus feeding may be more beneficial than continuous
EN feeding for critically ill patients. Further randomized
controlled studies comparing continuous with intermittent
and bolus feeding are required to confirm this hypotheses.
Intermittent vs Bolus Feeding
Several studies have been conducted to compare the out-
comes of feeding methods in the ICU; however, to our
knowledge, no study has compared intermittent with bolus
feeding. One reason for this could be that pump-assisted
continuous feeding is generally considered acceptable for
patients in the ICU to prevent EN intolerance.2,46 Another
reason could be that intermittent and bolus feeding are often
treated as the same feeding method because of the lack of a
clear definition of them.
Currently, no clinical guidelines strongly recommend
a specific method of feeding for either critically ill or
stable patients. According to the Canadian Critical Care
Nutrition Guidelines in 2015, there are insufficient data to
make recommendations on whether EN should be given
continuously or via other methods in critically ill patients.47
The American guidelines of Nutrition Support Therapy
in the Adult Critically Ill Patient in 2016 suggest that
for high-risk patients or those intolerant to bolus gastric
EN, the delivery of EN should be switched to continuous
infusion.2 The National Institute for Health and Care
Excellence Clinical Guidelines recommend that EN should
usually be continuously administered over 16–24 hours daily
Ichimaru 793

Table 1. Advantages, Disadvantages, and Indications of Each Feeding Method.

Feeding Method Advantages Disadvantages Indications

Continuous May improve tolerance Feeding pump required Initiation of feeding in

May reduce risk of aspiration May restrict ambulation critically ill patients
Increased time for nutrient absorption More expensive Promote tolerance
Compromised gastric
function
Feeding into small bowel
Intolerance to other
feeding methods
Cyclic Facilitates transition of support to oral diet Feeding pump required Transitioning from EN to
Allows daytime ambulation May require high oral nutrition (enhance
Encourages patient to eat normal meals and infusion rates appetite during the day)
snacks May promote intolerance Supplement inadequate
oral intake
Free patient from enteral
feedings during the day
Intermittent Feeding pump may not be required Increased risk for Intolerance to bolus
May enhance quality of life aspiration administration
Allows greater mobility between feedings Gastric distention Initiation of EN without
More physiological Delayed gastric emptying feeding pump
May be better tolerated than bolus feeding
Bolus More physiological Increased risk of Recommended for gastric
Feeding pump not required aspiration feeding
Inexpensive and easy administration Hypertonic, high-fat, or Normal gastric function
Limits feeding time high-fiber formulas
Patient is free to move about, participate in may delay gastric
rehabilitation therapies, and live a emptying or result in
relatively normal life osmotic diarrhea
More likely patient will receive all of formula

EN, enteral nutrition.

Reprinted by permission from Springer Nature: Diet and Nutrition in Critical Care, Intermittent and Bolus Methods of Feeding in Critical Care,
Ichimaru S and Amagai T, Springer Science+Business Media New York 2015.

for patients in the ICU, and if insulin administration is Summary

required, continuous feeding over 24 hours is safe and more
practical.48 The major points of this review are summarized as follows:

r There are 4 methods of EN administration: continu-

Application of Intermittent and/or Bolus
Feeding to Other Conditions
In the administration of EN at a lower acuity hospital
unit, a long-term care facility, or the home, it is important
to address practical factors, such as cost, convenience,
and patient mobility. Compared with other methods, bolus
feeding is inexpensive because it requires less equipment
such as feeding bags, administration sets, and pumps. In
addition, bolus feeding requires the least amount of time,
and it is easy to perform for caregivers at home. In patients
who frequently tug and dislodge nasogastric tubes, bolus
feeding may be safer than intermittent feeding. Especially
for alert and active patients, bolus feeding can allow them to
live a relatively normal life, because of shorter feeding time
and freedom from mechanical devices between feedings.
ous, cyclic, intermittent, and bolus feedings.
r At present, there is no evidence that suggests that any
particular feeding method is superior to others.
r In the ICU, pump-assisted continuous feeding is
generally acceptable to prevent EN-related complica-
tions.
r Regarding MPS and GI hormone secretion, inter-
mittent or bolus, rather than continuous, EN feeding
may be favorable even for critically ill patients.
r In a lower acuity hospital unit, a long-term care fa-
cility, or the home, bolus feeding is preferred because
it is inexpensive, easy to perform, requires the least
amount of time, and mimics normal eating patterns.
r Further research is required to determine whether in-
termittent or bolus feeding is beneficial for critically
ill patients.
794 Nutrition in Clinical Practice 33(6)

References patients treated under conditions of intensive therapy. Adv Clin

Exp. 2007;16:365-373.
1. Bankhead R, Boullata J, Brantley S, et al. Enteral nutrition practice
19. van Berge Henegouwen MI, Akkermans LM, van Gulik TM, et al.
recommendations. JPEN J Parenter Enteral Nutr. 2009;33(2):122-167.
Prospective, randomized trial on the effect of cyclic versus continu-
2. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the
ous enteral nutrition on postoperative gastric function after pylorus-
provision and assessment of nutrition support therapy in the adult
preserving pancreatoduodenectomy. Ann Surg. 1997;226:677-685.
critically ill patient: Society of Critical Care Medicine (SCCM) and
20. Ciocon JO, Galindo-Ciocon DJ, Tiessen C, Galindo D. Continuous
American Society for Parenteral and Enteral Nutrition (A.S.P.E.N.).
compared with intermittent tube feeding in the elderly. JPEN J Parenter
JPEN J Parenter Enteral Nutr. 2016;40:159-211.
Enteral Nutr. 1992;16:525-528.
3. Ichimaru S, Amagai T. Intermittent and bolus methods of feeding in
21. Serpa LF, Kimura M, Faintuch J, Ceconello I. Effects of continuous
critical care. In: Rajendram R, Preedy VR, Patel VB, eds. Diet and
versus bolus infusion of enteral nutrition in critical patients. Rev Hosp
Nutrition in Critical Care. New York: Springer; 2014:533-548.
Clin Fac Med Sao Paulo. 2003;58:9-14.
4. Parrish CR. Enteral feeding: the art and the science. Nutr Clin Pract.
22. MacLeod JB, Lefton J, Houghton D, et al. Prospective randomized
2003;18:76-85.
control trial of intermittent versus continuous gastric feeds for critically
5. Brantley SL, Mills ME. Overview of enteral nutrition. In: Mueller CM,
ill trauma patients. J Trauma. 2007;63:57-61.
ed. The A.S.P.E.N. Nutrition Support Core Curriculum. 2nd ed. Silver
23. Lee JS, Auyeung TW. A comparison of two feeding methods in
Spring, MD: American Society for Parenteral and Enteral Nutrition;
the alleviation of diarrhoea in older tube-fed patients: a randomised
2012:170-184.
controlled trial. Age Ageing. 2003;32:388-393.
6. Hiebert JM, Brown A, Anderson RG, Halfacre S, Rodeheaver GT,
24. Lee JS, Kwok T, Chui PY, et al. Can continuous pump feeding
Edlich RF. Comparison of continuous vs intermittent tube feedings
reduce the incidence of pneumonia in nasogastric tube-fed patients? A
in adult burn patients. JPEN J Parenter Enteral Nutr. 1981;5:73-75.
randomized controlled trial. Clin Nutr. 2010;29:453-458.
7. Kocan MJ, Hickisch SM. A comparison of continuous and intermit-
25. Steevens EC, Lipscomb AF, Poole GV, Sacks GS. Comparison of con-
tent enteral nutrition in NICU patients. J Neurosci Nurs. 1986;18:333-
tinuous vs intermittent nasogastric enteral feeding in trauma patients:
337.
perceptions and practice. Nutr Clin Pract. 2002;17:118-122.
8. Campbell IT, Morton RP, Macdonald IA, Judd S, Shapiro L, Stell
26. Epp L, Lammert L, Vallumsetla N, Hurt RT, Mundi MS. Use of
PM. Comparison of the metabolic effects of continuous postoperative
blenderized tube feeding in adult and pediatric home enteral nutrition
enteral feeding and feeding at night only. Am J Clin Nutr. 1990;52:1107-
patients. Nutr Clin Pract. 2017;32:201-205.
1112.
27. Ichimaru S, Amagai T, Wakita M, Shiro Y. Which is more effective
9. Sanz Parı́s A, Lázaro J, Guallar A, Gracia P, Caverni A, Albero R.
to prevent enteral nutrition-related complications, high- or medium-
Continuous enteral nutrition versus single bolus: effects on urine C
viscosity thickened enteral formula in patients with percutaneous
peptide and nitrogen balance. Med Clin (Barc). 2005;124:613-615.
endoscopic gastrostomy? A single-center retrospective chart review.
10. Shahriari M, Rezaei E, Bakht LA, Abbasi S. Comparison of the effects
Nutr Clin Pract. 2012;27:545-552.
of enteral feeding through the bolus and continuous methods on blood
28. Campbell IT, Morton RP, Cole JA, Raine CH, Shapiro LM, Stell PM.
sugar and prealbumin levels in ICU inpatients. J Educ Health Promot.
A comparison of the effects of intermittent and continuous nasogastric
2015;4:95.
feeding on the oxygen consumption and nitrogen balance of patients
11. Evans DC, Forbes R, Jones C, et al. Continuous versus bolus tube
after major head and neck surgery. Am J Clin Nutr. 1983;38:870-878.
feeds: does the modality affect glycemic variability, tube feeding
29. Pichard C, Roulet M. Constant rate enteral nutrition in bucco-
volume, caloric intake, or insulin utilization? Int J Crit Illn Inj Sci.
pharyngeal cancer care. A highly efficient nutritional support system.
2016;6:9-15.
Clin Otolaryngol Allied Sci. 1984;9:209-214.
12. Morgan LM, Dickerson RN, Alexander KH, Brown RO, Minard G.
30. Maurya I, Pawar M, Garg R, Kaur M, Sood R. Comparison of
Factors causing interrupted delivery of enteral nutrition in trauma
respiratory quotient and resting energy expenditure in two regimens of
intensive care unit patients. Nutr Clin Pract. 2004;19:511-517.
enteral feeding-continuous vs. intermittent in head-injured critically ill
13. Heyland DK, Dhaliwal R, Wang M, Day AG. The prevalence of ia-
patients. Saudi J Anaesth. 2011;5:195-201.
trogenic underfeeding in the nutritionally “at-risk” critically ill patient:
31. Puthucheary ZA, Rawal J, McPhail M, et al. Acute skeletal muscle
results of an international, multicenter, prospective study. Clin Nutr.
wasting in critical illness. JAMA. 2013;310:1591-1600.
2015;34(4):659-666.
32. Stevens RD, Dowdy DW, Michaels RK, et al. Neuromuscular dysfunc-
14. Heyland DK, Murch L, Cahill N, et al. Enhanced protein-energy
tion acquired in critical illness: a systematic review. Intensive Care Med.
provision via the enteral route feeding protocol in critically ill patients:
2007;33:1876-1891.
results of a cluster randomized trial. Crit Care Med. 2013;41:2743-
33. Herridge MS, Tansey CM, Matté A, et al. Functional disability 5 years
2753.
after acute respiratory distress syndrome. N Engl J Med. 2011;364:1293-
15. Taylor B, Brody R, Denmark R, Southard R, Byham-Gray L. Im-
1304.
proving delivery through the adoption of the “Feed Early Enteral
34. Bolster DR, Jefferson LS, Kimball SR. Regulation of protein synthesis
Diet Adequately for Maximum Effect (FEED ME)” protocol in a
associated with skeletal muscle hypertrophy by insulin-, amino acid-
surgical trauma ICU: a quality improvement review. Nutr Clin Pract.
and exercise-induced signalling. Proc Nutr Soc. 2004;63:351-356.
2014;29:639-648.
35. Katsanos CS, Kobayashi H, Sheffield-Moore M, Aarsland A, Wolfe
16. Declercq B, Deane AM, Wang M, Chapman MJ, Heyland DK.
RR. A high proportion of leucine is required for optimal stimulation
Enhanced Protein-Energy Provision via the Enteral Route Feeding
of the rate of muscle protein synthesis by essential amino acids in the
(PEPuP) protocol in critically ill surgical patients: a multicentre
elderly. Am J Physiol Endocrinol Metab. 2006;291:E381-E387.
prospective evaluation. Anaesth Intensive Care. 2016;44:93-98.
36. Anthony JC, Yoshizawa F, Anthony TG, Vary TC, Jefferson LS,
17. Bonten MJ, Gaillard CA, van der Hulst R, et al. Intermittent enteral
Kimball SR. Leucine stimulates translation initiation in skeletal muscle
feeding: the influence on respiratory and digestive tract colonization in
of postabsorptive rats via a rapamycin-sensitive pathway. J Nutr.
mechanically ventilated intensive-care-unit patients. Am J Respir Crit
2000;130:2413-2419.
Care Med. 1996;154:394-399.
37. Bohé J, Low JF, Wolfe RR, Rennie MJ. Latency and duration of
18. Tamowicz B, Mikstacki A, Grzymislawski M. The influence of
stimulation of human muscle protein synthesis during continuous
the Feeding Therapy Model on pulmonary complications in
infusion of amino acids. J Physiol. 2001;532:575-579.
Ichimaru
38. Atherton PJ, Etheridge T, Watt PW, et al. Muscle full effect after oral
protein: time-dependent concordance and discordance between human
muscle protein synthesis and mTORC1 signaling. Am J Clin Nutr.
2010;92:1080-1088.
39. Gazzaneo MC, Suryawan A, Orellana RA, et al. Intermittent bo-
lus feeding has a greater stimulatory effect on protein synthesis in
skeletal muscle than continuous feeding in neonatal pigs. J Nutr.
2011;141:2152-2158.
40. Gutierrez-Aguilar R, Woods SC. Nutrition and L and K-
enteroendocrine cells. Curr Opin Endocrinol Diabetes Obes. 2011;18:35-
41.
41. Ledeboer M, Masclee AA, Biemond I, Lamers CB. Gallbladder motil-
ity and cholecystokinin secretion during continuous enteral nutrition.
Am J Gastroenterol. 1997;92:2274-2279.
42. Stoll B, Puiman PJ, Cui L, et al. Continuous parenteral and enteral
nutrition induces metabolic dysfunction in neonatal pigs. JPEN J
Parenter Enteral Nutr. 2012;36:538-550.
43. Chowdhury AH, Murray K, Hoad CL, et al. Effects of bolus and
continuous nasogastric feeding on gastric emptying, small bowel water
content, superior mesenteric artery blood flow, and plasma hormone
795
concentrations in healthy adults: a randomized crossover study. Ann
Surg. 2016;263:450-457.
44. Jawaheer G, Shaw NJ, Pierro A. Continuous enteral feeding impairs
gallbladder emptying in infants. J Pediatr. 2001;138:822-825.
45. Shulman RJ, Redel CA, Stathos TH. Bolus versus continuous feedings
stimulate small-intestinal growth and development in the newborn pig.
J Pediatr Gastroenterol Nutr. 1994;18:350-354.
46. Mentec H, Dupont H, Bocchetti M, Cani P, Ponche F, Bleichner G.
Upper digestive intolerance during enteral nutrition in critically ill
patients: frequency, risk factors, and complications. Crit Care Med.
2001;29:1955-1961.
47. Heyland DK, Dhaliwal R, Drover JW, et al. Continuous vs.
other methods of administration. In: Canadian Critical Care Nu-
trition Guidelines in 2015. https://www.criticalcarenutrition.com/docs/
CPGs%202015/6.3%202015.pdf. Accessed November 4, 2017.
48. National Collaborating Centre for Acute Care. Enteral tube feeding
in hospital and the community. In: Nutrition Support for Adults:
Oral Nutrition Support, Enteral Tube Feeding and Parenteral Nutri-
tion. London: National Institute for Health and Care Excellence;
2006.
Invited Review

Nutrition in Clinical Practice

Volume 33 Number 6
Technology in Parenteral Nutrition Compounding December 2018 796–802

C 2018 American Society for

Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10086
wileyonlinelibrary.com

Caitlin Curtis, PharmD, BCNSP

Abstract
Technology is constantly being used in novel ways, and its use in the practice of medicine is no exception. Examples of this
include computer physician order entry, barcode-medication scanning, electronic health records (EHRs), and bedside patient charts,
to name a few. Compounding parenteral nutrition has been included in this technological revolution, with improvements such
as barcode-assisted medication preparation systems and EHR-to-compounder interfaces. Along with some of these electronic
advancements come the inevitable improvements and challenges, which are explored in this article. (Nutr Clin Pract. 2018;33:796–
802)

Keywords
parenteral nutrition; parenteral nutrition solutions; patient safety; medication errors; drug compounding

Compounding parenteral nutrition (PN) easily lends itself refer to PN that is compounded with fat emulsion. PN are
to a discussion regarding technology, as PN remains 1 of referred to as “formula” or “formulations” unless otherwise
the most complex medications prescribed, prepared, and specified.
administered in an inpatient or ambulatory setting. As a
result of its complexity, PN appropriately remains listed as Reducing Errors
a high-risk medication by the Institute for Safe Medication
As compounding PN is a complex process with multiple
Practices (ISMP).1 With every medication, there are several
steps leading to a finished product, it is prone to errors.1,2
steps in the medication-use process in which errors can
Unfortunately, these errors can lead to severe consequences
occur. With PN, the risk of error is high because of the
for patients. One event particularly stands out, as it involves
number of components in each formulation.2 Technology
technology yet it can be solved by technology.3 In 2007, a
seeks to decrease the risk of error by accomplishing the
reported error involves a zinc overdose administered via PN.
following: (1) reducing the need for order transcription with
The prescription was for a neonatal PN, with an additive
computer-prescriber order entry (CPOE) and electronic
of zinc correctly written for the dose of 330 mcg per
health record (EHR)-to-compounder interfaces; (2) intro-
100 mL of PN formula. Although the dose was prescribed
ducing a “photo-finish” check for hand-added ingredients
in units of mcg per 100 mL, the automated compounding
with barcode-assisted medication preparation (BCMP) sys-
device (ACD) required zinc to be entered in a mcg/kg dose.
tems; (3) introducing quality-assurance checks with refrac-
The pharmacist made the adjustment; however, when she
tometry and laboratory; and (4) introducing new techniques
entered it into the ACD, she accidentally entered it as a mg
to detect micro-organisms and endotoxin. CPOE also offers
dose, not mcg dose. This resulted in a final concentration
a unique opportunity to alert prescribers to medication
of 330 mg per 100 mL, which is a 1000 times more than
interactions as well as to medication dose adjustments
was prescribed (Figure 1). Although another pharmacist
needed because of changes in weight or renal function.
These clinical decision-making tools can be applied to PN to
alert prescribers to incompatibilities, volume limits, and the Financial disclosure: None declared.
appropriate dosing of ingredients. In summary, technology Conflicts of interests: None declared.
is able to greatly improve the safety of PN prescribing and This article originally appeared online on April 17, 2018.
compounding.
Corresponding Author:
For the use of this article, PN will be differentiated into Caitlin Curtis, PharmD, BCNSP, University of Wisconsin Hospital
on the following 2 groups: 2-in-1 PN and 3-in-1 PN. The and Clinics, Nutrition Support Pharmacist, Department of
terms 2-in-1 PN or 2-in-1s refer to PN that is compounded Pharmacy, 600 Highland Avenue, Madison, WI 53792, USA.
without fat emulsion, and the terms 3-in-1 PN or 3-in-1s Email: ccurtis@uwhealth.org
Curtis
Figure 1. Intended dose was 330 mcg per 100 mL or 0.33 mg
per 100 mL. The actual dose was 330,000 mcg per 100 mL or
330 mg per 100 mL.
checked the work, she missed the entry error. The pharmacy
technician mixing PN with the ACD needed to refill the zinc
syringe on the compounder—for a total of 11 times. A third
pharmacist checked the PN solution, not knowing about
the 11 syringe refills. The PN infusion was started on the
neonate. When an oncoming technician arrived to work, the
previous technician discussed the odd situation of needing
to change the zinc syringe so many times. This technician
quickly realized the error, and alerted the pharmacist, who
had the PN stopped and took measures to reverse the
zinc overdose with a chelate. However, sadly, the patient
perished. The ISMP concluded that this fatal event was due
to “systems-based causes” that, if corrected, may prevent
another such error from occurring. Could technology have
prevented this event? At the end of this section, this event
will be revisited to explore which issues could have been
prevented by the use of the technologies available today.
CPOE and EHR-to-Compounder Interfaces
CPOE and EHR-to-compounder interfaces go hand in
hand to providing the safest systems for compounding
PNs. (Note: During this discussion, it is implied that the
electronic medical record contains CPOE.) When EHR is
implemented, it eliminates many of the errors associated
with illegible handwriting and incorrect units (mcg or mg,
etc.). However, many EHRs do not interface with the
ACDs used to compound PN. A survey of hospitals in
2014 by Vanek et al4 showed that 28% of hospitals use
an EHR-to-ACD interface when compounding PN. If the
EHR does not interface with the ACD, then the EHR
prints a label with the PN formula in central pharmacy.
Then, the compounding pharmacist or technician still must
transcribe the PN formula into the ACD software before
the PN is compounded. This step allows for transcription
errors. Also, many of the safeguards for compounding PN,
797
including limits for electrolytes such as sodium, calcium,
and phosphate, are programmed into the software for the
ACD. These safeguards are visible to the compounding
pharmacist, but unavailable to the prescriber. It was not
until the advent of EHR-to-compounder interfaces that real
steps for safety could occur. These interfaces allow real-time
safeguards to be built into order entry so that prescribers can
see the limitations of PN mixture compatibility as they are
entering the orders.
A group in Salt Lake City, UT, implemented such an
EHR-to-compounder interface system that set soft-stop
and hard-stop limits as well as education prompts.5 These
limits and education prompts were set for macronutrients,
electrolytes, micronutrients (trace minerals and carnitine).
The “soft-limit prompts” allow the prescriber to override
them and then allow the verifying pharmacist to review the
override on verification. The verifying pharmacist is then
able to discuss the override with the prescriber and then they
can decide together if the override is reasonable. The “hard-
limit prompt” is not able not able to be overridden by the
prescriber or the pharmacist. For example, in the author’s
institution, the sodium has been preset to a hard limit of
154 mEq/liter. (This was an institutional decision.)
Recommendations for PN component ordering and la-
beling followed American Society for Parenteral and En-
teral Nutrition guidelines.6-9 As this system was designed
for a pediatric hospital, the choices for PNs were rather
elaborate, as the group determined “there were 8 natural
breakpoints for weight categories.” This means the provider
initially chooses the correct PN order based on the weight
of the patient and then the PN limits are built into the
order according to each weight category. Each PN weight
“category” (weight 0–1.5 kg, 1.6–3 kg, etc.) is different, and
therefore the soft and hard limits for each weight category
are different as well. After implementation of the system,
the group revealed a rate of medications errors related to
PN of 2.7 per 1000 PN solutions (a 0.27% error rate).
The researchers categorized the errors according to the
following traditional steps of the medication use process:
prescribing, transcription, preparation, and administration.
Transcription errors were eliminated, and most errors (95%)
occurred during administration.
This resulting decrease in errors is impressive. A prospec-
tive observational study by Sacks et al2 in 2004 (pre-EHR)
reported a frequency of 15.6 errors per 1000 PN prescrip-
tions or a 1.6% error rate. Most of the errors reported
occurred during the transcription and administration steps
of the medication use process, 39% and 35%, respectively.
The Salt Lake City group eliminated transcription, thus
eliminating a large source of error. This is likely to responsi-
ble for the lower error rate when compared with the study
by Sacks et al.2 The Salt Lake City group also reported
that 95% of the errors after the interface were related to
administration errors. This is similar to the report by Sacks
798
et al,2 as they reported that transcription and administration
were the steps that were most prone to error.
Of course, there are some challenges that come with
interfaces. For the interface to work properly, there must be
robust technical support in the setting of product shortages
and formulary additions. It is also important that the
verifying pharmacist role remains intact to review the PN
order for appropriateness regarding age, weight, disease
state, and other medications (correct dose in PN, inter-
actions/therapeutic duplication, and Y-site compatibility).
Also, there is concern is regarding downtime and emergency
procedures. What happens when the EHR goes down or if
there is a power outage and all orders must be converted to
paper—the limits are then not available to guide prescribers.
These are all challenges to consider when implementing an
EHR-to-ACD interface.
BCMP Systems
Although ACDs mix 90%–95% of the ingredients contained
in PN, there are some components that are, or should be,3
added by hand. These components, such as insulin, thiamin,
zinc, selenium, copper, folic acid, pyridoxine, and so on,
are too small in volume or weight to be measured by the
ACD and must be drawn up and added manually. Because
pharmacy technicians usually draw these components up,
a pharmacist must check them. Traditionally, the “pull-
down” method is used, whereas the pharmacy technician
adds the medication or component to the PN and then pulls
the syringe plunger “down” to the dose that was added. The
technician then includes this syringe and the medication vial
in with the bin for the pharmacist to check. This method is
prone to error because the pharmacist is not 100% certain
that the medication is added in the dose that was “pulled
down” on the syringe.10
The new BCMP system allows for safer compounding.
When an intravenous medication is compounded, the phar-
macy technician scans the label into the system. The BCMP
system will show, on a computer screen, the intravenous
fluid, medication, and dose of each (if applicable) needed
to compound the finished product. The technician will then
pick and scan each component into the system. The BCMP
system will detect if the technician has chosen the incorrect
medication, diluent, concentration, or expired medication,
and will not allow the technician to proceed to the next step
in the system until the technician scans the correct products.
Then the technician draws the amount of medication from
the vial into a syringe and the system takes a digital photo
of all the components, including the label, medication,
intravenous fluid, reconstituted medication (if applicable),
and “drawn up dose.” The technician then compounds the
sterile product, and the system takes another digital photo
of the finished preparation. The pharmacist is then able to
check each digital photo to observe what is actually added to
Nutrition in Clinical Practice 33(6)
the finished sterile product. For instance, if insulin is added
to a PN, then the technician takes the following actions:
(1) picks the insulin vial; (2) scans the insulin vial; (3) draws
up the correct insulin dose; (4) takes a digital photo of
the label, the insulin vial, and the syringe drawn up to
the correct dose; (5) adds the insulin to the PN. Then the
pharmacist can check in the system and per the digital
photos that the correct strength of insulin was picked and
the correct dose of insulin was drawn up. With this system,
there is extra assurance that the correct products are picked
and within expiration because the technician is required to
scan the barcodes before sterile product preparation. Also,
there is more assurance that correct doses are added because
the pharmacist can see a photo of the dose before it is added
instead of seeing what the technician thought he/she added
after the fact.
Improvement in error rate by implementation of a
BCMP system is well documented. Speth et al10 first studied
the system in 2009 and showed that after 2 years of
implementation, their error detection rate was 0.8%. The
researchers compared the error rate with the BCMP system
of 0.8% to a previously reported error rate of a traditional
paper system of 9%11 and concluded that the BCMP
system results in fewer preparation errors. Of the barcode-
intercepted errors, 60% were due to the technician scanning
the wrong drug, 28% due to the wrong fluid or diluent,
and 12% due to the wrong concentration of drug. Of the
pharmacist-intercepted errors, 75% were due to an incorrect
amount of drug or diluent, which were errors that would not
have been detected in the traditional paper system. The rest
of pharmacist-intercepted errors were attributed to unclear
digital photos or due to an improper sterile technique.
Another advantage that these authors identified was the
decrease in medication turnaround times. Turnaround times
decreased for new orders, stat orders, and chemotherapy
orders. The most dramatic decrease was for chemotherapy
orders from 52 minutes prior to BCMP down to 34 minutes
after BCMP implementation. The researchers posit this is
due to the following aspects of the BCMP system: (1) the
system notifies the pharmacist when a sterile product is
finished and ready to be checked, (2) the pharmacist can
check the sterile product remotely by viewing the digital
photos on a personal computer, (3) many errors are detected
and intercepted prior to mixing (i.e., when the technician
picks and scans the incorrect product), and wasted doses are
greatly reduced. The researchers reported a waste reduction
of approximately $30,000 during the first 12 months of
implementation. They attribute this to avoidance of errors
as well as due to the system’s ability to track doses discarded
after they have past the beyond-use date.
Moniz et al12 implemented a BCMP system in their
hospital pharmacy and collected data regarding the number
of doses prepared in the pharmacy, the number of errors
detected by pharmacists, and the number of reworked
Curtis
or rejected doses.12 The time of data collection was 5
months. The errors were reviewed for error detectability
(would the errors have been detected in the traditional
system?) and were also scored for harm. Using the BCMP
system, 425,683 medication doses were prepared during the
13-month study period. The error-detection rate by the
entire system (barcode and pharmacist final checks) was
0.68%, and the number of reworked or rejected doses was
1223 and 1667, respectively. Of the rejected doses, 375 doses
(22.4%) involved errors that were deemed “undetectable”
before implementation of the BCMP—that is, the pharma-
cist would not have been able to detect these errors due to the
lack of the digital photos they now had at their disposal with
BCMP. The other rejected doses involved errors deemed to
be detectable with the traditional system (prior to BCMP
system implementation) or errors introduced by the BCMP
system. Of the “undetectable” errors, 21 were judged to
be of a “severe harm” severity. The error rate with the
BCMP system of 0.68% was compared with a previously
reported error rate of a traditional paper system of 9%.11
The researchers concluded that the BCMP system results in
fewer preparation errors.
Deng et al13 also implemented a BCMP in their pediatric
hospital pharmacy and had similar results. After a year
of data collection, 421,730 doses were analyzed for errors.
The error detection rate was 0.74%. Of the 3101 errors, the
barcode scanning technology detected 2241 (72.27%) of
the errors before the pharmacist inspected the product on
the digital photos. The pharmacist was needed to detect the
remainder of the errors (27.73%).
All of these examples of BCMP were employed in
hospital pharmacies that compound sterile products. None
of these articles specifically mention data regarding the use
of BCMP in the compounding of PN. However, BCMP can
be applied to compounding PNs, as it is currently used in
the author’s institution.
Systems-Based Errors?
Returning to the fatal error in compounding3 described
earlier, how would the aforementioned technologies helped
to prevent them? For the zinc-overdose event, the first error
was a paper–ACD dose mismatch, as zinc was prescribed
as 330 mcg per 100 mL, but the ACD required the dose
to be in a mcg/kg dose. This error would have been solved
by both the CPOE and EHR-to-compounder interfaces,
as the ACD requires all doses to be entered correctly.
The next was a pharmacist calculation error, resulting in
a concentration of 330 mg per 100 mL. If a prescriber
entered this in error into a system with limits, the prescriber
would not be able to continue and the PN would not be
mixed. Next, the technician mixing the PN refilled the zinc
syringe in the compounder a total of 11 times. The safe
practice recommendation from ISMP is to “allow manual-
799
only” additions of low-volume ingredients. Although this
is a low-tech fix, it leads to a high-tech fix. If the zinc is
added as a manual ingredient, then it can go through the
BCMP workflow. Even if all of the other systems had failed,
the technician would then have scanned the zinc and the 11
syringes and taken digital photos. Hopefully, when seeing
the 11 syringes full of zinc, the pharmacist would have
known something was wrong at that point! Regardless, this
example illustrates how technology could have halted this
error before it reached the patient.
Quality Assurance
Refractometry
Maintaining quality assurance of ACDs is recommended
by United States Pharmacopeia (USP) Chapter <797>
as well as by the American Society for Hospital Phar-
macists and the American Society for Parenteral and
Enteral Nutrition.9,14,15 In USP Chapter <797>, quality
assurance can be accomplished by way of gravimetric or
volumetric tests. The chapter also suggests that “additional
tests of accuracy may be employed that determine the
content of certain ingredients in the final volume of the
PN admixture.”14 Some hospitals are using refractometry
or refractive index (RI) to aid in the quality assurance
of PN formulations. However, refractometry can only be
used for 2-in-1 PN formulations, that is, non-fat-containing
formulations. In the following 2 articles, pharmacists or
nurses use a predictive equation to calculate refractive index
and then compare it to the refractive index measured by a
small amount of PN formula dropped onto a refractometer
and then compare the 2 values. If the provider finds that the
values do not match (within a standard deviation), then the
provider asks the pharmacy to compound the PN again.
Refractive index of a substance is “the ratio of the
velocity of light in air to the velocity of light in the sub-
stance.” Refractive index increases with the concentration
of solutes in solution. Refractive index has been used to
verify concentrations of enteral nutrition and drugs.16-18
Chang and Yeh19 devised an equation to predict refractive
index for 2-in-1 PN formulations and validated the equation
with 500 prepared PNs. Chang and Yeh19 tested 4 predictive
equations and compared them with the values measured
by the refractometer. They used a hand-held refractometer,
which required 1 or 2 drops of the PN formulation for
measurement. Of the predictive equations, the first 3 were
variations based on concentrations of the PN ingredients
(equation 1 = amino acids (g/L) × 8 + glucose g/L × 7 +
sodium (mEq/L) × 2 + phosphate (mg/L) × 0.2 − 50), and
equation 4 used Brix values (equation 4 = 81.05 × Brix value
− 116.33). Brix values are the same as measured refractive
index. Chang and Yeh19 used Brix values, measured by their
refractometer, of known mixes of amino acid +dextrose
800
concentrations (eg, amino acid 4%, dextrose 6%, Brix
value reported is 11.8 ± 0.1) Of the 500 PNs, the authors
found a very high correlation of equation 4 with measured
refractometry. Equation 4, which uses a Brix value, was
successful in predicting 100% of the refractive index for
PNs within a measured osmolality range of 600 to 900
mOsm/kg for a CI range of 90%–110%. The authors were
confident in using the predictive equation and hand-held
refractometer as a quality assurance tool to check PNs for
compounding accuracy. The authors state that limitations
of using refractometry as a quality assurance method is the
time and labor needed to calculate predictive equations and
use the refractometer on the final PN formula.
Nelson et al20 also derived and validated an equation
for refractive index. Nelson et al20 used a refractometer
of a “reported accuracy of 0.00004” in an environment
controlled for temperature, as recommended by USP and
the manufacturer of the refractometer. They first measured
a refractive index for every ingredient used in the PN formu-
lations compounded at their institution. They found that the
ingredient indices were additive; that is, if 1 mL of sodium
chloride 2 mEq/mL had a refractive index of 1, then 2 mL of
sodium chloride 2 mEq/mL had a refractive index of 2. The
equation they tested for the refractive index of PN formula-
tions, when 1.333 is the RI of sterile water, was the following:
RIfinal = 1.333 + ࢣ(RIconcentrate [volume of concentrate
(mL)/total volume of PN solution (mL)]). They validated
this equation for refractive index on 1057 pediatric PN
solutions. Of the 1057 PNs, 99.8% of the refractometer
measurements fell within the acceptable range (within 2
standard deviations) of the predictive RI. The authors note
that limitations of the refractometer include cost, personnel
training, and that the main factors that drive refractive index
are concentrations of dextrose and amino acids. Electrolyte
changes are not readily picked up by refractometry. The
authors now use refractometry testing in their pharmacy
routinely for the first and last PN bag compounded each
day, as well as after each time the dextrose bag is changed on
the ACD. The authors state this reduced their error rate to
“2 compounding errors in every 10,000” PNs compounded;
however, they did not list the error rate before the refractom-
etry test was initiated.
Laboratory Testing
PN formulations may be tested by an institution’s laboratory
for quality assurance.14 Per USP Chapter <797>, this test
may only be used for quality assurance if the institution
standardizes its methods to USP references and procedures.
Laboratories are accustomed and calibrated to test blood
and body fluids. Thus if the laboratory is to test a PN,
the laboratory may need to validate to the high range of
dextrose that is typically found in a PN, instead of the range
found in blood.
Nutrition in Clinical Practice 33(6)
Sterility Detection
Keeping PN sterile using aseptic technique is essential
in delivering safe and effective nutrition support.9,15 A
painful consequence of interrupting sterile procedure is
the outbreak of Serratia marcescens infection due to the
contaminated bags of PN in 2011.21 As a result of the out-
break, 9 patients died. Although the pharmacy providing
PN in this example was checking for contamination and
endotoxin, the sample size was too small to detect bacterial
growth and the results were not available until after the PNs
were administered. New technologies are emerging to aid
in detecting bacterial and endotoxin contamination much
sooner so that this tragedy is not repeated.
Traditional sterility testing for compounded sterile prod-
ucts required by USP <797> is performed by taking a sam-
ple of the compounded sterile product, passing it through
at 0.45 μm filter, and dividing the filter into 2 parts.14
One part of the filter is transferred to tryptic soy broth
medium so as to grow aerobic microorganisms, and the
other part is transferred to fluid thioglycollate medium so as
to grow both aerobic and anaerobic microorganisms. Both
media solutions are stored at the appropriate temperatures
(defined by USP) and incubated for 14 days. Of course, this
type of testing is not feasible for PN for several reasons.
First, PN would need to be administered before the results
from testing returned, given the beyond-use date for PN
stored at room temperature (20–25°C) is 24 hours and
at refrigerated temperature (2–8°C) is 9 days. (If the PN
is stored at refrigerated temperature, the beyond-use date
is 9 days only if a multivitamin has not been added to
the PN.). Also, this method would only be feasible for
2-in-1 PNs, as 3-in-1 PNs cannot be filtered through a
0.45 μm filter. A faster detection method is available and
based on the detection of carbon dioxide production using
colorimetric methods.22 This system uses a 10 mL sample,
incubates the microorganisms, shakes the samples, and takes
measurements every 10 minutes. This system can detect
growth of certain microorganisms within 24 hours and is
approved by the U.S. Food and Drug Administration for
detecting microorganisms in blood and body fluids. The
system has limitations, however, as it is not approved by
the USP for verifying sterility in pharmaceuticals. Because
it uses a 10 mL sample that is not filtered, it may be
appropriate for testing both 2-in-1 and 3-in-1 formulations.
It is used to test blood and body fluids; however, there are
no examples in the literature of this system’s use in testing
emulsions.
A brand of this system, the BacT/Alert automated
system (bioMérieux, Durham, USA) has been used in South
Africa to detect microorganisms in PN.23 In this instance,
2-in-1 PNs are commercially mixed, irradiated, and sent to
the National Health Laboratory Services for testing. The
BacT/Alert system was used to monitor the pre-irradiated
Curtis
PNs for microorganisms in addition to soy broth. The
system isolated Candida parapsilosis, which was further cul-
tured. After further investigation, this organism was isolated
from work stations within the pharmacy. This report shows
that the BacT/Alert system is able to detect microorganisms
in PN solutions; however, many important barriers prevent
this system from being implemented in daily use. First, PNs
in this situation were being commercially produced, which
means that the PNs presumably had a long shelf life. The
BacT/Alert system may be useful when the end product
has a shelf life of 6 months or more, but perhaps not
useful if the product needs to be used immediately. However,
the BacT/Alert system might have a future role in home-
infusion pharmacies. As home-infusion pharmacies batch
PNs then refrigerate them (beyond-use date is 9 days), in
the future pharmacists might be able to use results from
the BacT/Alert system in time to dispose of a contaminated
batch. There is currently no data available that tests the use
of BacT/Alert system in home-infusion pharmacies for PN
sterility, although this is an opportunity for future research.
Endotoxin Detection
Currently, PN is not tested for endotoxin. However, as
many components are added to the PN, it is possible that
endotoxins are introduced. If a quick test for endotoxin
was available, it would be desirable to use for PN to
prevent avoidable fevers in the patient. The oldest endotoxin
detection approved by the USP is the rabbit pyrogen test
based on the rise in temperature of rabbits after injection
of a solution containing endotoxin.24 This method has
slowly been replaced by a more scientific and feasible test,
the limulus amoebocyte lysate test. This test is based on a
substance derived from the blood of the horseshoe crab.
The substance, limuluspolyphemus, reacts in an enzymatic
process with endotoxin and visibly clots when the endotoxin
reaches a certain concentration. However, due to the enzy-
matic nature of the process, some forms of glucose have
activated limulus amoebocyte lysate . Therefore, it may not
be suitable to test PN for endotoxin.
Newer ways of detection of endotoxin are on the fore-
front of technology. One way is the use of CD14 to bind
endotoxin. Through the use of recombinant CD14 and
the development of an electrochemilumninescent assay, it
is possible to detect endotoxin. However, many challenges
exist with this detection system, including the tendency of
CD14 to bind to a range of molecules other than endotoxin.
Other ways to detect endotoxin can be specific and sensitive,
such as the use of antibody sensors, similar to the use
of immunoassay. This method has been used to develop
monoclonal antibodies specific and sensitive to endotoxin.
Aptamers are also highly specific and sensitive to detecting
endotoxin, even in miniscule amounts. Aptamers are 3-
dimensional molecules designed from deoxyribonucleic acid
801
for a specific target. However, both antibody sensors and
aptamers are too costly and labor intensive to use for testing
PN on a daily basis. These tests are all in their infancy, and
it is also too early to know if they might be used for 2-in-1,
3-in-1, or both kinds of PNs.
Summary
In conclusion, technology has solved many issues related to
compounding PN, including transcription errors and direct-
to-prescriber safeguards. However, more challenges remain,
including the need for back-up systems in the event of EHR
failures or electric outages. Also, there is much opportunity
for wider installation of EHR-to-compounder systems in
the coming years.
With regard to quality assurance, refractometry has been
used in some hospitals as an additional quality assurance
check, but the cost of and equipment and training may
prohibit widespread application of use. Laboratory checks
of PN for quality assurance checks are available in select
hospitals, if the laboratory has the equipment and is able
to calibrate per USP Chapter <797> requirements. The
availability and use of this method of as a quality assurance
check would be an opportunity for future research.
Although the faster detection of microorganisms is
available, it is not approved for testing of sterility of
compounded medications. Endotoxin testing remains an
expensive and labor-intensive endeavor. There is much
opportunity for future research in sterility and endotoxin
testing of PN formulations. Technology is advancing expo-
nentially faster, and further research may address all of these
challenges more quickly than expected.
Statement of Authorship
C. Curtis contributed to conception/design of the research;
contributed to acquisition, analysis, or interpretation of
the data; drafted the manuscript; critically revised the
manuscript; and agrees to be fully accountable for ensuring
the integrity and accuracy of the work.
References
1. Institute for Safe Medication Practices. ISMP’s list of high-alert med-
ications. 2017. http://www.ismp.org/Tools/highalertmedications.pdf.
Accessed September 26,2017.
2. Sacks GS, Rough S, Kudsk KA. Frequency and severity of harm of
medication errors related to the parenteral nutrition process in a large
university teaching hospital. Pharmacotherapy. 2009;29(8):966-974.
3. Institute for Safe Medication Practices. Fatal 1,000-fold overdoses
can occur, particularly in neonates, by transposing mcg and mg.
https://www.ismp.org/newsletters/acutecare/articles/20070906.asp. Ac-
cessed October 31, 2017.
4. Vanek VW, Ayers P, Charney P, et al. Follow-up survey on function-
ality of nutrition documentation and ordering nutrition therapy in
currently available electronic health record systems. Nutr Clin Pract.
2016;31(3):401-415.
802
5. MacKay M, Anderson C, Boehme S, Cash J, Zobell J. Frequency
and severity of parenteral nutrition medication errors at a large
children’s hospital after implementation of electronic ordering and
compounding. Nutr Clin Pract. 2016;31(2):195-206.
6. Boullata JI, Gilbert K, Sacks G, et al. A.S.P.E.N. clinical guidelines:
parenteral nutrition ordering, order review, compounding, labeling,
and dispensing. JPEN J Parenter Enteral Nutr. 2014;38(3)334-377.
7. Druyan ME, Compher C, Boullata JI, et al. Clinical guidelines for the
use of parenteral and enteral nutrition in adult and pediatric patients:
applying the GRADE system to development of A.S.P.E.N. clinical
guidelines. JPEN J Parenter Enteral Nutr. 2012;36(1)77-80.
8. Ayers P, Adams S, Boullata J, et al. A.S.P.E.N. parenteral nutrition
safety consensus recommendations. JPEN J Parenter Enteral Nutr.
2014;38(3):296-333.
9. Mirtallo J, Canada T, Johnson D, et al. Safe practices for parenteral
nutrition. JPEN J Parenter Enteral Nutr. 2004;28(6)S39-S70.
10. Speth SL, Fields DB, Schlemmer CB, Harrison C. Optimizing i.v.
workflow. Am J Health-Syst Pharm. 2013;70:2076-2080.
11. Flynn EA, Pearson RE, Barker K. Observational study of accuracy
in compounding i.v. admixtures in five hospitals. Am J Health-Syst
Pharm. 1997;54:904-912.
12. Moniz TT, Chu S, Tom C, et al. Sterile product compounding using an
i.v. compounding workflow management system at a pediatric hospital.
Am J Health-Syst Pharm. 2014;71:1311-1317.
13. Deng Y, Lin AC, Hing J, et al. Risk factors for i.v. compounding errors
when using an automated workflow management system. Am J Health-
Syst Pharm. 2016;73:887-893.
14. Pharmaceutical compounding—sterile preparations (chapter <797>).
In: The United States Pharmacopeia, 38th Rev., and the National
Formulary, 33nd ed. Rockville, MD: United States Pharmacopeial
Convention, 2015:43-87.
Nutrition in Clinical Practice 33(6)
15. American Society of Health-System Pharmacists. ASHP guidelines
on the safe use of automated compounding devices for the prepa-
ration of parenteral nutrition admixtures. Am J Health-Syst Pharm.
2000;57:1343-1348.
16. Cheung JF, Chong S, Kitrenos JG. Use of refractometer to detect
controlled substance tampering. Am J Hosp Pharm. 1991;48:1488-
1492.
17. Frankenfield DL, Johnson RE. Refractometry of controlled sub-
stances. Am J Hosp Pharm. 1991;48:2129-2130.
18. Chang WK, Chen MZ, Chao YC. Use of the refractometer as a tool
to monitor dietary formula concentration in gastric juice. Clin Nutr.
2002;21:521-525.
19. Chang WK, Yeh MK. Prediction of parenteral nutrition osmolarity
by digital refractometry. JPEN J Parenter Enteral Nutr. 2011;35:412-
418.
20. Nelson S, Barrows J, Haftmann R, Helm M, MacKay M. Calculating
the refractive index for pediatric parenteral nutrient solutions. Am J
Health-Syt Pharm. 2013;70:350-355.
21. Gupta N, Hocevar S, Moulton-Meissner H, et al. Outbreak of Serratia
marcescens bloodstream infections in patients receiving parenteral
nutrition prepared by a compounding pharmacy. Clin Infect Dis.
2014;59:1-8.
22. Kaiser SJ, Mutters N, Backhaus J, et al. Sterility testing of in-
jectable products: Evaluation of the growth-based BacT/ALERT 3D
dual T culture system. PDA J Pharm Sci and Tech. 2016;70:568-
576.
23. Marais E, Stewart R, Dusé AG. Candida parapsilosis detected in TPN
using the BacT/Alert system and characterized by randomly amplified
polymorphic DNA. J Hosp Infect. 2004;56:291-296.
24. Su W, Ding X. Methods of endotoxin detection. J Lab Autom.
2015;20:354-364.
Review

Nutrition in Clinical Practice

Volume 33 Number 6
Effect of Early vs Late Start of Oral Intake on Anastomotic December 2018 803–812

C 2017 American Society for

Leakage Following Elective Lower Intestinal Surgery: Parenteral and Enteral Nutrition
DOI: 10.1177/0884533617711128
A Systematic Review wileyonlinelibrary.com

Boudewijn J. J. Smeets, MD1,2 ; Emmeline G. Peters, MD1,3 ;

Eelco C. J. Horsten, BSc1 ; Teus J. Weijs, PhD1 ; Harm J. T. Rutten, PhD1,2 ;
Willem A. Buurman, PhD4 ; Wouter J. de Jonge, PhD3 ; and Misha D. P. Luyer, PhD1

Abstract
Background: Experimental and clinical studies have demonstrated a beneficial effect of early enteral nutrition (EN) on anastomotic
leakage following colorectal surgery. Early oral intake is a common form of early EN with various clinical benefits, but the effect
on anastomotic leakage is unclear. This systematic review investigates the effect of early vs late start of oral intake on anastomotic
leakage following lower intestinal surgery. Methods: A systematic literature search was performed using the PubMed, Embase,
Medline, and Cochrane databases. Randomized controlled trials were included that compared early (within 24 hours) vs late start
of oral intake following elective surgery of the small bowel, colon, or rectum. Meta-analysis was performed for anastomotic leakage,
overall complications, length of stay, and mortality. Sensitivity analysis was performed in which studies of inferior methodological
quality were excluded. Results: Nine studies including 879 patients met eligibility criteria. Early start of oral intake significantly
reduced overall complications (odds ratio [OR], 0.65; 95% confidence interval [CI], 0.46–0.93; P = .02), length of stay (mean
difference, −0.89; 95% CI, −1.22 to −0.57; P < .001), and anastomotic leakage (OR, 0.40; 95% CI, 0.17–0.95; P = .04) compared
with late start of oral intake. However, in the sensitivity analysis only the overall reduction of length of stay remained significant.
Conclusion: The effect of early oral intake on anastomotic leakage is unclear as existing studies are heterogeneous and at risk of bias.
High-quality studies are needed to study the potential benefit of EN on anastomotic healing. (Nutr Clin Pract. 2018;33:803–812)

Keywords
enteral nutrition; surgery; wound healing; nutritional support; colorectal surgery; anastomotic leak; length of stay; mortality; meta-
analysis

Anastomotic leakage (AL) is a severe complication fol-
lowing colorectal surgery as it is associated with increased
morbidity, mortality, and cancer recurrence rates.1-4 Despite
ongoing efforts, strategies that effectively reduce AL are
lacking and the incidence has remained stable over the
years.5
Several experimental studies have demonstrated that en-
teral nutrition (EN) may improve anastomotic healing.6-12
Moreover, in 2 recent randomized controlled trials, AL
following colorectal surgery was significantly reduced by
means of early postoperative EN13 and direct perioperative
sham feeding.14 Taken together, these results suggest that
EN administered close to surgery may provide new thera-
peutic opportunities to reduce AL.
A common method to provide EN close to surgery
is the early postoperative start of oral intake. Systematic
reviews on the effects of early start of oral intake following
gastrointestinal (GI) surgery have demonstrated clear ben-
efits, including a reduction in length of stay (LOS), overall
complications, and mortality.15-18 However, in these reviews,
early oral intake did not affect AL.15-18 Importantly, these
systematic reviews included various types of GI surgery or
included studies that combined other elements of fast-track
protocols in the intervention group but not in the control
group.15-18 To further investigate the potential beneficial
effects of EN on anastomotic healing, this systematic review
From the 1 Department of Surgery, Catharina Hospital, Eindhoven,
the Netherlands; 2 GROW School of Oncology and Developmental
Biology, Maastricht University, Maastricht, the Netherlands; 3 Tytgat
Institute for Intestinal and Liver Research, Academic Medical Center,
Amsterdam, the Netherlands; and the 4 School for Mental Health and
Neuroscience, Maastricht University, Maastricht, Limburg, the
Netherlands.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on June 19, 2017.
Corresponding Author:
Boudewijn J. J. Smeets, MD, Department of Surgery, Catharina
Hospital Eindhoven, Michelangelolaan 2, Eindhoven, Noord-Brabant
5623 EJ, the Netherlands.
Email: boudewijn.smeets@catharinaziekenhuis.nl
804
compares the effect of early vs late start of oral intake on AL
following elective lower intestinal surgery.
Methods
This systematic review was performed according to
the Cochrane Handbook for Systematic Reviews for
Interventions19 and the Preferred Reporting Items for
Systematic Reviews and Meta-Analyses (PRISMA)
statement guidelines.20 The entire review process (ie, article
search, critical appraisal, data extraction, and analysis) was
conducted by 3 independent researchers (B.J.J.S., E.G.P.,
and E.C.J.H.). Any disagreements were resolved through
discussion.
Eligibility Criteria
We included only randomized controlled clinical trials that
reported on the effects of early vs late start of oral intake
on AL following elective surgery of the small bowel, colon,
or rectum in patients aged ࣙ18 years. We defined early start
of oral intake as any caloric intake started within 24 hours
after surgery. Late start of oral intake was defined as nil by
mouth until resolution of postoperative ileus (ie, passage of
flatus or stool without presence of nausea or vomiting). To
assess the true effect of early oral intake alone, we excluded
studies that combined early oral intake with other elements
of fast-track protocols in the intervention group but not in
the control group.
Information Sources and Search Strategy
The PubMed, Medline, Embase, and Cochrane databases
were systematically searched. The search strategy combined
all synonyms regarding the intervention “early oral intake”
and the domain “lower intestinal surgery” with the Boolean
operator “AND.” All synonyms were combined with the
Boolean operator “OR.” An example set of search terms
is provided in Supplementary Table S1. We tested the
sensitivity of the search strategy by screening all references
of included articles for relevant publications that were not
retrieved in the initial search. Furthermore, we screened
all citing articles and related articles using Web of Science
version 5.15.1. Identification of additional eligible articles
led to evaluation and improvement of the search strategy
until it retrieved all eligible articles. We contacted authors
by email if articles were not available in full text. The search
was updated until September 28, 2016.
Study Selection
Three authors screened all records on title and abstract.
Records were excluded if they clearly did not address the
domain and intervention under investigation. The remain-
ing articles were screened in full text. Only articles fulfilling
all eligibility criteria were included.
Nutrition in Clinical Practice 33(6)
Data Extraction and Outcomes
The relevant published data were collected in pilot-tested
tables. Extracted information from each study included (1)
study information, including name of first author, year of
publication, number of participants in each group, and
reported outcomes; (2) patient information, including type
of surgery, disease, sex, age, and perioperative protocols
used affecting AL21 ; and (3) postoperative feeding proto-
cols. Furthermore, we extracted data on the incidence and
definition of AL, overall complications, mortality, and LOS.
Risk of Bias in Individual Studies
All studies included in the review were investigated for risk
of bias with the Cochrane collaboration’s tool for assessing
risk of bias.19 Risk of bias was assessed on the follow-
ing items: randomization method, allocation concealment,
blinding of participants and personnel, blinding of outcome
assessment, incomplete outcome data, selective reporting,
and any other item in study design.
Data Analysis
Meta-analysis was performed for AL, overall complications,
LOS, and mortality. We performed data analysis using
Review Manager Software version 5.3 as recommended
by the Cochrane Handbook for Systematic Reviews for
Interventions.19,22 Dichotomous results were analyzed using
the random-effects model in the Mantel-Haenszel method
and are presented as odds ratio (ORs) with corresponding
95% confidence intervals (CIs). Continuous results were an-
alyzed using the inverse variance method and are presented
as mean difference (MD) with corresponding 95% CI. A P
value <.05 was considered to be statistically significant. We
assessed presence and amount of statistical heterogeneity
using the I2 statistic. Furthermore, a sensitivity analysis
was performed in which studies of inferior methodological
quality were excluded.
Results
Description of Studies
Figure 1 presents the search results and study selection
process. The search revealed 80 potentially relevant stud-
ies, of which 9 randomized controlled trials fitted in-
clusion criteria. Study characteristics are shown in Ta-
ble 1. Types of surgical procedures were evenly matched
between groups in all studies. Laparoscopic surgery was
performed only in 1 study23 ; in other studies, open surgery
was performed.24-31 When reported, there was substantial
heterogeneity between studies in the use of perioperative
protocols, including use of epidural anesthesia, preopera-
tive bowel lavage, nonsteroid anti-inflammatory drugs, and
Smeets et al 805

Figure 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flowchart for search results and study
selection process.

opioids. Postoperative feeding protocols are described in
Table 2.
AL was a rare event in most studies and rarely a
primary outcome. Four studies provided definitions for AL
(Table 3).25,26,28,31
Methodological Quality of Included Studies
Due to the nature of the intervention, blinding could not
be applied in any study. Three studies gave no explicit
description of the randomization method, resulting in
an unclear risk of bias.26,27,30 One open-label study used
blocked randomization with a fixed block size of 6; hence,
risk of selection bias was present.29 In the same study, it was
unclear whether the nasogastric tube was postoperatively
removed in both groups at similar time points. In 1 study,28
patients allocated to delayed start of oral intake had a
significantly greater amount of intraoperative blood loss
(early oral intake median 300 mL vs delayed oral intake
median 800 mL, P = .002), which is a known risk factor for
AL.21 The summary of risk of bias assessment is shown in
Table 4. Overall, 3 studies were identified to have the best
available methodological quality and were entered in the
sensitivity analysis.23,24,31
AL
First, all studies were entered in the meta-analysis regardless
of methodological quality. As shown in Figure 2A, early
start of oral intake significantly reduced AL compared with
late start of oral intake (OR, 0.40; 95% CI, 0.17–0.95; P =
.04). However, when trials with risk of bias were excluded
in the sensitivity analysis, the overall effect on AL was no
longer significant (OR, 0.39; 95% CI, 0.09–1.76; P = .22)
(Figure 2B).
Overall Complications
All studies provided data on overall complications. As
shown in Figure 3A, early start of oral intake significantly
reduced overall complications compared with late start
of oral intake (OR, 0.65; 95% CI, 0.46–0.93; P = .02).
However, when trials with risk of bias were excluded in the
sensitivity analysis, the overall effect on overall complica-
tions was no longer significant (OR, 0.74; 95% CI, 0.42–
1.31; P = .30) (Figure 3B).
LOS
Seven studies provided data on LOS as mean ± standard
deviation and were entered in the meta-analysis.23-26,28-30 As
shown in Figure 4A, early start of oral intake significantly
reduced LOS compared with late start of oral intake (MD,
−0.89; 95% CI, −1.22 to −0.57; P < .001). Excluding
studies with risk of bias increased the overall effect of
LOS (MD, −3.47; 95% CI, −4.73 to −2.21; P < .001)
(Figure 4B).
806
Table 1. Study Characteristics.a

Mean Age, y Sex (Male/Female), No.

Author and Publication Early Oral Late Oral Early Oral Late Oral
Year Site of Surgery Disease Intake Intake Intake Intake Reported Postoperative Outcomes

da Fonseca et al,23 2011 Colon NR 57 52 8/16 10/16 LOS, recovery of GI function, overall
morbidity, anastomotic leakage,
readmission, surgical
reintervention after discharge,
mortality
Dag et al,24 2011 Colon, rectum Neoplasia 62 61 52/47 61/39 LOS, recovery of GI function,
complications
El Nakeeb et al,25 2009 Colon, rectum Neoplasia 52 56 39/21 42/18 LOS, recovery of GI function,
complications, readmission,
mortality
Hartsell et al,26 1997 Colon, rectum Neoplasia, diverticular 66 68 NR NR LOS, recovery of GI function,
disease, inflammatory complications
bowel disease
Lucha et al,27 2005 Colon, rectum NR 51 51 17/9 16/9 LOS, recovery of GI function,
aspiration pneumonia,
anastomotic leakage, antiemetic
medication, admission costs
Minig et al,28 2009 Small bowel, Gynecological neoplasia 54 58 0/18 0/22 LOS, recovery of GI function,
colon, rectum complications, intensity of
abdominal pain, patient
satisfaction level, quality of life,
antiemetic and analgesic
medication
Pragatheeswarane et al,29 Small bowel, Neoplasia, familial 47 47 33/27 32/28 LOS, recovery of GI function,
2014 colon, rectum adenomatous complications
polyposis, stricture
Reissman et al,30 1995 Small bowel, NR 51 56 34/46 43/38 LOS, recovery of GI function,
colon, rectum complications
Stewart et al,31 1998 Colon, rectum NR 58 59 19/21 18/22 LOS, recovery of GI function,
complications, antiemetic and
analgesic medication, mobilization

GI, gastrointestinal; LOS, length of stay; NR, not reported.

a All values are presented as patient numbers unless indicated otherwise.
Smeets et al 807

Table 2. Postoperative Feeding Protocols. Table 3. Definitions for Anastomotic Leakage.

Author and Late Oral Author and Publication

Publication Year Early Oral Intake Intake Year Definition

da Fonseca et al,23 POD1 oral liquid diet, SC El Nakeeb et al,25 2009 Symptoms such as fever and
2011 advance to regular diet leakage of intestinal contents
within 24 hours as Hartsell et al,26 1997 Resulting in sepsis and eventual
tolerated death
Dag et al,24 2011 12 hours postoperatively, SC Minig et al,28 2009 Requiring surgical reexploration
start fluids, advance to Stewart et al,31 1998 Fecal discharge from drain tube,
solid diet as tolerated which settled without
El Nakeeb et al,25 POD1 fluids, advance to SC intervention
2009 regular diet within
24–48 hours as
tolerated
Hartsell et al,26 1997 POD1 full liquid diet, SC Experimental studies have suggested that EN can im-
advance to regular diet
prove anastomotic healing via several mechanisms.6-12
if >1000 mL was
consumed within 24 These experimental findings may be corroborated by 2
hours randomized clinical trials of good methodological qual-
Lucha et al,27 2005 8 hours after surgery, SC ity that demonstrated a reduction of AL by means of
start regular diet perioperative sham feeding14 (ie, gum-chewing) and early
Minig et al,28 2009 POD0 CL, advance to SC postoperative enteral tube feeding.13 Early start of oral
regular diet on POD1 intake is a more common form of early EN and has
as tolerated
been extensively described as part of fast-track protocols
Pragatheeswarane et POD1 CL, advance to SC
al,29 2014 full fluid diet within 48 in colorectal surgery. Individual randomized trials have
hours, start solid diet described no effect of early oral intake on AL, but most
over next 24 hours studies had a relatively small sample size and were there-
Reissman et al,30 1995 POD1 CL, advance to SC fore inadequately powered to detect a potential effect on
regular diet as AL.23-31 Previous systematic reviews on early EN also did
tolerated not support an effect on AL.15-18 However, these reviews
Stewart et al,31 1998 4 hours postoperatively, SC
may have been inadequate to assess the true effect of EN,
start free fluids,
advance to solid diet since they included studies with patients undergoing upper
as tolerated on POD1 GI surgery,32 studies that applied immunonutrition,33 or
studies that applied other aspects of fast-track protocols
CL, clear liquids; POD, postoperative day; SC, standard care (nil by only in the treatment group but not in the control group (eg,
mouth until resolution of ileus).
early nasogastric tube removal).34 Furthermore, no review
performed a sensitivity analysis to minimize the risk of
bias.15-18 This study therefore aimed to provide an update
of the available literature and to perform a rigorous critical
Mortality appraisal and sensitivity analysis to examine the true effect
Seven studies provided data on mortality and were entered of EN on anastomotic healing in a clinical setting. In the
in the meta-analysis.23,25,26,28-31 As shown in Figure 5A, early current meta-analysis, the pooling of all 9 randomized trials
start of oral intake did not affect mortality compared with regardless of methodological quality resulted in several
late start of oral intake (OR, 0.61; 95% CI, 0.17–2.22; P = beneficial effects in favor of early oral intake. However,
.45). Excluding studies with risk of bias did not alter the the strict exclusion of 6 studies with a modest to high risk
overall effect of early start of oral intake on mortality (OR, of bias25-30 in the sensitivity analysis significantly reduced
1.04; 95% CI, 0.10–10.35; P = .97) (Figure 5B). overall sample size and made the effect on AL and overall
complications no longer significant. As such, the results
from our study suggest that early start of oral intake is only
Discussion associated with a reduction of length of stay (LOS).
The current review demonstrates that the effect of early The overall reduction of LOS following early oral intake
oral intake on AL is unclear in a clinical setting. This is ranges from almost 1 day in the general meta-analysis to
mainly due to a lack of high-quality evidence, since existing approximately 3 days in the sensitivity analysis. This may
randomized trials are clinically heterogeneous and at risk of be explained by a faster return of bowel function in the
bias. early feeding group, as demonstrated by various indicators
808
Table 4. Critical Appraisal of Included Studies.

da Fonseca Dag et al,24 El Nakeeb Hartsell Lucha Minig Pragatheeswarane Reissman Stewart
Item et al,23 2011 2011 et al,25 2009 et al,26 1997 et al,27 2005 et al,28 2009 et al,29 2014 et al,30 1995 et al,31 1998

Randomization? + + + ? ? + + ? +
Allocation concealment? + + ? ? ? + − ? +
Blinding of participants and personnel? − − − − − − − − −
Blinding of outcome assessment? − − − − − − − − −
Complete outcome data? + + + + + + + + +
No selective reporting? + + + + + + + + +
No other bias? + + + + + − ? + +

+, yes; –, no; ?, unclear.

Smeets et al 809

Figure 2. (A) Forest plot for studies that examined the effect of early vs late start of oral intake for anastomotic leakage. (B)
Sensitivity analysis for studies that examined the effect of early vs late start of oral intake for anastomotic leakage. M-H,
Mantel-Haenszel.

Figure 3. (A) Forest plot for studies that examined the effect of early vs late start of oral intake for overall complications. (B)
Sensitivity analysis for studies that examined the effect of early vs late start of oral intake for overall complications. M-H,
Mantel-Haenszel.
810 Nutrition in Clinical Practice 33(6)

Figure 4. (A) Forest plot for studies that examined the effect of early vs late start of oral intake for length of stay. (B) Sensitivity
analysis for studies that examined the effect of early vs late start of oral intake for length of stay. IV, inverse variance; M-H,
Mantel-Haenszel.

Figure 5. (A) Forest plot for studies that examined the effect of early vs late start of oral intake for mortality. (B) Sensitivity
analysis for studies that examined the effect of early vs late start of oral intake for mortality. M-H, Mantel-Haenszel.
Smeets et al
of GI motility (eg, time to first flatus or defecation) in
multiple studies.23-25,29 However, differences in discharge
criteria between the included studies may have confounded
the effect of early oral intake on LOS, as suggested by the
high statistical heterogeneity.
Despite the attempt to minimize risk of bias by means
of a sensitivity analysis, several other limitations remain
present in the current systematic review. First, perioperative
protocols varied greatly between studies, including the use
of nonsteroid anti-inflammatory drugs and preoperative
bowel lavage. However, while differences in these protocols
can affect various clinical outcomes, the specific effect on
AL may be limited except for nonsteroid anti-inflammatory
drugs.21 Second, no study was blinded; however blinding
is difficult to apply due to the nature of the intervention.
Third, the included studies involved various sites of lower
intestinal surgery; it is well known that the a priori risks of
AL vary in the small bowel, colon, and rectum. However,
in the sensitivity analysis, only studies including colorectal
surgery were included. Last, a clear definition for AL
lacked in most studies and varied between studies when
provided. While this review attempts to summarize best
available evidence, the generalizability remains limited by
the heterogeneity of the included studies.
In conclusion, the results of this systematic review
suggest that early oral intake is associated with a reduction
in LOS, while the effect on anastomotic healing remains
unclear as existing literature is clinically heterogeneous
and at risk of bias. More well-designed, high-quality
randomized studies are needed to further study the potential
benefit of EN, since alternative strategies that reduce AL are
lacking.
Statement of Authorship
B. J. J. Smeets and M. D. P. Luyer equally contributed to the
conception and design of the research; T. J. Weijs contributed
to the design of the research; and B. J. J. Smeets, E. G.
Peters, and E. C. J. Horsten contributed to the acquisition and
analysis of the data and drafted the manuscript. All authors
contributed to the interpretation of the data, critically revised
the manuscript, agree to be fully accountable for ensuring the
integrity and accuracy of the work, and read and approved the
final manuscript.
Supplementary Information
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
References
1. Jannasch O, Klinge T, Otto R, et al. Risk factors, short and long
term outcome of anastomotic leaks in rectal cancer. Oncotarget.
2015;6(34):36884-36893.
2. Kang CY, Halabi WJ, Chaudhry OO, et al. Risk factors for anasto-
motic leakage after anterior resection for rectal cancer. JAMA Surg.
2013;148(1):65-71.
811
3. Krarup PM, Nordholm-Carstensen A, Jorgensen LN, Harling H.
Anastomotic leak increases distant recurrence and long-term mortality
after curative resection for colonic cancer: a nationwide cohort study.
Ann Surg. 2014;259(5):930-938.
4. Mirnezami A, Mirnezami R, Chandrakumaran K, Sasapu K, Sagar
P, Finan P. Increased local recurrence and reduced survival from
colorectal cancer following anastomotic leak: systematic review and
meta-analysis. Ann Surg. 2011;253(5):890-899.
5. Daams F, Luyer M, Lange JF. Colorectal anastomotic leakage: as-
pects of prevention, detection and treatment. World J Gastroenterol.
2013;19(15):2293-2297.
6. Demetriades H, Botsios D, Kazantzidou D, et al. Effect of early
postoperative enteral feeding on the healing of colonic anastomoses
in rats: comparison of three different enteral diets. Eur Surg Res.
1999;31(1):57-63.
7. Guven A, Pehlivan M, Gokpinar I, Gurleyik E, Cam M. Early
glutamine-enriched enteral feeding facilitates colonic anastomosis heal-
ing: light microscopic and immunohistochemical evaluation. Acta
Histochem. 2007;109(2):122-129.
8. Khalili TM, Navarro RA, Middleton Y, Margulies DR. Early postop-
erative enteral feeding increases anastomotic strength in a peritonitis
model. Am J Surg. 2001;182(6):621-624.
9. Fukuzawa J, Terashima H, Ohkohchi N. Early postoperative oral
feeding accelerates upper gastrointestinal anastomotic healing in the
rat model. World J Surg. 2007;31(6):1234-1239.
10. Tadano S, Terashima H, Fukuzawa J, Matsuo R, Ikeda O, Ohkohchi
N. Early postoperative oral intake accelerates upper gastrointestinal
anastomotic healing in the rat model. J Surg Res. 2011;169(2):202-208.
11. Kiyama T, Efron DT, Tantry U, Barbul A. Effect of nutritional
route on colonic anastomotic healing in the rat. J Gastrointest Surg.
1999;3(4):441-446.
12. Kiyama T, Onda M, Tokunaga A, Yoshiyuki T, Barbul A. Effect
of early postoperative feeding on the healing of colonic anastomoses
in the presence of intra-abdominal sepsis in rats. Dis Colon Rectum.
2000;43(10)(suppl):S54-S58.
13. Boelens PG, Heesakkers FF, Luyer MD, et al. Reduction of postop-
erative ileus by early enteral nutrition in patients undergoing major
rectal surgery: prospective, randomized, controlled trial. Ann Surg.
2014;259(4):649-655.
14. van den Heijkant TC, Costes LM, van der Lee DG, et al. Randomized
clinical trial of the effect of gum chewing on postoperative ileus and
inflammation in colorectal surgery. Br J Surg. 2015;102(3):202-211.
15. Andersen HK, Lewis SJ, Thomas S. Early enteral nutrition within
24h of colorectal surgery versus later commencement of feed-
ing for postoperative complications. Cochrane Database Syst Rev.
2006;(4):CD004080.
16. Lewis SJ, Andersen HK, Thomas S. Early enteral nutrition within 24 h
of intestinal surgery versus later commencement of feeding: a system-
atic review and meta-analysis. J Gastrointest Surg. 2009;13(3):569-575.
17. Osland E, Yunus R, Khan S, Memon MA. Early enteral nutrition
within 24 h of intestinal surgery versus later commencement of
feeding: a systematic review and meta-analysis. J Gastrointest Surg.
2009;13(6):1163-1167.
18. Zhuang CL, Ye XZ, Zhang CJ, Dong QT, Chen BC, Yu Z. Early versus
traditional postoperative oral feeding in patients undergoing elective
colorectal surgery: a meta-analysis of randomized clinical trials. Digest
Surg. 2013;30(3):225-232.
19. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews
of Interventions. Version 5.1.0 [updated March 2011]. Chichester,
United Kingdom: The Cochrane Collaboration; 2011. http://www.
cochrane-handbook.org. Accessed February 13, 2017.
20. Moher D, Liberati A, Tetzlaff J, Altman DG ; PRISMA Group.
Preferred reporting items for systematic reviews and meta-analyses: the
PRISMA statement. J Clin Epidemiol. 2009;62(10):1006-1012.
812
21. McDermott FD, Heeney A, Kelly ME, Steele RJ, Carlson GL,
Winter DC. Systematic review of preoperative, intraoperative and
postoperative risk factors for colorectal anastomotic leaks. Br J Surg.
2015;102(5):462-479.
22. Review Manager (RevMan) [Computer program]. Version 5.3. Copen-
hagen, Denmark: The Nordic Cochrane Centre, The Cochrane Collab-
oration; 2014.
23. da Fonseca LM, Profeta da, Luz MM, Lacerda-Filho A, Correia MI,
Gomes da, Silva R. A simplified rehabilitation program for patients
undergoing elective colonic surgery—randomized controlled clinical
trial. Int J Colorectal Dis. 2011;26(5):609-616.
24. Dag A, Colak T, Turkmenoglu O, Gundogdu R, Aydin S. A random-
ized controlled trial evaluating early versus traditional oral feeding after
colorectal surgery. Clinics. 2011;66(12):2001-2005.
25. El Nakeeb A, Fikry A, El Metwally T, et al. Early oral feed-
ing in patients undergoing elective colonic anastomosis. Int J Surg.
2009;7(3):206-209.
26. Hartsell PA, Frazee RC, Harrison JB, Smith RW. Early postoperative
feeding after elective colorectal surgery. Arch Surg. 1997;132(5):518-
521.
27. Lucha PA Jr, Butler R, Plichta J, Francis M. The economic impact of
early enteral feeding in gastrointestinal surgery: a prospective survey of
51 consecutive patients. Am Surg. 2005;71(3):187-190.
Nutrition in Clinical Practice 33(6)
28. Minig L, Biffi R, Zanagnolo V, et al. Early oral versus “traditional”
postoperative feeding in gynecologic oncology patients undergoing
intestinal resection: a randomized controlled trial. Ann Surg Oncol.
2009;16(6):1660-1668.
29. Pragatheeswarane M, Muthukumarassamy R, Kadambari D, Kate V.
Early oral feeding vs. traditional feeding in patients undergoing elective
open bowel surgery—a randomized controlled trial. J Gastrointest
Surg. 2014;18(5):1017-1023.
30. Reissman P, Teoh TA, Cohen SM, Weiss EG, Nogueras JJ, Wexner SD.
Is early oral feeding safe after elective colorectal surgery? A prospective
randomized trial. Ann Surg. 1995;222(1):73-77.
31. Stewart BT, Woods RJ, Collopy BT, Fink RJ, Mackay JR,
Keck JO. Early feeding after elective open colorectal resections:
a prospective randomized trial. Aust N Z J Surg. 1998;68(2):125-
128.
32. Beier-Holgersen R, Boesby S. Influence of postoperative enteral nutri-
tion on postsurgical infections. Gut. 1996;39(6):833-835.
33. Heslin MJ, Latkany L, Leung D, et al. A prospective, randomized
trial of early enteral feeding after resection of upper gastrointestinal
malignancy. Ann Surg. 1997;226(4):567-580.
34. Ortiz H, Armendariz P, Yarnoz C. Is early postoperative feeding
feasible in elective colon and rectal surgery? Int J Colorectal Dis.
1996;11(3):119-121.
Review

Nutrition in Clinical Practice

Volume 33 Number 6
The Prognostic Role of Phase Angle in Advanced Cancer December 2018 813–824

C 2018 American Society for

Patients: A Systematic Review Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10100
wileyonlinelibrary.com

Mayane Marinho Esteves Pereira, BS; Mariana dos Santos Campello Queiroz, BS;
Nathália Masiero Cavalcanti de Albuquerque, BS; Juliana Rodrigues, PhD;
Emanuelly Varea Maria Wiegert, MD; Larissa Calixto-Lima, MD; and Livia Costa de
Oliveira, PhD

Abstract
Phase angle (PA) is a ratio between the reactance and resistance obtained by bioelectric impedance analysis and has been interpreted
as a cell membrane integrity indicator and a predictor of total body cell mass. A low PA may suggest deterioration of the cell
membrane, which in advanced cancer patients may result in a reduced overall survival (OS). This systematic review sought to
investigate the current evidence regarding whether there is an association between PA and OS in patients with advanced cancer (ie,
metastatic disease). The search was conducted on electronic databases in August 2017. A total of 34 articles were identified in the
initial literature search. Nine studies reporting on 1496 patients were deemed eligible according to our inclusion criteria. PA data
were analyzed as continuous variables or according to different cutoffs, under a frequency of 50 Khz. Low PA was associated with
worse nutrition status evaluated by body mass index, serum albumin level, transferrin, and fat-free mass. The median OS of the
included papers varied from 25.5–330 days, and all studies analyzed showed a significant association between PA and OS, in that
patients with low PA had worse OS. Future studies are necessary to justify the use of PA in therapeutic decisions for this population
and to evaluate whether nutrition status can influence the association between PA and survival. (Nutr Clin Pract. 2018;33:813–824)

Keywords
bioelectric impedance; cancer; neoplasms; nutrition assessment; phase angle; survival

Introduction negatively associated with resistance.11 Lower PA suggests

Cancer is recognized as a global public health problem. In
developing countries, at the time of diagnosis, the majority
of patients with cancer have late-stage disease.1 Malnutri-
tion is a frequent manifestation in patients with advanced
cancer and is correlated with poor prognosis and high
mortality.2-5
Bioelectrical impedance analysis (BIA) has been increas-
ingly used to assess nutrition status.6,7 BIA consists of a
rapid, relatively inexpensive, non-invasive, and reproducible
method for providing indirect estimates of the body’s
compositional compartments, as well as the distribution
of fluids in the intracellular and extracellular spaces.6,8
In this context, BIA can be useful in clinical practice to
assess changes in body composition.9 BIA measures the
parameters of body resistance (opposition offered by the
body to the flow of an alternate electrical current) and
reactance (resistive effect produced by the tissue interfaces
and cell membranes).10
The phase angle (PA) is a ratio between reactance and
resistance, and it has been suggested as a marker of cel-
lular function and, consequently, of nutrition status.11 By
definition, PA is positively associated with capacitance and
cell death or decreased cell integrity, while higher PA
suggests large quantities of intact cell membranes. Current
evidence presents PA as an indicator of nutrition status,
a predictive factor for risk of complications and death
in patients suffering from different clinical conditions8
and an independent prognostic factor in advanced cancer
patients.7,12-14 In the present study, we conducted a system-
atic review of the literature (SRL) to investigate the evidence
regarding whether there is an association between PA and
overall survival (OS) in patients with advanced cancer (ie,
metastatic disease).
From the Palliative Care Unit, National Cancer Institute José Alencar
Gomes da Silva (INCA), Rio de Janeiro, RJ, Brazil.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on May 22, 2018.
Corresponding Author:
Larissa Calixto-Lima, MD, 274 Visconde de Santa Isabel Street, Vila
Isabel, 20560-120, Rio de Janeiro, RJ, Brazil; +55-21-991729948.
Email: larissa_calixto@hotmail.com
814 Nutrition in Clinical Practice 33(6)

Table 1. Keywords Used in Search Strategy in Electronic Data Extraction

Databases.
Data extraction tables were specifically developed for this
Electronic Databases Keywords SRL, and the following information was selected indepen-
dently by 2 reviewers: the general characteristics of the stud-
MEDLINE/PubMed ("advanced cancer") AND
ies (first author, year of publication, study design, sample
(https://www.ncbi.nlm. ("electric impedance" OR
nih.gov/pubmed) "phase angle") AND size, country, study aims, and statistical test), participant
("survival" OR characteristics (age and cancer population), information
"outcomes" OR about the BIA measurements (model, current frequency,
"mortality" OR and PA thresholds) and main results about associations
"prognosis") between PA and nutrition status and OS rates.
Scopus (advanced cancer) AND
(https://www.scopus.com) (electric impedance)
AND (survival OR Quality Assessment
outcomes OR mortality
OR prognosis) The quality assessment was performed by 2 independent
LILACS (advanced cancer) AND reviewers using the Newcastle-Ottawa Scale (NOS)16 (Ap-
(http://lilacs.bvsalud.org/) (electric impedance OR pendix 1). The scale consists of 3 quality criteria: selection,
phase angle) AND comparability, and outcome. The maximum score is 9 points
(survival OR outcomes (4 for selection, 2 for comparability, and 3 for outcome).
OR mortality OR Study quality was defined as poor when the score was 1–3,
prognosis)
fair when the score was 4–6, and good when the score was
Cochrane Library (advanced cancer) AND
(http://onlinelibrary. (electric impedance OR 7–9 points.
wiley.com/cochranelibrary/ phase angle) AND Regarding the comparability domain, in addition to age
search) (survival OR outcomes and sex, other confounding factors considered for the sta-
OR mortality OR tistical analysis controls were prognostic evaluation through
prognosis) the Palliative Prognostic Index (PPI) or Palliative Prognostic
(PAP) Score, Karnofsky Performance Score (KPS), and the
Eastern Cooperative Oncology Group (ECOG) Scale of
Performance Status. The score assigned for each paper is
described in Figure 1.
Inter-reviewer reliability was determined using Cohen’s
κ statistics. The interpretation of the coefficient was based
Methods
on the proposal of Shrout.17 The inter-reviewer reliability
Literature Search and Study Selection presented an optimal accuracy (κ = 0.89), representing an
agreement of 92.5%.
A comprehensive search of the literature was conducted ac-
cording to the Preferred Reporting Items for Systematic Re-
views and Meta-Analyses (PRISMA) criteria15 using well- Results
known indexed databases, including MEDLINE/PubMed, Literature Search and Characteristics of the
Scopus, LILACS, and the Cochrane Library in August
2017; a combination of search terms is described in
Included Studies
Table 1. No restrictions were made regarding language or The search resulted in a total of 34 papers. After the
publication date. We selected the studies by the following exclusion for study design, 27 papers were selected for title
inclusion criteria: 1) abstract available on-line, 2) original and abstract review. Subsequently, 18 papers were selected
articles, 3) cohort or case-control design, 4) performed in for full-text review (Figure 2). Papers were excluded by
humans, 5) participants aged ࣙ18 years, and 6) presented the careful reading of titles and abstracts, with reasons for
the relationship between PA and OS. The reference lists of exclusion listed in Table S1. In addition, the examination of
related and included papers were also screened to search for the reference lists of the papers did not recover any further
additional potential studies. studies. Finally, 9 studies reporting on 1496 patients were
Two authors independently reviewed search results. Re- deemed eligible for this SRL.
viewers assessed each title and abstract and considered each The selected papers were all published within the
study for a full-text review. Any disagreements in either last 14 years,7,12,14,18-23 and most were performed
title/abstract or in the full-text paper review phases were on outpatients14,19,21-23 and conducted in the United
resolved by consensus. The opinion of a third reviewer was States,12,14,20-23 Mexico,18,19 and South Korea.7 Five
sought when necessary. studies were prospective,7,12,18-20 and the others were
Pereira et al 815

Figure 1. Evaluation of the methodological quality of selected papers.

Figure 2. Flow chart of studies selection.

816
Table 2. Description of Studies Regarding Authors, Year of Publication, Origin, Age of Participants, Study Design, Sample Size, Cancer Type, Objectives, and Type of
Bioelectrical Impedance Analysis.

Sample
Author/Year Origin Study Design Size Cancer Population Age Objectives BIA

Gupta et al., 200422 USA Retrospective 52 Outpatients, stage IV colorectal 55.8 ± 10.8 Investigate prognostic role BIA-101Q analyzer (RJL
cohort cancer yearsa of PA in advanced Systems, Clinton
colorectal cancer. Township, MI, USA),
single frequency (50
kHz)
Gupta et al., 200423 USA Retrospective 42 Outpatients, stage IV 56.2 ± 10.7 Investigate prognostic role BIA-101Q analyzer (RJL
cohort pancreatic cancer yearsa of PA in advanced Systems, Clinton
pancreatic cancer. Township, MI, USA),
single frequency (50
kHz)
Davis et al., 200920 USA Prospective 50 Inpatients, without defining 63.0 ± 12.0 Determine if BIA correlates (RJL Systems, Clinton
cohort advanced cancer. Cancer yearsa before hydration or Township, MI, USA),
type: pancreatic (12.0%), changes during hydration single frequency (50 kHz)
lung (12.0%), breast (12.0%), and determine if these
myeloma (6.0%), renal changes were
(6.0%), colon (6.0%), gastric prognostically important.
(6.0%).
Gupta et al., 200921 USA Retrospective 165 Outpatients, stages IIIB and IV 56.0 ± 9.1 Investigate prognostic role SFB7 BioImp v1.55
cohort with non-small-cell lung yearsa of BIA-derived PA in analyzer
cancer, treated at Cancer patients with advanced (ImpediMed,
Treatment Centers of non-small-cell lung Brisbane, Australia),
America. cancer. single frequency (50
kHz)
Sánchez-Lara et al., Prospective 119 Outpatients, diagnosis of 60.5 ± 12.5 Evaluate association of Bodystat Quadscan
201219 Mexico cohort histopathologic-confirmed yearsa nutrition parameters in 4000,
or cytologic-confirmed stage health-related quality of multi-frequency
III or IV non-small-cell lung life and OS in patients
cancer. with advanced
non-small-cell lung
cancer.
Lee et al., 20147 South Prospective 28 Inpatients, without defining >70 years Investigate BIA-derived PA Biodynamics model 450
Korea cohort advanced cancer. Cancer = 53.6% as prognostic indicator (Biodynamics Co.,
type: digestive tract (39.2%), 50–70 years for survival in advanced Seattle, WA, USA),
lung (10.7%), hematology = 35.7% cancer patients. single frequency (50
(10.7%), bladder/renal 30–50 years kHz)
(10.7%), and other (28.5%). = 10.7%

(continued)
 Table 2. (continued)

Sample
Author/Year Origin Study Design Size Cancer Population Age Objectives BIA

Hui et al., 201412 USA Prospective 222 Inpatients, without defining 55.0 Determine association of Quantum IV (RJL
cohort advanced cancer. Cancer (22.0–79.0) PA, hand grip strength, Systems, Clinton
type: digestive tract most yearsb and maximal inspiratory Township, MI, USA),
common (33.0%), pressure with OS in single frequency (50
respiratory (16.0%), breast patients with advanced kHz)
(13.0%), gynecologic cancer.
(11.0%), genitourinary
(9.0%), head and neck
(5.0%), hematologic (5.0%),
and other (9.0%).
Hui et al., 201714 USA Retrospective 366 Outpatients with advanced 58.0 Determine association of Inbody 720
cohort cancer defined as locally (21.0–90.0) PA obtained from
advanced, recurrent or yearsb multi-frequency BIA with
metastatic disease for solid OS in patients with
tumors, or progres- advanced cancer.
sive/refractory/incurable
disease for hematologic
tumors. Cancer type:
gastrointestinal (30.0%),
breast (14.0%), head and
neck (13.0%), respiratory
(11.0%), gynecologic (9.0%),
and other (15.0%).
(Inbody, Cerritos,
CA, USA),
multi-frequency
(5, 50, and 250
kHz)
Camargo et al., Prospective 452 Inpatients, without defining 57.6 ± 14.6 Associate PA and OS in Inbody 720
201718 Mexico cohort advanced cancer. Cancer yearsa palliative patients at
type: gastric (39.2%), lung National Cancer Institute
(18.8%), gynecologic of Mexico.
(16.6%), and other (25.4%).
(Inbody, Cerritos,
CA, USA),
single frequency
(50 kHz)

BIA, bioelectrical impedance analysis; OS, overall survival; PA, phase angle.
a Mean (± standard deviation).
b Median (interquartile range).

817
818
retrospective.11,18-20 The age of the population included in
the studies varied between the fifth and sixth decade of life.
Four studies included a population with specific types of
cancer. One study was exclusively with colorectal cancer
patients,22 1 with pancreatic cancer patients,23 and 2 studies
with lung cancer patients.19,21 For studies that included
different cancer types,7,12,14,18,20 digestive tract cancer was
the most common type (Table 2).
Type of BIA Used to Assess PA
Regarding the BIA equipment, different models (Biody-
namics 450, Inbody 720, Bodystat quadscan 400, SFB7
Bioimp v1.55 analyzer, and RJL systems) were used to
assess PA; most were manufactured in the U.S. Only 2
studies used a multi-frequency BIA,14,20 but only 1 study
presented PA data using these different frequencies (5, 50,
and 250 kHz).14 Further studies were conducted at a single
frequency of 50 kHz (Table 2).7,12,18-23
PA Association With Nutrition Status
The association between PA and nutrition status or body
composition was included only in 3 of the selected papers
and evaluated by pre–serum albumin level23 : 1 for body
mass index (BMI),18 1 for subjective global assessment
(SGA),23 and 1 for fat-free mass.12 With the exception
of SGA, these nutrition markers showed a negative and
significant association with PA, indicating that subjects with
worse nutrition status had lower values of PA (Table 3).
PA Association With OS
Survival analyses were performed using Cox proportional
hazards models with bivariate and multivariate logistic
regression in all the studies. Kaplan-Meier curves were
constructed to analyze the probability of OS according to
PA,12,14,18-23 and the log-rank test was used7,12,14,18,20-23 to
verify survival differences between the groups.
Regarding the confounder variables used for controls
in Cox regression analyses, a high variability among the
studies was observed: 6 studies controlled for age,7,14,18-21 3
controlled for gender,14,18,20 and only 1 controlled for race.17
None of the studies controlled for physical activity. The
nutrition status of patients was not considered in 3 of the 9
papers.11,20,21 Some studies controlled for different markers,
such as BMI,7,15,19 SGA,19,22 pre–serum albumin level,23
and transferrin,23 and only 2 of these studies adjusted for
body composition, such as fat-free mass.12,22 Four studies
included information about useful tools for predicting sur-
vival; 3 publications11,19,20 included the ECOG scale, and 17
included the KPS and PPI (Table 3).
Different PA cutoffs were used varying from 4.0°→6.0°,
and in 3 of the studies, no cutoffs were used for the Cox
analysis; the PA was used as a continuous variable.7,12,20 In
Nutrition in Clinical Practice 33(6)
general, patients with a PA ࣘ5.6° presented a significantly
shorter survival time than those with a PA >5.6°.
The median OS of the included papers varied from 25.5–
330 days, and the papers that compared the median OS
among those with low and high PA showed decreased values
in OS for those with low PA (Table 3).12,18-23
It was observed that patients with colorectal cancer
presented a median OS of 8.6 months when the PA was
ࣘ5.6°,22 lung cancer patients presented a median OS of
7.6 months when PA was ࣘ5.4°,16 and in pancreatic cancer
patients, the median OS was 6.3 months among those with
PA ࣘ5.0°.20 For studies that included a population with
different types of cancer, median OS varied from 35–162
days with the previously mentioned cutoffs (Table 3).7,12,14,18
In the Cox analyses, the hazard ratio was approximately
20% higher for each unit increase of PA.7,12,14,20-23 When
different cutoffs were tested, the hazard ratio for mortality
varied between 1.8-fold–10.7-fold risk for death.18,19,22 In
addition, it was observed that all the studies included in this
SRL showed a significant association between PA and OS
(Table 3).
Discussion
This is the first SRL that evaluated the association be-
tween PA and OS in advanced cancer patients. Our results
confirmed that a lower PA value was associated with OS
in advanced cancer patients. These findings suggest PA
as a relevant indicator for unfavorable outcomes, with a
decrement in the survival rate. To determine reliable and
useful prognostic factors that can be used in clinical practice,
the improvement of therapeutic approaches for this type
of population is necessary.24,25 In this context, PA is an
objective and non-invasive method that can be used for
prognosis.
Patients with advanced cancer may present a PA lower
than those observed in a healthy population.26 This reduced
PA may be related to the metabolic disarrangement that
this population experiences, such as malnutrition and cancer
cachexia.27 In fact, a poor nutrition status can result in
a body fluid imbalance and cell membrane changes.12,27,28
Thus, the relationship of PA with body composition and
nutrition status makes it an indicator for the predictive risk
of morbidity and mortality.12
In this context, a significant association was found be-
tween PA and the nutrition markers such as pre–serum albu-
min level, BMI, and fat-free mass. Pre–serum albumin level
is a reliable indicator of acute changes in nutrition status,
and it is unaffected by hydration status. On the other hand,
the use of BMI has been vastly discussed in the literature
as a nutrition marker that does not take into account the
evaluation of fat-free mass and body fat. Hui et al. tested
the association between PA and fat-free mass, and found
a positive and significant association, emphasizing the idea
 Table 3. Description of Statistical Tests, Cutoffs for Phase Angle, Survival, Potential Confounding, and Main Findings Applied in Studies.

Association
Cutoff for PA (50 Potential Between PA and Association Between PA and
Author/Year Statistical Test kHz) Confounding Nutrition Status OS Main Findings

Gupta et al., Cox regression; 5.57° Diagnostic time; – Patients with PA ࣘ5.6° had PA and age at diagnosis presented
200422 Kaplan-Meier; body weight; median survival of 8.6 positive correlation with OS.
log-rank test. lean mass; months, while others Furthermore, SGA was not
serum albumin presented median OS of 40.4 associated with OS. PA was
level; pre–serum months. In multivariate prognostic indicator in patients
albumin level; analysis, PA (ࣘ5.6°) was a with advanced colorectal cancer.
transferrin; prognostic indicator in The risk was approximately 11-fold
SGA; stage of patients with advanced higher for those with PA ࣘ5.6°.
tumor; prior colorectal cancer (HR =
treatment. 10.7; 95% CI: 1.9–60.2;
P-value = 0.007).
Gupta et al., Spearman 5.0° Pre–serum PA was associated Patients with PA ࣘ5.0° had PA associated positively with OS.
200423 correlation; albumin level; with pre–serum median OS of 6.3 months, Pre–serum albumin level
Pearson serum albumin albumin level (r while others presented a (<150mg/L) was also associated
correlation; Cox level; prior = 0.32; P = median OS of 10.2 months with OS. PA was strong prognostic
regression; treatment. 0.04), (P-value = 0.02). indicator in advanced pancreatic
Kaplan-Meier; transferrin (r = Multivariate analysis after cancer patients. Therefore, each
log-rank test. 0.19; P = 0.25) adjusting for pre–serum unit increase on PA showed
and SGA (r = albumin level demonstrated reduction of mortality of 31%.
-0.26; P = 0.10). relative risk reduction (=
0.69; 95% CI: 0.49; 0·97) that
was associated with each unit
increase in PA. PA continued
to be statistically significant
(P-value = 0.02) after jointly
controlling for serum
albumin level, pre–serum
albumin level, and previous
treatment history.

(continued)

819
820
Table 3. (continued)

Association
Cutoff for PA (50 Potential Between PA and Association Between PA and
Author/Year Statistical Test kHz) Confounding Nutrition Status OS Main Findings

Davis et al., Spearman Survival Resistance; total – Higher PA before hydration Weight loss was associated with
200920 correlation; curves: body fluid; predicted a borderline and shorter OS. PA positively predicted
Pearson Decreased or sodium; gender; significantly lower OS (HR borderline and significantly lower
correlation; Cox not altered age; skin color; = 0.8; 95% CI: 0.6–1.0; OS in advanced cancer patients on
regression; vs increased ECOG. P-value = 0.057); first day of their hospital stay,
Kaplan-Meier; PA paradoxically, an increase in indicating that each unit increase
log-rank test. Regression: PA during hydration on PA presented reduction of 20%
Continuous predicted lower OS (HR = in mortality.
variable 1.2; 95% CI: 1.1–1.4; P-value
= 0.007).
Gupta et al., Cox regression; 5.3° Age; stage of – Patients with PA ࣘ5.3 had Stage at diagnosis and treatment
200921 Kaplan-Meier; tumor; prior median OS of 7.6 months, history was associated with OS.
log-rank test. treatment. while those >5.3 had 12.4 BIA-derived PA was positive and
months (P = 0.02). In independently prognostic indicator
multivariate analysis, each of OS in patients with stage IIIB
degree of increase in PA, and IV non-small-cell lung cancer.
adjusted for age, stage at Nutrition interventions targeted at
diagnosis, and prior improving PA could potentially
treatment, history was lead to improved OS in patients
associated with OS (HR = with advanced non-small-cell lung
0.8; 95% CI: 0.6–1.0; P-value cancer. Therefore, each unit
= 0.020). increase on PA presented reduction
of 20% in mortality.
Sánchez-Lara Mann-Whitney U 5.8° Gender; age; – Median follow-up was 6 (± 5) Malnutrition measured by SGA,
et al., 201219 test; Cox ECOG; stage of months. In multivariate weight loss >10%, and BMI >20
regression; tumor; weight analysis, patients with was associated with lower related
Kaplan-Meier. loss; BMI; ECOG = 2 (HR = 2.7; 95% quality of life. Patients with ECOG
SGA; serum CI: 1.5–4.7; P-value = = 2, high content serum IL-6, and
albumin level; 0.001), PA ࣘ5.8° (HR = 3.0; lower PA showed lower OS. PA was
PLR; NLR; 95% CI: 1.2–7.1; P-value = presented to be an independent
CRP; IL-6; 0.011) and malnutrition (HR prognostic factor in advanced
TNF-α. = 2.7; 95% CI: 1.3–5.5; non-small-cell lung cancer, where
P-value = 0.005) had worse those with PA ࣘ5.8° presented risk
OS. 3-fold higher than those with
higher values.

(continued)
 Table 3. (continued)

Association
Cutoff for PA (50 Potential Between PA and Association Between PA and
Author/Year Statistical Test kHz) Confounding Nutrition Status OS Main Findings

Lee et al., Spearman Survival Age; BMI; PPI; – Median OS time was 25.5 days. PA showed positive and significant
20147 correlation; Cox curves: KPS. PA ࣙ4.4° showed a longer association with OS, where each
regression; 4.4° survival time compared with unit increase on PA presented a
log-rank test. Regression: PA <4.4° (100 vs 125 days, reduction of 36% in mortality.
Continuous respectively; P-value = 0.01). However, age, PPS, and BMI did
variable In multivariate analysis, each not show significant relationship
degree of increase in PA, with OS. BIA-derived PA may
adjusted for age, PPI, BMI, serve as independent prognostic
and PA was associated with indicator in advanced cancer
OS (HR = 0.64; 95% CI: patients.
0.42–0.95; P-value = 0.028).
Hui et al., Spearman Survival PAP Score; serum PA was associated Median OS was 106 days (95% PA was positively associated with OS,
201412 correlation; Cox curves: albumin level; with fat free CI: 71–128 days). Patients where each unit increase on PA
regression; 4.4° vs >4.4° fat-free mass. mass (r = 0.29; with lower PA (<3°) had presented a reduction of 14% in
Kaplan-Meier; 2.0–2.9° vs P <0.001) and lower survival (35 vs 220 mortality. Furthermore, patients
log-rank test. 3.0–3.9° vs the fat free mass days) when compared with with lower KPS, PPS,
4.0–4.9° vs index (r = 0.33; those with higher values (PA hypoalbuminemia (<3.0 g/dL), and
5.0–5.9° vs P <0.001). ࣙ6.0°). In multivariate lower fat-free mass (ࣘ51.6 kg) had
ࣙ6.0° analysis, each degree of poor survival. PA was shown to be
<P5 vs >P95 increase in PA was an independently established
Regression: associated with OS (HR = prognostic factor in advanced
Continuous 0.86; 95% CI: 0.74–0.99; cancer setting during hospital stay.
variable P-value = 0.04).
Hui et al., Spearman Survival Gender; age; race; – Median OS was 250 days. PA was positively associated with OS,
201711 correlation; curves: tumor type; Patients with PA ࣘ4.5° where each unit increase on PA
Contal & Right side of ECOG; weight showed a significant lower presented reduction of 15% in
O’Quigley; Cox body: 4.5° loss in 6 median of OS (162 days) mortality. In addition,
regression; Left side of months; serum when compared with those hypoalbuminemia (<4.0g/dL)
Kaplan-Meier; body: 4.3° albumin level; with higher (>4.5°) values elevated lactate dehydrogenase
log-rank test. Regression: calcium; (403 days). In multivariate (>618 unit), serum albumin levels,
Right side of hemoglobin; analysis, each degree increase and neutrophil-lymphocyte ratio
body: 4.4° leukocytes; in PA was associated with (>3) were correlated with poor
Left side of lactate; OS (HR = 0.85; 95% CI: survival. PA represented novel
body: 4.2° lymphocytes; 0.72–0.99; P-value = 0.048). objective prognostic factor in an
neutrophils. outpatient palliative cancer-care
setting, regardless of frequency and
body sides.

(continued)

821
822
Table 3. (continued)

Association
Cutoff for PA (50 Potential Between PA and Association Between PA and
Author/Year Statistical Test kHz) Confounding Nutrition Status OS Main Findings

Camargo et al., Spearman 4.0° Gender; age; Patients with BMI Average survival was 86 vs 163 PA and BMI were positively
201715 correlation; Cox diagnose; BMI. <24.9kg/m² had days for patients with PA correlated with OS. Therefore, risk
regression; significantly ࣘ4.0° and >4.0°, for mortality was approximately
Kaplan-Meier; lower mean PA respectively (P >0.0001). In 2-fold higher for those with PA
log-rank test. (3.73° ±0.92) the multivariate analysis, ࣘ4.0°.
than those with patients with PA ࣘ4.0° had
BMI ࣙ 25 kg/m² HR = 1.8; 95% CI: 1.4–2.3;
(4.37° ±0.98; P P-value >0.0001.
= 0.011).

BIA, bioelectrical impedance analysis; BMI, body mass index; CI, confidence interval; CRP, C-reactive protein; ECOG, Eastern Cooperative Oncology Group; HR, hazard ratio; IL-6,
interleukin 6; KPS, Karnofsky Performance Status; NLR, neutrophils/lymphocytes ratio; OS, overall survival; PA, phase angle; PAP Score, Palliative Prognostic Score; PLR,
platelets/lymphocytes ratio; PPI, Palliative Prognostic Index; PPS, Palliative Performance Scale; SGA, subjective global assessment; TNF-α, tumor necrosis factor-α; vs, versus.
Pereira et al
that body composition has an important influence on PA
values.12
It is worth noting that despite all the papers included in
this SRL that assessed PA by using the BIA, most studies
did not evaluate body composition in their study popu-
lation. Based on the assumption that body composition
may influence OS, it is reasonable to say that controlling
for this potential confounder in these analyses is necessary.
However, most of the studies controlled for at least 1 marker
of nutrition status in the Cox regression.
Other potential confounders used in the Cox analyses in
these studies should be discussed. The first is performance
status (KPS) or prognostics scales (PPI and PAP score),
which are useful tools for predicting survival of palliative
subjects.13 Only 4 studies, however, included this informa-
tion. Nevertheless, these confounders were considered in
the evaluation of quality assessment in this SRL. Other
confounders that should be included in analyses that are
influences on PA values are ethnicity/race, gender, age,
weight and height.12,29,30 Therefore, these aspects should
be considered in the study design when involving mea-
surements of PA because these confounders may lead to a
prejudice in the associations between PA and OS.
Another factor that may also play a role in these sta-
tistical analyses could be the population included in the
studies; some studies included subjects with different cancer
types, and no statistic control was observed for cancer
type. The lack of control for this variable might influence
the PA analysis and its association with OS. Thus, further
investigation regarding this issue is necessary.
Another important issue is related to the different PA
cutoffs in the analyses of the studies. Currently, there is
no known precise PA value capable to identify the length
of survival in individuals with advanced cancer. In general,
the mean, median, or lower PA quartiles found in the
groups studied were used as a parameter. However, it can
be affirmed that sick individuals present lower values of PA,
which correlates with the severity of the disease.7,12,14,18-23
In all the studies, the data were collected in a sin-
gle care center and were related to inpatient and/or out-
patient follow-ups.7,12,14,18-23 It is worth mentioning that
survival may differ substantially between inpatients and
outpatients.14 Inpatients present specific features that de-
mand specialized treatment, including better symptom con-
trol, which could limit the generalization of these findings.
The use of different methods to evaluate PA was influ-
enced by the BIA frequency that was used to evaluate it,
i.e., single-frequency or multi-frequency.31,32 In this SRL,
different methods of BIA were used in the included studies,
while only 2 studies used a multi-frequency BIA.14,19 How-
ever, only 1 of these studies showed results that were related
to different frequencies in patients with advanced cancer
in palliative care; this study was the first with differential
analyses relating PA and OS.14 Lower frequencies of PA
823
(<50 kHz) are associated with a flow through the extra-
cellular compartment, whereas higher frequency currents
(>200 kHz) penetrate the cell membranes and pass through
thin tissues.33 This differential tissue penetration at a higher
frequency current allows for better accuracy in assessing
the body’s composition in healthy individuals.34 Still, more
studies are needed to evaluate the usefulness of different
frequencies in the evaluation of PA in advanced cancer
patients.
With regard to strengths, the SRL methodology was
consistent with the PRISMA statement.15 The search was
conducted in different databases and in the reference lists of
the included papers; there were also no restrictions made for
language, the year of publication, or the place of execution.
Conclusions
Considering that survival prediction remains a challenge in
individuals with advanced cancer, the results of this SRL
show that PA, derived from BIA, is an important objective
predictive factor of OS for this population. Nonetheless,
there is still a lack of information regarding the use of
thresholds in this population. Future studies are necessary
to identify cutoffs of PA to guide therapeutic decisions
and to evaluate whether nutrition status can influence the
association between PA and survival. While no specific
threshold can be used as a prognostic factor in this popu-
lation, the reduction in values of PA, evaluated routinely,
could indicate a worse OS.
Statement of Authorship
L. Calixto-Lima and L.C. Oliveira equally contributed to
the conception and design of the research; L.C. Oliveira,
M.M.E. Pereira, M.S.C. Queiroz, E.M. Wiegert, and J.
Rodrigues contributed to the acquisition, analysis, and in-
terpretation of the data; L.C. Oliveira and L. Calixto-Lima
drafted the manuscript; M.M.E. Pereira, M.S.C. Queiroz,
N.M.C. de Albuquerque, J. Rodrigues, E.M. Wiegert,
L. Calixto-Lima, and L.C. Oliveira critically revised the
manuscript; M.M.E. Pereira, M.S.C. Queiroz, N.M.C. de
Albuquerque, J. Rodrigues, E.M. Wiegert, L. Calixto-Lima,
and L.C Oliveira agree to be fully accountable for ensuring
the integrity and accuracy of the work. All authors read and
approved the final manuscript.
Supplementary Information
Additional supporting information may be found online in the
Supporting Information section at the end of the article.
References
1. WHO. World Health Organization. National Cancer Control Programs:
Policies and Managerial Guidelines. 2nd ed. Geneva: World Health
Organization; 2002.
824
2. Gangadharam A, Choi SE, Hassan A, et al. Protein calorie malnutri-
tion, nutritional intervention and personalized cancer care. Oncotarget.
2017;8(14):24009-24030.
3. Aktas A, Walsh D, Galang M, et al. Under-recognition of malnutrition
in advanced cancer: the role of the dietitian and clinical practice
variations. Am J Hosp Palliat Care. 2017;34(6):547-555.
4. Arends J, Bachmann P, Baracos V, et al. ESPEN guidelines on
nutrition in cancer patients. Clin Nutr. 2017;36(1):11-14.
5. Santarpia L, Marra M, Montagnese C, Alfonsi L, Pasanisi F, Contaldo
F. Prognostic significance of bioelectrical impedance phase angle in
advanced cancer: Preliminary observations. Nutrition. 2009;25(9):930-
931.
6. Gupta D, Lis CC, Dahlk SL, et al. The relationship between bio-
electrical impedance phase angle and subjective global assessment in
advanced colorectal cancer. Nutr J. 2008;7(19):1-6.
7. Lee SY, Lee YJ, Yang J, Kim C, Choi W. The association between
phase angle of bioelectrical impedance analysis and survival time
in advanced cancer patients: preliminary study. Korean J Fam Med.
2014;35(5):251-256.
8. Llames L, Baldomero V, Iglesias ML, Rodota LP. Values of the phase
angle by bioelectrical impedance; nutritional status and prognostic
value. Nutr Hosp. 2013;28(2): 286–295
9. Sarhill N, Mahmoud FA, Christie R, Tahir A. Assessment of nutri-
tional status and fluid deficits in advanced cancer. Am J Hosp Palliat
Care. 2003,20:465-473.
10. Zarowitz BJ, Pilla AM. Bioelectrical impedance in clinical practice.
DICP. 1989; 23:548-555.
11. Baumgartner RN, Chumlea WC, Roche AF. Bioelectric impedance
phase angle and body composition. Am J Clin Nutr. 1988,48:
16-23
12. Hui D, Bansal S, Morgado M, Dev R, Chisholm G, Bruera E. Phase
angle for prognostication of survival in patients with advanced cancer:
preliminary findings. Cancer. 2014;120(14):2207-2214.
13. Hui D. Prognostication of survival in patients with advanced
cancer: predicting the unpredictable? Cancer Control. 2015;22(4):
489-497.
14. Hui D, Dev R, Pimentel L, et al. Association between multi-frequency
phase angle and survival in patients with advanced cancer. J Pain
Symptom Manage. 2017;53(3):571-577.
15. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group.
Preferred reporting items for systematic reviews and meta-
analyses: The PRISMA statement. J Clin Epidemiol. 2009;62(10):
1006-1012.
16. Wells GA, Shea B, O´Connell D, et al. The Newcastle-Ottawa Scale
(NOS) for assessing the quality of non-randomized studies in meta-
analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.
asp. Accessed March 15, 2017.
17. Shrout PE. Measurement reliability and agreement in psychiatry. Stat
Methods Med Res. 1998;7:301-317.
18. Camargo DAP, Pérez SRA, Franco MMR, Licona NEA, Ceniceros
VIU, Baldenegro LEF. Phase angle of bioelectrical impedance analysis
as prognostic factor in palliative care patients at the national cancer
institute in Mexico. Nutr Cancer. 2017;69(4):601-606.
Nutrition in Clinical Practice 33(6)
19. Sánchez-Lara K, Turcott JG, Juárez E, et al. Association of nutrition
parameters including bioelectrical impedance and systemic inflam-
matory response with quality of life and prognosis in patients with
advanced non-small-cell lung cancer: a prospective study. Nutr Cancer.
2012;64(4);526-534.
20. Davis MP, Yavuzsen T, Khoshknabi D, et al. Bioelectrical impedance
phase angle changes during hydration and prognosis in advanced
cancer. Am J Hosp Palliat Care. 2009;26(3):180-187.
21. Gupta D, Lammersfeld CA, Vashi PG, et al. Bioelectrical impedance
phase angle in clinical practice: implications for prognosis in stage IIIB
and IV non-small cell lung cancer. BMC Cancer 2009;9(37):1-6.
22. Gupta D, Lammersfeld CA, Burrows JL, et al. Bioelectrical impedance
phase angle in clinical practice: implications for prognosis in advanced
colorectal cancer. Am J Clin Nutr. 2004;80(6):1634-1638.
23. Gupta D, Lis CG, Dahlk SL, Vashi PG, Grutsch JF, Lammersfeld
CA. Bioelectrical impedance phase angle as a prognostic indicator in
advanced pancreatic cancer. Br J Nutr. 2004;92(6):957-962.
24. Krishnan M, Temel JS, Wright AA, Bernacki R, Selvaggi K, Balboni T.
Predicting life expectancy in patients with advanced incurable cancer:
a review. J Support Oncol. 2013;11(2):68-74.
25. Partridge M, Fallon M, Bray C, McMillan D, Brown D, Laird B. Prog-
nostication in advanced cancer: a study examining an inflammation-
based score. J Pain Symptom Manage. 2012;44(2):161-167.
26. Mulasi U, Kuchnia AJ, Cole AJ, Earthman CP. Bioimpedance at the
bedside: current applications, limitations, and opportunities. Nutr Clin
Pract. 2015;30:180-193.
27. Grundmann O, Yoon SL, Williams JJ. The value of bioelectrical
impedance analysis and phase angle in the evaluation of malnutrition
and quality of life in cancer patients – a comprehensive review. Eur
J Clin Nutr. 2015;69(12):1290-1297.
28. Norman K, Stobaus N, Zocher D, et al. Cutoff percentiles of bioelec-
trical phase angle predict functionality, quality of life, and mortality in
patients with cancer. Am J Clin Nutr. 2010:92(3):612-619.
29. Barbosa-Silva MCG, Barros AJD, Wang J, Heymsfield SB, Pierson
RN Jr. Bioelectrical impedance analysis: population reference values
for phase angle by age and sex. Am J Clin Nutr. 2005;82(1):49-52.
30. Gonzalez MC, Heymsfield SB. Bioelectrical impedance analysis for
diagnosing sarcopenia and cachexia: what are we really estimating?
J Cachexia Sarcopenia Muscle. 2017;8(2):187-189.
31. Kyle UG, Bosaeus I, De Lorenzo AD, et al. Bioelectrical impedance
analysis-part I: review of principles and methods. Clin Nutr.
2004;23(5):1226-1243.
32. Kyle UG, Bosaeus I, De Lorenzo AD, et al. Bioelectrical impedance
analysis-part II: utilization in clinical practice. Clin Nutr. 2004;
23(6):1430-1453.
33. Utter AC, Nieman DC, Mulford GJ, et al. Evaluation of leg-to-leg BIA
in assessing body composition of high-school wrestlers. Med Sci Sports
Exerc. 2005;37(8):1395-1400.
34. Gaba A, Kapus O, Cuberek R, Botek M. Comparison of multi-
and single-frequency bioelectrical impedance analysis with dual-energy
X-ray absorptiometry for assessment of body composition in post-
menopausal women: effects of body mass index and accelerometer-
determined physical activity. J Hum Nutr Diet. 2015;28(4):390-400.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Effect of Vitamin D Level on Clinical Outcomes in Patients December 2018 825–830

C 2018 American Society for

Undergoing Left Ventricular Assist Device Implantation Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10078
wileyonlinelibrary.com

Fadi Abou Obeid, MD1 ; Gardner Yost, MS2 ; Geetha Bhat, PhD, MD2 ;
3 4
Erin Drever, MD ; and Antone Tatooles, MD

Abstract
Background: Vitamin D is an important hormone that regulates cardiac myocyte function. Low levels contribute to the
development of cardiovascular disease and have been implicated in immune function and the inflammatory cascade. Patients who
undergo left ventricular assist device (LVAD) implantation are at risk for driveline infection, stroke, and gastrointestinal (GI)
bleeding. We investigated whether serum 25-hydroxy (25-OH) vitamin D levels affect clinical outcomes after LVAD. Methods:
212 patients who underwent LVAD implantation between 2010 and 2015 were included. We measured preoperative 25-OH
vitamin D level and postoperative adverse events during the first year. Vitamin D level was classified into 3 categories: normal
(>30 ng/mL), insufficient (20–30 ng/mL), and deficient (<20 ng/mL). Clinical outcomes in both insufficient and deficient categories
were compared with the normal category. Results: The odds ratio (OR) of being admitted ࣙ2 times was 2.46 (95% confidence
interval [CI]: 1.067–5.769) for deficiency and 2.5 (95% CI: 0.970–6.443) for insufficiency. The OR of driveline infection was 6.185
(95% CI: 0.80–49.2; P = .07) for insufficiency and 11.467 (95% CI: 1.204–109.26; P = .03) for deficiency. Vitamin D levels were
not associated with GI bleeding, length of stay, or stroke. Conclusions: In patients with LVAD, both deficiency and insufficiency
of 25-OH vitamin D levels are independently associated with increased postoperative driveline infection risk and higher rate of
readmission. Further trials are needed to confirm whether a repletion regimen could be a promising means of decreasing the risk
for these postoperative adverse events. (Nutr Clin Pract. 2018;33:825–830)

Keywords
heart assist devices; heart failure; vitamin D; vitamin D deficiency

Introduction is associated with a doubled risk for stroke compared with

Vitamin D is a steroidal hormone that functions to maintain
bone health and calcium homeostasis. Recently, vitamin D
has been found to optimize the function of multiple organs,
including the heart.1
Vitamin D deficiency is very prevalent in patients with
congestive heart failure, and may contribute to the devel-
opment of heart failure and other cardiovascular diseases
by activating the renin-angiotensin-aldosterone system, in-
ducing chronic inflammation, and causing endothelial dam-
age that can lead to hypertension, obesity, and diabetes
mellitus.2,3 At present, some data suggest that vitamin D
deficiency is associated with a poor prognosis and low
survival rate among patients with chronic heart failure
(CHF).4
Liu et al5 reported that, among 548 patients, lower
vitamin D levels were associated with higher levels of brain
natriuretic peptide (BNP) and higher plasma renin activity,
as well as increased CHF hospitalization and overall greater
incidence of all-cause mortality. Vitamin D deficiency is
also associated with a greater stroke incidence. A recent
study of healthy adults indicated that vitamin D deficiency
those with normal levels.6 This increase in risk may be
related to the role of vitamin D in regulation of neu-
ronal death, as well as its function as an antithrombotic
agent.
Vitamin D has important roles in regulation of the
immune system, influencing proliferation and differentia-
tion of B and T cells and immunoglobulin production.7
From the 1 Department of Cardiology, Advocate Christ Medical
Center, Oak Lawn, IL, USA; 2 Center for Heart Transplant and Assist
Devices, Heart Institute, Advocate Christ Medical Center, Oak Lawn,
IL, USA; 3 Department of Endocrinology, Advocate Christ Medical
Center, Oak Lawn, IL, USA; and 4 Department of Cardiothoracic
Surgery, Advocate Christ Medical Center, Oak Lawn, IL, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on March 30, 2018.
Corresponding Author:
Geetha Bhat, PhD, MD, Medical Director, Center for Heart
Transplant and Assist Devices, Heart Institute, Advocate Christ
Medical Center, 4440 W 95th Street, OPP 6409P, Oak Lawn, IL 60453.
Email: geetha.bhat@advocatehealth.com
826
In addition, vitamin D may have an important role in
moderating the severity of bloodstream infections and
sepsis by binding to and inhibiting bacterial lipopolysac-
charides, and by reducing the sepsis cascade in fungal
infection by controlling the production of proinflammatory
cytokines.8,9
These important functions of vitamin D are particularly
germane to patients with advanced heart failure who un-
dergo left ventricular assist device (LVAD) implantation.
Candidates for LVAD implantation are patients with end-
stage heart failure who are at high risk for vitamin D deple-
tion and concomitant neurohormonal dysregulation.2-5,10
Further, after implantation of an LVAD, patients are at
increased risk for infection of the surgical device pocket,
driveline exit site, and bloodstream. Recognizing the role
of vitamin D in regulation of neurohormonal activation
and immune-related processes, we sought to evaluate the
relationship of pre-LVAD vitamin D levels with postop-
erative clinical outcomes including length of stay after
LVAD, readmission rate, driveline infection, stroke, and
gastrointestinal (GI) bleeding.
Methods
Study Design
We conducted a retrospective chart review of 212 consecu-
tive patients who underwent implantation of a continuous
flow LVAD between January 1, 2010, and January 1, 2015.
This study was approved by a local ethics committee and
conforms with the principles outlined in the Declaration of
Helsinki.
Participants
Patients ࣙ18 years of age who underwent LVAD
implantation between January 1, 2010, and January 1,
2015, were included. Patients who required extracorporeal
membrane oxygenation or emergent biventricular support
were excluded from this study.
Data Collection
The study included a total of 212 patients who were eligible
by inclusion and exclusion criteria. Data from eligible pa-
tients’ charts were collected from electronic medical records
beginning from the time of each patient’s consent for LVAD
implantation until 1 year after LVAD implantation.
Demographic features including age, sex, race, and type
of device were collected. 25-Hydroxy-vitamin D (25-OH
vitamin D) level was collected within a week of implantation
or within the last hospitalization for pre-LVAD workup.
Other laboratory tests and biomarkers such as creatinine,
blood urea nitrogen, calcium, serum albumin, and BNP
Nutrition in Clinical Practice 33(6)
were collected before LVAD implantation. Vitamin D levels
were not adjusted for shifts in fluid status or administration
of intravenous (IV) fluids. Clinical outcomes of interest
were collected for a 1-year period after LVAD implantation.
The outcome measures included length of stay after LVAD,
readmission rate, driveline infection, stroke, and GI bleeding
events.
Outcome Measure Stratification and Statistical
Analysis
Postoperative length of stay was categorized as <3 weeks
of postoperative hospital stay or >3 weeks based on mean
length of stay for LVAD implantation, as established by
Hasin et al.11
The number of total readmissions in the first year after
LVAD implantation was recorded. A recent article that
analyzed readmissions after LVAD implantation found that
the overall readmission rate was 1.64 ± 1.97 during a 1.4 ±
0.9-year follow-up period.12 In this study, we thus divided
the number of readmissions within a year after LVAD into
2 categories: 0–1 admission and ࣙ2 admissions.
Driveline infection was defined as any event identified as
exit site driveline infection in postoperative year 1. Stroke
events included both ischemic and hemorrhagic stroke, evi-
dent clinically and radiologically, observed in the first year
after LVAD implantation. GI bleeding events were defined
as any unexpected decline in hemoglobin level with evidence
of melena, hematemesis, or hematochezia, regardless of the
source.
We conducted all analyses using bivariate logistic regres-
sion and calculated odds ratio (OR) with 95% confidence
interval (CI). We made no imputation for missing values.
The main independent variable, serum 25-OH vitamin D,
was categorized, as defined by the Endocrine Society, into
3 categories: normal 25-OH vitamin D with a level >30
ng/mL, insufficient 25-OH vitamin D with a level ranging
between 20 and 30 ng/mL, and deficient 25-OH vitamin D
with a level <20 ng/mL.13 Dependent variables and outcome
measures were compared across the groups with insufficient
and deficient vitamin D levels. The model was adjusted for
baseline characteristics that differed between the deficient,
insufficient, and normal vitamin D level groups with P < .05
by univariate analysis. Statistical operations were performed
using SAS 9.4 (SAS, Cary, NC, USA).
Results
From January 1, 2010, through January 1, 2015, 212 pa-
tients with LVAD were eligible for chart review. Of these
patients, 29 were excluded from analysis because of missing
preoperative 25-OH vitamin D levels. Data on the outcomes
measures were available for the entire population.
Obeid et al 827

Table 1. Demographics for the Study Population (N = 183). insufficient in 51 (27.87%), and deficient in 100 (54.64%).
Although patients with vitamin D deficiency were more
Demographics n %
likely to be African American than white, there were no
Race other significant differences in preoperative demographics
White 95 51.91 or laboratory values between the 3 groups (Table 3).
Black 75 40.98 A total of 102 (55.7%) patients remained in the hos-
Hispanic 7 3.82 pital >3 weeks after LVAD. The frequency of having ࣙ2
Asian 1 0.54 readmissions within the first year was higher in those pa-
Other 5 2.73 tients with vitamin D insufficiency or deficiency. In patients
Sex
Female 46 24.14
with normal vitamin D levels preoperatively, 50% had
Male 137 75.86 ࣙ2 admissions in the first year after LVAD implantation,
Age (Mean ± SD) 58.84 ± 13.43 compared with 70.6% in the insufficient and 71.0% in
Type of device the deficient groups, respectively (P = .0737) (Table 4). A
HeartMate II 146 79.79 logistic regression model determined that the risk for >2
HeartWare HVAD 37 20.21 readmissions in the first postoperative year was 2.46 times
higher for the vitamin D–deficient group and 2.5 times
higher for the vitamin D–insufficient group compared with
Table 2. Preoperative Laboratory Values for the Study the normal vitamin D group (adjusted OR: 2.46, 95% CI:
Population (N = 183). 1.067–5.769, and adjusted OR: 2.5, 95% CI: 0.97–6.433,
Laboratory Tests n Mean SD respectively).
Of the 183 patients in the study, 31 (16.9%) patients had
Vitamin D (ng/mL) 183 20.33 11.70 at least 1 episode of driveline infection in the first year
Parathyroid hormone (IU) 83 18.81 8.25 after LVAD. Of these 31 patients, 17 of 100 (17.2%) were
Calcium (mEq/L) 183 8.99 4.05 in the vitamin D–deficient category, 13 of 51 (26%) in the
Brain natriuretic peptide (pg/mL) 183 771.01 816.65
insufficient category, and only 1 of 32 (3.1%) in the normal
Serum albumin (g/dL) 183 2.89 0.52
Thyroid-stimulating hormone (mIU/mL) 183 3.88 8.77 category. A significantly larger percentage of patients who
Total bilirubin (mg/dL) 183 1.73 6.41 were vitamin D deficient had driveline infections compared
with those who had a normal vitamin D level (17.2% vs
3.1%, P = .027, respectively; Table 4). A logistic regression
model indicated that patients with vitamin D deficiency
Baseline characteristics for the study population are were at an 11.47 times greater risk for development of
summarized in Tables 1 and 2. The population was divided driveline infection compared with patients with normal
into 3 groups based on their 25-OH vitamin D level. 25- vitamin D level (95% CI: 1.204–109.26, P = .03), and that
OH vitamin D level was normal in 32 patients (17.49%), patients with vitamin D insufficiency were at 6.185 times

Table 3. Comparison of Demographic and Laboratory Data for Patients with Normal, Insufficient, and Deficient Preoperative
Vitamin D.

Preoperative Vitamin D

Demographics Normal (n = 32) Insufficient (n = 51) Deficient (n = 100) P Value

Race, n (%) .036

White 22 (22.68) 33 (34.00) 42 (43.30)
Black 10 (13.16) 16 (21.05) 50 (65.79)
Male sex, n 24 36 76 .768
Age, y 64.62 ± 10.35 59.22 ± 12.18 58.4 ± 13.53 .0532
White cell count, 1000/μL 7.441 ± 2.92 8.09 ± 3.00 8.50 ± 3.98 .297
Neutrophils, 1000/μL 5.17 ± 2.13 5.77 ± 2.55 5.89 ± 3.59 .504
Lymphocytes, 1000/μL 1.41 ± 1.49 1.52 ± 0.68 1.27 ± 0.67 .207
Sodium, mg/dL 135.13 ± 3.83 135.59 ± 3.42 134.77 ± 4.59 .240
Creatinine, mg/dL 1.74 ± 1.58 1.44 ± 0.42 1.64 ± 2.51 .551
Serum calcium, mEq/L 8.76 ± 0.40 8.73 ± 0.40 8.76 ± 68 .964
Serum albumin, g/dL 2.91 ± 0.51 2.98 ± 0.51 2.85 ± 0.50 .354
Total bilirubin, mg/dL 1.30 ± 1.17 1.12 ± 1.00 1.24 ± 1.38 .805
828
Table 4. Comparison of Outcome Measures for Patients With Normal Preoperative Vitamin D, Insufficient Preoperative Vitamin
D, and Deficient Preoperative Vitamin D.
Outcome Measures Normal (n = 32)
1-Year mortality, n (%) 6 (18.75)
Driveline infection, n (%) 1 (3.1)
Gastrointestinal bleed, n (%) 11 (34.38)
Length of stay, n (%) 10 (31.25)
Readmission,a n (%) 16 (50)
Cerebrovascular accident, n (%) 3 (9.4)
All outcome measures indicate first year of postoperative left ventricular assist device support.
a ࣙ2 readmissions.
greater risk for having driveline infection when compared
with patients with a normal vitamin D level (95% CI: 0.80–
49.2, P = .07).
In their first year of LVAD support, 52 patients (28.4%)
had at least 1 episode of GI bleeding. Among those
with normal vitamin D levels, 11 (34.38%) experienced GI
bleeding, whereas 51 (21.57%) and 30 (30%) of those with
insufficient and deficient levels experienced GI bleeding,
respectively. The rates of GI bleed were not significantly
different between these groups (Table 4).
Finally, 15 patients (8.19%) experienced a cerebrovascu-
lar accident (CVA) within the first year. Among those with
normal vitamin D levels, 3 patients (9.38%) had a stroke,
whereas 2 (3.92%) and 10 (10%) of those in the insuffi-
cient and deficient groups, respectively, had a stroke. The
rates of CVA were not significantly different between these
groups.
Discussion
Vitamin D deficiency is highly prevalent among patients
with heart failure. The 3-year National Health and
Nutrition Examination Survey completed in 2004 reported
an 89% prevalence rate of hypovitaminosis D (defined as
vitamin D level <30 ng/mL), after controlling for age, race,
and sex.13,14 A study of 154 patients with LVAD reported
34.4% had vitamin D levels <10 ng/mL, relatively lower
than the percentage of patients in our study with levels
<10 ng/mL, which was 21.6%.15
Despite its prevalence in the heart failure population, few
studies are analyzing vitamin D status and its relationship
to important clinical outcomes in patients with LVAD. A
recent study analyzing patients undergoing elective cardiac
surgery for coronary artery bypass and valve replacement
did not find a relationship between preoperative vitamin D
levels and several postoperative outcomes including length
of stay.16
Our study is the largest to evaluate the clinical impact
of vitamin D levels in patients with LVAD. Among all
Nutrition in Clinical Practice 33(6)
Preoperative Vitamin D
Insufficient (n = 51) Deficient (n = 100) P Value
1 (1.96) 10 (10.00) .0349
13 (26.0) 17 (17.2) .027
11 (21.57) 30 (30.0) .3950
14 (27.45) 39 (39.0) .3381
36 (70.59) 71 (71.0) .0737
2 (4.0) 10 (10.0) .430
patients, it was found that 51 of 183 patients (27.88%)
were vitamin D insufficient and that 100 (54.64%) were
deficient. Recognizing the wide range of pleiotropic effects
associated with vitamin D, we sought to investigate the role
of hypovitaminosis D with several of the most common
adverse events in patients with LVAD, including infection,
GI bleeding, and stroke.
Driveline Infection
Device-associated infection remains 1 of the most common
complications in long-term LVAD support with a preva-
lence rate ranging up to 50% in the first 3 years after
implantation.17 Infection is associated with higher mortality
and morbidity. Recent studies have elucidated the role of
vitamin D receptors in activating nearly all types of immune
cells, including CD4+ and CD8+ T cells, B cells, neutrophils,
macrophage, and dendritic cells.18 In addition, as part of
adaptive immunity, vitamin D plays a major role in the pro-
liferation and differentiation of B and T cells, and modulates
immunoglobulin production.7 Further studies have revealed
the importance of vitamin D in the endothelial response
to lipopolysaccharide (LPS). Specifically, endothelial cells
treated with 1,25(OH) vitamin D generated a significantly
lower response when stimulated with LPS compared with
untreated cells.8 These findings, which indicate the impor-
tant role of vitamin D in immune system function, suggest
vitamin D depletion may be associated with greater risk
for infections. Indeed, the recently published CALCITOP
study, an analysis of 3340 cardiac surgery patients, identified
2.57 times greater risk for postoperative infection in patients
with vitamin D deficiency compared with those with higher
levels.19
This trend was clearly seen in our LVAD population in
those patients with deficient or insufficient vitamin D levels.
The risk for LVAD driveline infections, when compared
with patients with normal vitamin D level, was 6.185
times greater with vitamin D insufficiency and 11.467 times
greater with vitamin D deficiency.
Obeid et al
Readmission Rate
Strategies to reduce hospital readmission rates are essential
to improve quality of life for patients with LVAD. Multiple
studies have identified several causes leading to recurrent
readmissions. One large study showed that recurrent heart
failure (33.3%) and GI bleeding (22.2%) were the most
common causes of LVAD readmissions.20 A separate study
of 126 patients indicated that GI bleeding and LVAD-
related infections were the leading causes of readmissions.21
Our analysis identified vitamin D level as an independent
predictor for increased postoperative readmission rate in
patients with LVAD. Patients with insufficient or deficient
vitamin D levels had 2.5 times increased risk for being
readmitted in the year after LVAD implantation. A sig-
nificant percentage of hospital readmissions occur because
of driveline infections. The increased rate of infection in
patients with low vitamin D level may, in part, drive the
increased readmission rate in that group.
Stroke
The presence of 25-OH vitamin D receptors within central
nervous system tissues indicates the capability of vitamin
D to cross the blood–brain barrier and to interact with
neural tissue. Several studies have suggested that vitamin
D may have a neuroprotective effect, reducing the rate of
stroke and cognitive decline.22,23 In addition, data from
prospective observational studies indicate that vitamin D
deficiency is an independent risk factor for stroke. In 1 meta-
analysis, after adjusting for cardiovascular risk factors, the
risk for cerebrovascular disease was significantly reduced in
individuals with higher 25-OH vitamin D levels compared
with those who were deficient.22,23 Vitamin D levels have
also been shown to be associated with stroke severity and
worsened early functional capacity after stroke.24-27
Known predictors of stroke in patients with LVAD
include diabetes, aortic cross-clamping with cardioplegic
arrest, duration of LVAD support, and international nor-
malized ratio.27 Recently, a study by Zittermann et al,15
analyzing 154 patients with LVAD, indicated that those with
25-OH vitamin D level <10 ng/mL had an increased risk
for stroke and mortality within 1 year of implantation. In
our study, despite the increased risk for readmission and
driveline study, there was no significant association between
vitamin D levels and risk for stroke. This discrepancy in
relationship of hypovitaminosis D to risk for postoperative
stroke may be because of the relatively larger number of
patients with very low preoperative vitamin D levels in
Zittermann et al’s15 study compared with ours.
Conclusions
Our study suggests that deficient and insufficient levels of
25-OH vitamin D before LVAD implantation are associated
829
with significantly higher risk for driveline infection and
readmission within 1 year. These effects may be a result of
the ubiquitous presence of vitamin D receptors on immune
cells and the role these receptors play in leukocyte differ-
entiation and proliferation. Consequently, vitamin D levels
may alter the immune response to infectious organisms and
contribute to the risk for driveline infection.
This study identified preoperative vitamin D depletion
as an important risk factor in those patients undergoing
LVAD implantation. Further study is required to determine
whether a preoperative repletion regimen will result in
protective effects or risk reduction. Limitations to this study
include that it was a single-center, retrospective study and
that longitudinal vitamin D levels were not available. In
addition, recent work has shown that vitamin D levels may
be diluted in patients who receive bolus IV fluids, an effect
that may be significant in patients with advanced heart
failure. Future investigation should control for this potential
confounding variable.28
Statement of Authorship
F. A. Obeid, G. Bhat, E. Drever, and A. Tatooles contributed
to conception/design of the research; F. A. Obeid, G.
Yost, and G. Bhat contributed to acquisition, analysis, or
interpretation of the data; F. A. Obeid and G. Yost drafted
the manuscript; F. A. Obeid, G. Yost, G. Bhat, E. Drever,
and A. Tatooles critically revised the manuscript; and F.
A. Obeid, G. Yost, G. Bhat, E. Drever, and A. Tatooles
agree to be fully accountable for ensuring the integrity and
accuracy of the work. All authors read and approved the
final manuscript.
References
1. Shor R, Tirosh A, Shemesh L, et al. 25 Hydroxyvitamin D levels in
patients undergoing coronary artery catheterization. Euro J Int Med.
2012;23(5):470-473.
2. Wang TJ, Pencina MJ, Booth SL, et al. Vitamin D deficiency and risk
of cardiovascular disease. Circulation. 2008;117(4):503-511.
3. Kunadian V, Ford GA, Bawamia B, Qiu W, Manson JE. Vitamin D
deficiency and coronary artery disease: a review of the evidence. Am
Heart J. 2014;167(3):283-291.
4. Liu L, Chen M, Hankins SR, et al; Drexel Cardiovascular Health
Collaborative Education, Research and Evaluation Group. Serum 25-
hydroxyvitamin D concentration and mortality from heart failure
and cardiovascular disease and premature mortality from all-cause in
United States adults. Am J Cardiol. 2012;110(6):834-839.
5. Liu LC, Voors AA, van Veldhuisen DJ, et al. Vitamin D status and
outcomes in heart failure patients. Eur J Heart Fail. 2011;13:619-625.
6. Makariou SE, Michel P, Tzoufi MS, Challa A, Milionis HJ. Vitamin
D and stroke: promise for prevention and better outcome. Curr Vasc
Pharmacol. 2014;12(1):117-124.
7. Bartley J. Vitamin D: emerging roles in infection and immunity. Expert
Rev Anti Infect Ther. 2010;8:1359-1369.
8. Equils O, Naiki Y, Shapiro AM, et al. 1,25-Dihydroxyvitamin D
inhibits lipopolysaccharide-induced immune activation in human en-
dothelial cells. Clin Exp Immunol. 2006;143:58-64.
830
9. Khoo AL, Chai LY, Koenen HJ, et al. 1,25-dihydroxyvitamin D3
modulates cytokine production induced by Candida albicans: impact
of seasonal variation of immune responses. J Infect Dis. 2011;203:122-
130.
10. Sayer G, Bhat G. The renin-angiotensin-aldosterone system and heart
failure. Cardiol Clin. 2014;32(1):21-32.
11. Hasin T, Marmor Y, Kremers W, et al. Readmissions after implan-
tation of axial flow left ventricular assist device. J Am Coll Cardiol.
2013;61(2):153-163.
12. Cotts WG, McGee EC Jr, Myers SL, et al. Predictors of hospital
length of stay after implantation of a left ventricular assist device:
an analysis of the INTERMACS registry. J. Heart Lung Transplant.
2014;33(7):682-688.
13. Holick MF, Binkley NC, Bischoff-Ferrari HA, et al; Endocrine Society.
Evaluation, treatment, and prevention of vitamin D deficiency: an
Endocrine Society clinical practice guideline. J Clin Endocrinol Metab.
2011;96(7):1911-1930.
14. Kim DH, Sabour S, Sagar UN, Adams S, Whellan DJ. Prevalence
of hypovitaminosis D in cardiovascular diseases (from the National
Health and Nutrition Examination Survey 2001 to 2004). Am J Cardiol.
2008;102(11):1540-1544.
15. Zittermann A, Morshuis M, Kuhn J, et al. Vitamin D metabolites
and fibroblast growth factor-23 in patients with left ventricular assist
device implants: association with stroke and mortality risk. Eur J Nutr.
2016;55(1):305-313.
16. Sriram K, Perumal K, Alemzadeh, et al. The relationship between im-
mediate preoperative serum 25-hydroxy-vitamin D3 levels and cardiac
function, dysglycemia, length of stay, and 30-d readmission in cardiac
surgery patients. Nutrition. 2015;31:820-826.
17. Leuck AM. Left ventricular assist device driveline infections: recent
advances and future goals. J Thorac Dis. 2015;7(12):2151-2157.
18. Baeke F, Takiishi T, Korf H, Gysemans C, Mathieu C. Vitamin D:
modulator of the immune system. Curr Opin Pharmacol. 2010;10:482-
496.
Nutrition in Clinical Practice 33(6)
19. Zittermann A, Kuhn J, Ernst JB, et al. Circulating 25-hydroxyvitamin
D and 1,25-dihydroxyvitamin D concentrations and postoperative
infections in cardiac surgical patients: the CALCITOP-study. PLoS
ONE. 2016;11(6):e0158532.
20. Tsiouris A, Paone G, Nemeh HW, Brewer RJ, Morgan JA. Fac-
tors determining post-operative readmissions after left ventricular
assist device implantation. J Heart Lung Transplant. 2014;33(10):1041-
1047.
21. Akhter SA, Badami A, Murray M, et al. Hospital readmissions
after continuous-flow left ventricular assist device implantation: inci-
dence, causes, and cost analysis. Ann Thorac Surg. 2015;100(3):884-
889.
22. Pilz S, Tomaschitz A, Drechsler C, Zittermann A, Dekker JM, März
W. Vitamin D supplementation: a promising approach for the pre-
vention and treatment of strokes. Curr Drug Targets. 2011;12:88-
96.
23. Buell JS, Dawson-Hughes B. Vitamin D and neurocognitive dys-
function: preventing “D” ecline? Mol Aspects Med. 2008;29:415-
422.
24. Sun Q, Pan A, Hu FB, Manson JE, Rexrode KM. 25-hydroxyvitamin
D levels and the risk of stroke: a prospective study and meta-analysis.
Stroke. 2012;43:1470-1477.
25. Chowdhury R, Stevens S, Ward H, Chowdhury S, Sajjad A, Franco
OH. Circulating vitamin D, calcium and risk of cerebrovascular
disease: a systematic review and meta-analysis. Eur J Epidemiol.
2012;27:581-591.
26. Daubail B, Jacquin A, Guilland JC, et al. Serum 25-hydroxyvitamin
D predicts severity and prognosis in stroke patients. Eur J Neurol.
2013;20:57-61.
27. Morgan JA, Brewer RJ, Nemeh HW, et al. Stroke while on long-
term left ventricular assist device support: incidence, outcome, and
predictors. ASAIO J. 2014;60(3):284-289.
28. Sriram K, Meguid M. The ongoing story of vitamin D. Nutrition.
2015;31:1293-1294.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Diagnostic Accuracy of Bioelectrical Impedance Analysis December 2018 831–842

C 2018 American Society for

Parameters for the Evaluation of Malnutrition in Patients Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10098
Receiving Hemodialysis wileyonlinelibrary.com

Angela Teodósio da Silva, MSc1 ; Daniela Barbieri Hauschild, MSc1 ;

1
Yara Maria Franco Moreno, PhD ; João Luiz Dornelles Bastos, PhD2 ;
and Elisabeth Wazlawik, PhD1

Abstract
Background: In the absence of a gold standard technique for assessing nutrition status in patients receiving hemodialysis
(HD), we aimed to determine the diagnostic accuracy of single-frequency (50 kHz) bioelectrical impedance analysis parameters,
resistance/height (R/H), reactance/height (Xc/H), and impedance/height (Z/H), and their cutoff points for malnutrition. Methods:
The reference standards, Subjective Global Assessment (SGA), Malnutrition Inflammation Score, and Nutritional Risk Screening
2002, were performed at baseline and then once a year for 2 years. At least 2 assessments for each reference standard were performed
during the monitoring period, and those patients who were assessed as malnourished on at least 2 consecutive occasions were
classified as malnourished. Results: A total 101 patients receiving HD were evaluated. R/H and Z/H demonstrated low to moderate
accuracy to diagnose malnutrition in men and low accuracy in women, whereas the accuracy of Xc/H was uncertain. The cutoff
points of bioelectrical impedance vector analysis (BIVA) parameters, determined based on the SGA to maximize sensitivity and
specificity simultaneously, were: R/H ࣙ330.05 and ࣙ420.92 ohms/m for men and women, respectively; Z/H ࣙ332.71 and ࣙ423.19
ohms/m for men and women, respectively. In men, sensitivity based on the cutoff points of R/H and Z/H together ranged from
73% to 89% and specificity ranged from 49% to 50%. In women, sensitivity ranged from 58% to 80% and specificity from 48% to
55%. Conclusion: BIVA parameters demonstrated low to moderate accuracy in men and low accuracy in women for the diagnosis
of malnutrition. (Nutr Clin Pract. 2018;33:831–842)

Keywords
bioelectrical impedance; chronic renal insufficiency; nutritional assessment; renal dialysis

Background Bioelectrical impedance analysis (BIA) is based on the

Protein-energy wasting is characterized as the depletion of
body protein and fat and is 1 of the most prevalent nutrition
disorders in patients with chronic kidney disease receiving
hemodialysis (HD).1 Moreover, this condition is associated
with an increase in morbidity and mortality rates as well as
reduction in physical function and quality of life.2,3
The nutrition status assessment and monitoring of pa-
tients receiving HD are fundamental to prevention, diag-
nosis, and treatment of malnutrition.1 However, a precise
assessment is complex due to frequent changes in the
body hydration, occurrence of comorbidities, and chronic
inflammation.4,5 In the absence of a gold standard for
the assessment of nutrition status, efforts have been made
to identify indicators capable of an accurate diagnosis of
malnutrition, such as handgrip strength (HGS), 50 kHz
phase angle, and 50 kHz bioelectrical impedance vector
analysis (BIVA).5-9 However, few studies have evaluated the
validity of each indicator and this remains unclear.10
principle that body tissues offer different opposition to the
passage of electrical current. Impedance (Z) refers to this
From the 1 Federal University of Santa Catarina, Health Sciences
Center, Post-Graduate Program in Nutrition, Trindade Florianópolis,
Santa Catarina, Brazil; 2 Federal University of Santa Catarina, Health
Sciences Center, Public Health Department, Trindade Florianópolis,
Santa Catarina, Brazil.
Financial disclosures: This study was carried out with financial
support from the Brazilian governmental agency CAPES
(Coordination for the Improvement of Higher Education Personnel)
by way of scholarships awarded to the students involved in the project.
Conflicts of interest: None declared.
This article originally appeared online on May 22, 2018.
Corresponding Author:
Elisabeth Wazlawik, PhD, Federal University of Santa Catarina,
Health Sciences Center, Post-Graduate Program in Nutrition,
Campus Universitário, Trindade Florianópolis, Santa Catarina,
Brazil 88040–970; 55-48-3721-5138.
Email: e.wazlawik@ufsc.br
832
opposition and is composed of 2 vectors: resistance (R)
and reactance (Xc).11 The R component of BIA represents
the opposition to flow of an alternating current through
ionic solutions and is inversely related to intracellular
and extracellular water volume. Muscle tissue is highly
conductive of electrical current due to a large amount
of water and electrolytes, and presents low resistance to
the passage of electric current.12 In situations of illness, a
higher R value may be related to malnutrition,13 and so
a higher R value may be related to the lower amount of
muscle mass and, consequently, of water and electrolytes.
Higher Xc values suggest a greater quantity of body cell
mass and lean body mass.6,11 The parameters R, Xc, and
Z were standardized by the height (H) in meters of each
patient (resistance/height [R/H], reactance/height [Xc/H],
and impedance/height [Z/H]) to account for the effect of
the length of the conductor because the impedance of a
conductor is related to its length, cross-sectional area, and
frequency of the electrical current applied.14 The phase
angle is calculated based on R and Xc and is considered an
indicator of both nutrition status and functional state; thus,
the phase angle determined at 50 kHz has been explored
as a prognostic indicator in patients receiving HD.15,16
The BIVA is based on the evaluation of the values of
normalized R and Xc for H. After normalization, the BIVA
parameters (R/H, Xc/H, Z/H) are plotted as non-graph
resistance-reactance (RXc) vectors. The position is thought
to provide information about hydration status, cell mass,
and cell membrane integrity.17 Some limitations of the use
of BIA to evaluate nutrition status in clinical populations
have been reported and have led researchers to explore the
use of raw impedance data in nutrition evaluation.18-20
None of the nutrition indicators currently used are
considered complete for patients receiving HD.21 Besides
BIA, the International Society of Renal Nutrition and
Metabolism recommends the use of anthropometric in-
dicators, such as body mass index (BMI), percentage of
fat mass (%FM), mid-upper arm circumference (MUAC)
and mid-upper arm muscle circumference (MUAMC), and
biochemical indicators, such as serum albumin level.1 The
single-frequency (SF) BIA approach is more affordable
and available than bioimpedance spectroscopy. Considering
the scarcity of studies and the necessity to explore the
application of BIVA parameters (R/H, Xc/H, Z/H) for
the evaluation of malnutrition in patients receiving HD,
this study aimed to determine the diagnostic accuracy of
SF-BIA parameters in the evaluation of malnutrition in
patients receiving HD using 3 reference standards: Subjec-
tive Global Assessment (SGA), Malnutrition Inflammation
Score (MIS), and Nutritional Risk Screening (NRS) 2002.
We also aimed to establish cutoff points for BIVA parame-
ters, investigate correlations with other clinically important
nutrition indicators (BMI, %FM, MUAC, MUAMC, HGS,
Nutrition in Clinical Practice 33(6)
and serum albumin level), and determine the prevalence of
malnutrition diagnosed using BIVA.
Materials and Methods
Study Design and Subjects
A prospective, longitudinal, diagnostic validation study was
conducted with a convenience sample recruited from 2 HD
clinics. The sample consisted of 101 patients aged ࣙ19
years, undergoing HD 3 times a week, and who agreed to
participate from April to September 2011, June to Septem-
ber 2012, and July to October 2013. Exclusion criteria
were patients receiving HD for <3 months; those with
a BMI >34 kg/m²; patients with amputated or atrophied
limbs; those wearing a pacemaker; patients with a current
diagnosis of cancer, stroke sequelae, or unable to respond;
and those hospitalized for any reason and not undergoing at
least 2 evaluations with the reference standards during the
evaluation period.
Demographic and clinical data were obtained from the
medical record. In the beginning of the study, a nutrition
status assessment was performed after a session of HD
using different indicators: BIVA parameters (R/H, Xc/H,
and Z/H), BMI, %FM, MUAC, MUAMC, and HGS. The
reference standard indicators (SGA, MIS, and NRS 2002)
were determined at the beginning of the study as well as
once a year throughout the follow-up period. The minimum
follow-up period was 10 months (1 patient) and the mean
duration was 19 months (Figure 1). The clinical informa-
tion, the BIVA parameters, and the reference standards were
available to different evaluators, and the data were analyzed
by another appraiser without influence on data collection.
This study received approval by the local Research Ethics
Committee under process number No. 1821 and CAAE
14375113.8.0000.0121; all participants signed a statement
of informed consent.
BIA
SF-BIA was performed for the determination of R and Xc
using a portable tetrapolar device (Biodynamics R
model
310e, Biodynamics Corporation, Seattle, WA) with a current
intensity of 800 μA and at a single frequency of 50 kHz. The
electrodes were placed on the midline between the promi-
nent ends of the radius and ulna of the wrist and midline
between the medial and lateral malleoli of the ankle. The
patients were evaluated approximately 20 minutes following
HD on the side of the body without vascular access,22 as
recommended by the U.S. National Institutes of Health.23
The SF-BIA was chosen because it presents a single result
for the entire body of R (ohm), Z (ohm), and Xc (ohm).
BIA parameters. R (ohm) and Xc (ohm) to obtain the
Z (ohm)24 :
da Silva et al
Figure 1. Flowchart of sample selection process.
*Body mass index >34 kg/m2 , limbs amputated or atrophied, pacemaker, cancer, heart attack, inability to respond, hospitalized
or positive for acquired immunodeficiency syndrome.
**Death, underwent kidney transplant, change in treatment to peritoneal dialysis, transference from dialysis center.
√
Z= (R2 + Xc2 ).
R, Xc, and Z were standardized by H in meters: R/H, Xc/H,
Z/H (ohm/m).
Bioelectrical impedance vector analysis (BIVA 2002). The
BIVA 2002 software developed by Piccoli and Pastori.11
was used for the determination of BIVA, with the creation
of 50%, 75%, and 95% tolerance ellipses as well as confi-
dence ellipses. The normalization of R and Xc by H was
expressed as z-scores (z-R/H and z-Xc/H) with the reference
population described by Piccoli et al.25 Patients with vectors
outside the 75% ellipsis in the upper quadrant (underweight)
and lower quadrant (cachexia) on the right side of the graph
were classified as malnourished (Figure 2A).
Phase angle. R and Xc for the calculation of:
phase angle (◦ ) = arctangent [(Xc()/R()) × (180/π)]26
Nutrition Screening
We used 3 reference standards; 2 tools are used to do a
complete nutrition assessment, SGA and MIS, and 1 tool
is meant to identify nutrition risk, NRS 2002. The reference
standards are composite indicators, which include objective
and subjective criteria in the evaluation, and thus use >1
833
measurement parameter in the evaluation of the nutrition
status. They have the advantage of being easily applicable
in clinical practice and were used as a reference standard in
another of our research studies.8 Patients were classified as
malnourished when diagnosed with malnutrition at least 2
consecutive times by the same indicator during the follow-
up period (2011–2013). The SGA was based on clinical
history and clinical examination and classifies the patients
as: A) well nourished, B) moderately malnourished or sus-
pected of being malnourished, or C) severely malnourished.
Categories B and C were grouped for the analysis.27 The
MIS considers the SGA components as well as the duration
of HD, comorbidities, BMI, serum albumin level, and total
iron binding capacity.28 It was performed according to
Kalantar-Zadeh et al, and the patients were classified as well
nourished (<6) and malnourished (ࣙ6).29 NRS 2002 was
evaluated according to Kondrup et al and the patients were
classified as without (<3) or at nutrition risk (ࣙ3).30
Anthropometric Variables
Weight (kg), H (m), MUAC (cm), and skin-fold thickness
(mm) were determined after the session of HD. MUAC
and skin-fold thickness were measured on the arm without
834 Nutrition in Clinical Practice 33(6)

Figure 2. (A) Graph with regions of elliptic probability on RXc plane normalized by height (R/H and Xc/H, in /m) 11 ;
(B) Position of vector mean for men (•) and women () after transformations of impedance measures into z-scores; (C)
Distribution of men on resistance-reactance graph; (D) Distribution of women on resistance-reactance graph; (E) Confidence
ellipses for men (black line) and women (dashed line).

vascular access. All data were collected by trained nutrition-
ists following reference procedures.31
Weight (kg) was determined using an electronic scale
(Marte R
, Marte Balanças e Aparelhos de Precisão Ltda,
Santa Rita do Sapucaı́, MG, Brazil) and height (cm) was de-
termined using a portable stadiometer (Sanny R
, American
Medical do Brasil, São Bernardo do Campo, SP, Brazil).32
Skin-fold thicknesses (mm) were determined using
calipers (Lange R
adipometer, Beta Technology Incorpo-
rated, Cambridge, MD). Biceps, triceps, subscapular, and
suprailiac skin-fold measurements were performed follow-
ing the recommendations proposed by Lohman et al.31
%FM was determined using the formula described by Siri.33
The equation proposed by Durnin and Womersley was used
to calculate body density.34
MUAC (cm) was measured using a flat steel measuring
tape (Cescorf 
R
, Cescorf Equipamentos para Esporte Ltda.,
Porto Alegre, RS, Brazil) following the method described
by Frisancho.35 MUAMC was determined by the following
equation:
da Silva et al
[MUAMC (cm) = MUAC (cm) − π × [triceps skin
fold (mm)/10]35
HGS
HGS (kg) was measured using a hydraulic dynamometer
(Saehan R
model SH 5001, Saehan Corporation, Yangdeok-
Dong, Masan, Korea). The patients were instructed about
the use of the dynamometer; explanation and demonstra-
tion were given on the need to tighten the handle to their
full strength. At the moment of assessment, each participant
was seated with hips and knees at 90° flexion, adducted
shoulder close to the trunk, flexed elbow at 90°, with
the forearm in neutral position (between supination and
pronation), and wrist between 0° and 30° of extension and
0° and 15° of ulnar deviation.36
HGS was evaluated on the arm side without vascular
access. Three measurements were performed, with the pe-
riod of maximum continuous contraction of 3 seconds,
with a 30-second interval between trials.37 For the analysis,
the highest value of the 3 measurements was taken into
consideration.38,39
Biochemical Variable
Serum albumin level (g/dL) was determined using the col-
orimetric method with bromocresol green.40
Statistical Analysis
Description of the sample was performed by absolute
and relative frequencies, mean and standard deviation for
parametric variables, or median and interquartile range for
non-parametric variables. To check the difference between
these variables according to gender, the χ 2 test was used
for categoric variables and the Student’s t-test or Mann-
Whitney test was used for numeric variables.
Pearson (parametric variables) or Spearman (non-
parametric variables) correlation was performed to assess
the relationship between the BIVA parameters and other
nutrition status indicators (SGA, MIS, NRS 2002, phase
angle, BMI, %FM, MUAC, MUAMC, HGS, and serum
albumin level). For dichotomous variables (SGA, MIS,
NRS 2002): 0, malnourished; or 1, not malnourished. The
coefficients were interpreted as follows: 0–0.29 = weak
correlation; 0.30–0.69 = moderate correlation; and 0.70–1.0
= strong correlation.41
To verify the diagnostic accuracy of BIVA parameters in
the identification of patients with malnutrition according to
SGA, MIS, and NRS 2002, receiver operating characteristic
(ROC) curves were constructed and areas under the curve
(AUC) were obtained. When AUC was >0.90, accuracy was
considered high, while 0.70–0.90 was considered moderate;
0.50–0.69 was considered low and <0.50 meant uncertain
835
accuracy.42 Using the SGA as the reference standard, the
best cutoff points for R/H and Z/H were obtained for gender
because it maximizes sensitivity and specificity simultane-
ously when compared with NRS 2002 and MIS.
Using the cutoff of R/H and Z/H together, sensitivity,
specificity, positive predictive value (PPV) and negative
predictive value (NPV) were determined for the usefulness
of the BIVA parameters. Individuals with R/H and Z/H
below the cutoff points were considered nourished and
those with R/H and Z/H above the cutoff points were con-
sidered malnourished. Statistical analyses were performed
using software (Stata version 11, Stata Corporation, College
Station, TX). The vector means of the resistance-reactance
graph were analyzed using Hotelling’s T2 test and univariate
analysis (F test). P < 0.05 was considered statistically
significant.
Results
A total of 101 individuals (60 men) participated in the
present study. Mean age was 53.5 ± 15.7 years (range,
22–81 years) and median duration receiving HD was 40
months (range, 3 months–40 years). Almost one-third of
the patients (n = 28, 28%) were aged ࣙ65 years. Figure 1
displays the flow chart of the sample selection process.
The causes of chronic kidney disease were systemic arte-
rial hypertension (42%), diabetes mellitus (24%), glomeru-
lonephritis (13%), polycystic kidney disease (5%), and other
or undetermined causes (17%). Table 1 shows the main
clinical characteristics of the patients stratified by gender.
The mean HGS, weight, height, and MUAMC were higher
in men, while women showed higher %FM, R/H, Xc/H, and
Z/H (P < 0.001).
The results of the BIVA are displayed in Figure 2. The
vector mean for the women was near the 75% ellipsis of
tolerance in the lower right quadrant (malnourished), and
the vector mean for the men was near the 95% ellipsis of
the lower right quadrant (malnourished; Figure 2B). Based
on the resistance-reactance graph, the malnutrition rate was
71.6% (n = 43) among the men (Figure 2C) and 51.2% (n =
21) among the women (Figure 2D). Figure 2E displays the
confidence intervals of the R/H and Xc/H for the women
(ellipsis with dashed line) and men (ellipsis with black
line). The Hotelling test (T² = 54.7) and F test (F = 27.1)
revealed significant differences between sexes (P < 0.001),
demonstrating the importance of establishing specific cutoff
points for the BIA variables (R/H, Xc/H, and Z/H) for each
sex. The prevalence of malnutrition (patients classified as
malnourished on at least 2 consecutive evaluations) based
on SGA was 23.8% (n = 24) among the total sample, 23.3%
(n = 14) among the men, and 24.4% (n = 10) among the
women. Based on the MIS, malnutrition was 26.7% (n = 27)
among the total sample, 25.0% (n = 15) among the men,
and 29.2% among the women (n = 12), and based on the
836 Nutrition in Clinical Practice 33(6)

Table 1. Clinical Characteristics of Patients Receiving Dialysis Stratified by Gender: Florianópolis, Brazil, 2011–2012.

Characteristics Total (n = 101) Men (n = 60) Women (n = 41) P-Value

Age (years) 53.5 ± 15.7 53.2 ± 16.1 53.7 ± 15.5 0.884b

Nutrition variables
HGS (kg) 26.5 ± 11.3 31.0 ± 11.4 19.9 ± 7.20 <0.001c
Weight (kg) 65.4 ± 12.9 69.0 ± 11.5 60.2 ± 13.3 <0.001b
Height (m) 1.63 ± 0.10 1.68 ± 0.09 1.56 ± 0.09 <0.001b
BMI (kg/m²) 24.5 ± 3.62 24.4 ± 3.41 24.6 ± 3.95 0.715b
FM (%) 32.6 ± 6.54 30.7 ± 5.81 35.4 ± 6.63 <0.001b
MUAC (cm) 27.8 ± 4.04 27.5 ± 3.25 28.2 ± 5.01 0.756c
MUAMC (cm) 23.3 ± 3.00 23.9 ± 2.63 22.5 ± 3.32 0.012b
Phase angle (°) 6.35 ± 1.36 6.43 ± 1.26 6.24 ± 1.50 0.501b
Serum albumin level 4.00 ± 0.28 4.01 ± 0.34 3.99 ± 0.18 0.507c
BIA parameters
R/H (ohm/m) 372.4 ± 73.0 336.6 ± 53.0 424.8 ± 66.6 <0.001b
Xc/H (ohm/m) 41.3 ± 11.6 37.8 ± 8.73 46.6 ± 13.2 <0.001c
Z/H (ohm/m) 374.8 ± 73.4 338.8 ± 53.2 427.5 ± 67.0 <0.001b
Comorbidities (n, [%])*
Diabetes mellitus 30 (29.7) 15 (25.0) 15 (36.6) 0.211d
Systemic arterial hypertension 84 (83.2) 47 (78.3) 37 (90.2) 0.116d
Heart disease 27 (26.7) 15 (25.0) 12 (29.3) 0.634d
Kidney function
Time on dialysis (months)a 40.4 (16;84) 28.4 (12;76) 50.6 (29;97) 0.027c

BIA, bioelectrical impedance analysis; BMI, body mass index; FM, fat mass; HGS, handgrip strength; MUAC, mid-upper arm circumference;
MUAMC, mid-upper arm muscle circumference; R/H, resistance for height; Xc/H, reactance for height; Z/H, impedance for height.
*Absolute and relative frequency for categoric variables.
a Median and interquartile range.
b Student’s t-test.
c Mann-Whitney test.
d χ 2 test.

NRS 2002, it was 16.8% (n = 17) among the total sample,
15.0% (n = 9) among the men, and 19.5% (n = 8) among
the women.
Positive correlations were found between R/H and Z/H
with SGA (r = 0.20, P ࣘ 0.05) and NRS 2002 (dichotomous
variables: 0, malnourished; or 1, not malnourished): (r =
0.24, P <0.05). R/H and Z/H were negatively correlated with
MUAC (r = −0.27, P <0.05), BMI (r = −0.31, P <0.05),
MUAMC (r = −0.48, P <0.05), and HGS (r = −0.61, P
<0.05). Xc/H exhibited a positive correlation with % FM
(r = 0.27, P <0.05) and phase angle (r = 0.72, P <0.05;
Table 2).
The AUC for R/H demonstrated low accuracy for the
total sample, and the highest was the AUC of the NRS 2002
(0.68). For men, R/H demonstrated moderate accuracy with
the NRS 2002 (0.75) and low accuracy with the SGA and
MIS. For women, the accuracy of R/H was low using all
3 reference standards, with the highest AUC found for the
SGA (0.66; Table 3).
The AUCs for Xc/H demonstrated accuracy <50%,
ranging from 42% to 46% for the total sample, 43% to
49% for men, and 33% to 44% for women, demonstrating
uncertain accuracy (data not shown).
The AUCs for Z/H demonstrated low accuracy for the
total sample, and the highest was that of the NRS 2002
(0.68). For men, Z/H demonstrated moderate accuracy with
the NRS 2002 (0.75) and low accuracy with the SGA and
MIS. For women, the accuracy of Z/H was low for all
reference standards, with the highest AUC found for the
SGA (0.66; Table 3).
Table 3 displays the cutoff points of R/H and Z/H. The
SGA was the reference standard for the classification of
malnutrition because it maximizes sensitivity and specificity
simultaneously. The cutoff points established using SGA
were as follows: R/H ࣙ 330.05 ohms/m for the total sample
and for men, and ࣙ 420.92 ohms/m for women; Z/H
ࣙ340.47 ohms/m for the total sample, ࣙ 332.71 ohms/m for
men, and ࣙ423.19 ohms/m for women.
Figure 3 shows the ROC curves with the values of
diagnostic accuracy of R/H and Z/H in the identification
of men and women with malnutrition according to SGA,
where R/H and Z/H showed low accuracy among women
and men.
The estimated prevalence of malnutrition based on the
cutoff points of R/H and Z/H together, established using the
SGA, was 65.4% (n = 66) among the total sample, 56.7%
da Silva et al 837

Table 2. Correlation Between Bioelectrical Impedance Analysis Parameters and Nutrition Indicators Among Patients Receiving
Hemodialysis: Florianópolis, Brazil, 2011–2013.

Parameter R/H P Xc/H P-Value Z/H P-Value

SGA 0.1975 0.048a –0.085 0.398a 0.1955 0.050a

MIS 0.1661 0.097a –0.068 0.500a 0.1657 0.098a
NRS 2002 0.2364 0.017a –0.1048 0.297a 0.2351 0.0180a
Phase angle (°) –0.0701 0.486b 0.7161 <0.001b –0.0560 0.578b
BMI (kg/m²) –0.3156 0.001b –0.0798 0.428b –0.3135 0.001b
FM (%) 0.0990 0.325b 0.2664 0.007b 0.1024 0.308b
MUAC (cm) –0.2746 0.006b 0.1092 0.277b –0.2698 0.006b
MUAMC (cm) –0.4886 <0.001b –0.0402 0.690b –0.4841 <0.001b
HGS (kg) –0.6169 <0.001b –0.1492 0.137b –0.6129 <0.001b
Serum albumin level (g/dL) 0.0011 0.991b 0.0738 0.466b 0.0026 0.979b

BMI, body mass index; FM, fat mass; H, height; HGS, handgrip strength; MIS, Malnutrition Inflammation Score; MUAC, mid-upper arm
circumference; MUAMC, mid-upper arm muscle circumference; NRS 2002, Nutritional Risk Screening 2002; R, resistance; SGA, Subjective
Global Assessment; Xc, reactance; Z, impedance.
a Spearman correlation; significant values in bold (P ࣘ 0.05).
b Pearson correlation; significant values in bold (P ࣘ 0.05).

Table 3. Resistance/Height (R/H) and Impedance/Height (Z/H) Cutoff Points for Identification of Nourished and Malnourished
Patients Based on SGA, MIS, and NRS 2002, Stratified by Gender.

Reference Standard Sensitivitya Specificitya AUCa Cutoff Point (ohm/m)

Total sample (n = 101) R/H

SGA 87.50 32.37 63.39 ࣙ330.05
MIS 88.89 31.08 60.84 ࣙ326.18
NRS 2002 94.12 32.14 68.24 ࣙ330.05
Men (n = 60) R/H
SGA 78.57 50.00 66.23 ࣙ330.05
MIS 80.00 46.67 64.37 ࣙ326.18
NRS 2002 100.0 33.33 75.49 ࣙ317.65
Women (n = 41) R/H
SGA 80.00 54.84 66.13 ࣙ420.92
MIS 58.33 72.41 56.90 ࣙ437.18
NRS 2002 62.50 72.73 64.39 ࣙ443.40
Total sample (n = 101) Z/H
SGA 83.33 40.26 63.26 ࣙ340.47
MIS 88.89 31.08 60.81 ࣙ329.24
NRS 2002 94.12 32.14 68.14 ࣙ332.71
Men (n = 60) Z/H
SGA 78.57 50.00 65.84 ࣙ332.71
MIS 80.00 46.67 64.00 ࣙ329.24
NRS 2002 100.0 33.33 74.95 ࣙ319.29
Women (n = 41) Z/H
SGA 80.00 54.84 66.13 ࣙ423.19
MIS 58.33 72.41 56.90 ࣙ438.92
NRS 2002 62.50 72.73 64.39 ࣙ445.00

AUC, area under curve; MIS, Malnutrition Inflammation Score; NRS 2002, Nutritional Risk Screening 2002; SGA, Subjective Global
Assessment.
a Data expressed as percentage.

(n = 34) among the men, and 53.7% (n = 22) among the gether established for the identification of malnutrition. In
women. the total sample, the sensitivity ranged from 78% to 88%,
Table 4 shows the diagnostic accuracy of the BIVA specificity 39% to 40%, PPV 23% to 32%, and NPV 38%
parameters based on the R/H and Z/H cutoff points to- to 94%. In men, sensitivity ranged from 73% to 89% and
838
Figure 3. Receiver operating characteristic (ROC) curves of R/H with ASG in men (A) and women (B), and of Z/H in men (C)
and women (D).
specificity from 49% to 50%. In women sensitivity ranged
between 58% and 80%, and specificity between 48% and
55%.
Discussion
In this study we compared BIA with a risk tool and assess-
ment tools. Screening tools, such as NRS 2002, indicated
risk factors for a deprived nutrition condition. Also, these
tools are the first step in determining nutrition problems,43,44
and assessment tools (SGA and MIS) allow the clinician
to formulate a nutrition diagnosis.44 Our results show that
BIVA parameters RH and Z/H offer low to moderate
accuracy in men and low accuracy in women for the diag-
nosis of malnutrition using the risk and assessment tools
as reference standards. BIVA parameters were correlated
with SGA, NRS 2002, anthropometric variables, HGS, and
phase angle. R/H and Z/H together carry high sensitivity
but low specificity compared with the reference standards.
Z is a combination of R and Xc. R is the opposition
to flow of an alternating current through intracellular and
extracellular ionic solutions. Xc has been suggested to
represent the capacitive component of tissue interfaces, cell
Nutrition in Clinical Practice 33(6)
membranes, and organelles, and it is thought by some to be
related to the amount of soft tissue structures.11
Based on the difference in the mean SF-BIVA parameters
(R/H, Xc/H, and Z/H) between men and women, and the
expression of sexual dimorphism in body composition, the
highest values of R/H, Xc/H, and Z/H in the women in our
sample were indicative of a higher amount of %FM than
in men (P < 0.001). Highest values of the BIVA parameters
were similar to those of the healthy American population,7
Italian elderly individuals,45 Indians with chronic kidney
disease,46 and elderly patients hospitalized in the United
Kingdom.13
Phase angle appeared to be higher in males than in
females, although this did not reach statistical significance.
The higher phase angle in men has been reported in the
literature,7,13,45-47 suggesting a higher amount of body cell
mass,17 and particularly greater skeletal muscle mass.7 In
our sample, males had higher mean values of MAUMC (P
= 0.021) and HGS (P <0.001) than women, which may be
related to the greater amount of muscle mass, resulting in
greater muscle strength.8
A weak positive correlation was found between the
BIVA parameters R/H and Z/H and the reference standards
da Silva et al 839

Table 4. Diagnostic Accuracy of Bioelectrical Impedance Analysis Parameters (R/H and Z/H) for Identification of Nourished
and Malnourished Patients Based on SGA, MIS, and NRS 2002, Stratified by Gender.

Reference Standard Sensitivitya Specificitya PPVa NPVa

Total sample (n = 101)

(R/H ࣙ330.05; Z/H ࣙ 340.47)
SGA
Value 83.33 40.26 30.30 88.57
95% CI 62.62–95.26 29.23–52.06 19.59–42.85 73.26–96.80
MIS
Value 77.78 39.19 31.82 82.86
95% CI 57.74–91.38 28.04–51.23 20.89–44.44 66.35–93.44
NRS 2002
Value 88.24 39.29 22.73 94.29
95% CI 63.56–98.54 28.80–50.55 13.31– 34.70 80.84–99.30
Men (n = 60)
(R/H ࣙ330.05; Z/H ࣙ332.71)
SGA
Value 78.57 50.00 32.35 88.46
95% CI 49.20–95.34 34.90–65.10 17.39–50.53 69.85–97.55
MIS
Value 73.33 48.89 32.35 84.62
95% CI 44.90–92.21 33.70–64.23 17.39–50.53 65.13–95.64
NRS 2002
Value 88.89 49.02 23.53 96.15
95% CI 51.75–99.72 34.75–63.40 10.75– 41.17 80.36–99.90
Women (n = 41)
(R/H ࣙ420.92; Z/H ࣙ423.19)
SGA
Value 80.00 54.84 36.36 89.47
95% CI 44.39–97.48 36.03–72.68 17.20–59.34 66.86–98.70
MIS
Value 58.33 48.28 31.82 73.68
95% CI 27.67–84.83 29.45–67.47 13.86–54.87 48.80–90.85
NRS 2002
Value 62.50 48.48 22.73 84.21
95% CI 24.49–91.48 30.80–66.46 7.82–45.37 60.42–96.62

CI, confidence interval; MIS, Malnutrition Inflammation Score; NPV, negative predictive value; NRS, Nutritional Risk
Screening; PPV, positive predictive value; R/H, resistance/height; SGA, Subjective Global Assessment; Z/H, impedance/height.
a Data expressed as percentage.

SGA and NRS 2002. Moreover, a weak/moderate negative
correlation was found between anthropometric variables
and HGS. In an elderly Spanish population with dementia,
R/H and Z/H were found to have a moderate negative
correlation with BMI only in women.48 Regarding the
correlations between BMI and both R/H and Z/H, the
data observed in this study are in agreement with data
reported for the American population.7 In the present
study, Xc/H demonstrated a weak positive correlation with
%FM. In the American population, a weak and moderate
negative correlation was found between Xc/H and %FM
in men and women, respectively, and the same result was
found regarding correlations between %FM and both R/H
and Z/H.7 The strong positive correlation between Xc/H
and the phase angle may be explained by the fact that
both parameters reflect the integrity of cell membranes,
and the phase angle is calculated based on the Xc/R
ratio.6
The AUC of the cutoff points of R/H, Xc/H, and Z/H for
the evaluation of malnutrition ranged from 0.63 to 0.68 (low
accuracy), 0.42 to 0.46 (uncertain accuracy), and 0.6 to 0.68
(low accuracy), respectively, in the total sample. A similar
finding was reported in a study involving 1590 American
adults in which the accuracy of the BIVA for the evaluation
of %FM (AUC: 0.49–0.61) was tested using dual-energy X-
ray absorptiometry.7 In the present study, the accuracy of
Xc/H for the identification of malnutrition was <50% in
the total sample as well as for men and women separately.
In a recent study, age was the most important determinant
of phase angle variation, and a greater age was associated
with a lower phase angle in 1442 healthy subjects.49 Thus,
the reduction in Xc seems to be physiologic, accompanying
840
the aging process. Because approximately one-third (n =
28) of the patients in the present investigation were aged
ࣙ65 years, it is likely that this factor exerted an influence
on the results.13 It is possible that the vector length by
BIVA allows a better comprehension of body composition
than does the consideration of only PA independent of
vector length, or the resistance component independent of
reactance.50 Moreover, it may be suggested that the occur-
rence of comorbidities, such as diabetes and high blood
pressure, also contributes to the reduction in Xc/H. Previous
studies with specific BIVA,48 classic BIVA,51 and both52
with elderly individuals report a significant difference in R
values between well-nourished and malnourished patients,
with the malnourished demonstrating higher values of R,
whereas no difference was found in Xc.48,51,52
This is the first study to define cutoff points of the
R/H and Z/H parameters of BIA for the identification of
malnutrition in patients undergoing HD. The SGA was
the reference standard based on the highest sensitivity and
specificity values. The U.S. National Kidney Foundation,
Kidney Disease/Dialysis Outcomes and Quality Initiative
has recommended the use of the SGA for the evaluation of
nutrition status in patients receiving HD.53 The SGA was
used as the reference standard in 1 other study that aimed
to evaluate the diagnostic accuracy of the BIVA for the
evaluation of malnutrition in this group of patients.5
In the present study, the BIVA parameters R/H and
Z/H demonstrated greater sensitivity than specificity in the
identification of malnutrition. Sensitivity ranged from 78%
to 88% in the total sample, which is similar to that described
in a study conducted in Italy with patients receiving HD.5
In contrast, specificity ranged from 39% to 40%, which
differs significantly from the figure reported in the Italian
study on the use of BIVA for the diagnosis of malnutri-
tion in patients undergoing HD (79%).5 A sensitive test
needs to be chosen when the consequences of overlooking
a diagnosis are considerable.54 Malnutrition is common
in patients with chronic kidney disease, and it has been
identified as an important risk factor for complications and
mortality.53 Thus, it is important to identify patients who
are truly malnourished. Accurate diagnosis can lead to a
prompt nutrition intervention to minimize the occurrence
of clinical complications and lower the risk of death. BIA
parameters are potential prognostic indicators, as changes
on the cellular level may occur prior to anthropometric and
biochemical changes,55 and they could be an important tool
for evaluating clinical outcome or for monitoring disease
progression.47
The SGA is not a specific indicator for patients on HD,
but it has been used in different studies involving HD
and is recommended for this population.53 SGA assesses
subjective aspects of nutrition status that are different
from the objective BIVA parameters, and that can be a
limitation. The small sample size and the number of patients
Nutrition in Clinical Practice 33(6)
excluded for not undergoing at least 2 evaluations with
the reference standards may have diminished the statistical
power of some results. However, the age and gender of
the individuals who remained in the study were similar to
those who were excluded. Nonetheless, comparisons of the
data should be made with caution, considering adult and
elderly patients undergoing HD 3 times a week. Moreover,
the sociodemographic characteristics and health conditions
of the patients evaluated should be taken into account.
The validity of the present findings is strengthened by the
fact that all researchers were trained for the establishment
of intraexaminer and interexaminer agreement and the ad-
equate standardization of the anthropometric measures and
administration of the reference standards for the screening
of malnutrition.
To the best of our knowledge, there are no previous
prospective longitudinal studies on the diagnostic accuracy
of BIVA parameters (R/H, Xc/H). Thus, the present in-
vestigation is important because the patients classified as
malnourished based on the SGA, MIS, and NRS 2002 had
at least 2 consecutive evaluations in which the classification
of malnutrition was given by the same indicator during
the study follow-up, which is strong evidence that these
individuals truly were and remained malnourished.
Conclusions
This study showed BIVA parameters R/H and Z/H offer
low to moderate accuracy in men and low accuracy in
women for the diagnosis of malnutrition using SGA as a
reference standard. However, BIVA parameters were cor-
related with SGA, NRS 2002, anthropometric variables,
HGS, and phase angle. R/H and Z/H carry high sensitivity
but low specificity compared with the reference standards.
Further studies using the specific BIVA approach can be
performed. The validation with physiologically relevant
reference standards (eg, dual-energy X-ray absorptiometry,
lean tissue measures) is necessary to determine if the BIVA
parameters R/H, Xc/H, and Z/H are appropriate bedside
tools for the assessment of nutrition status receiving HD
patients.
Acknowledgments
We are grateful to the patients and staff of Apar Vida and
Clinirim clinics in the region of Florianópolis, Southern Brazil,
where data collection was performed, and the Coordination for
the Improvement of Higher Education Personnel (CAPES) for
the scholarship to the researcher.
Statement of Authorship
A. T. da Silva, D. B. Hauschild, E. Wazlawik, and Y. M. Franco
Moreno contributed to conception/design of the research; A.
T. da Silva, D. B. Hauschild, E. Wazlawik, J. L. Dornelles
Bastos, and Y. M. Franco Moreno contributed to acquisition,
analysis, or interpretation of the data; A. T. da Silva and
da Silva et al
E. Wazlawik drafted the manuscript; A. T. da Silva, D. B.
Hauschild, Y. M. Franco Moreno, J. L. Dornelles Bastos, and
E. Wazlawik critically revised the manuscript; A. T. da Silva,
D. B. Hauschild, Y. M. Franco Moreno, J. L. Dornelles Bastos,
and E. Wazlawik agree to be fully accountable for ensuring
the integrity and accuracy of the work. All authors read and
approved the final manuscript.
References
1. Fouque D, Kalantar-Zadeh K, Kopple J, et al. A proposed nomen-
clature and diagnostic criteria for protein–energy wasting in acute and
chronic kidney disease. Kidney Int. 2008;73(4):391-398.
2. Beberashvili I, Sinuani I, Azar A, et al. Serum uric acid as a clinically
useful nutritional marker and predictor of outcome in maintenance
hemodialysis patients. Nutrition. 2015;31(1):138-147.
3. Komatsu M, Okazaki M, Tsuchiya K, Kawaguchi H, Nitta K. Geriatric
Nutritional Risk Index is a simple predictor of mortality in chronic
hemodialysis patients. Blood Purif. 2015;39(4):281-287.
4. Mancini A, Grandaliano G, Magarelli P, Allegretti A. Nutritional
status in hemodialysis patients and bioimpedance vector analysis. J Ren
Nutr. 2003;13(3):199-204.
5. Piccoli A, Codognotto M, Piasentin P, Naso A. Combined evalua-
tion of nutrition and hydration in dialysis patients with bioelectrical
impedance vector analysis (BIVA). Clin Nutr. 2014;33(4):673-677.
6. Norman K, Stobäus N, Pirlich M, Bosy-Westphal A. Bioelectrical
phase angle and impedance vector analysis—clinical relevance and
applicability of impedance parameters. Clin Nutr. 2012;31(6):854-861.
7. Buffa R, Saragat B, Cabras S, Rinaldi AC, Marini E. Accuracy of
specific BIVA for the assessment of body composition in the United
States population. PLoS One. 2013;8(3):e58533.
8. Garcia MF, Wazlawik E, Moreno YMF, Führ LM, González-Chica
DA. Diagnostic accuracy of handgrip strength in the assessment of
malnutrition in hemodialyzed patients. e-SPEN Journal. 2013;8:e181-
e186.
9. Buffa R, Mereu E, Comandini O, Ibanez ME, Marini E. Bioelec-
trical impedance vector analysis (BIVA) for the assessment of two-
compartment body composition. Eur J Clin Nutr. 2014;68(11):1234-
1240.
10. Locatelli F, Fouque D, Heimbürguer O, et al. Nutritional status
in dialysis patients: a European consensus. Nephrol Dial Transplant.
2002;17(4):563-572.
11. Piccoli A, Pastori G. BIVA software. In: Sciences DoMaS. Padova,
Italy: University of Padova; 2002.
12. Eickemberg M, Oliveira CC, Roriz AKC, Sampaio LR. Bioelec-
tric impedance analysis and its use for nutritional assessments
(Bioimpedância elétrica e sua aplicação em avaliação nutricional). Rev
Nutr. 2011;24(6):873-882.
13. Slee A, Birc D, Stokoe D. Bioelectrical impedance vector analysis,
phase-angle assessment and relationship with malnutrition risk in
a cohort of frail older hospital patients in the United Kingdom.
Nutrition. 2015;31(1):132-137.
14. Pupim LCB, Ribeiro CB, Kent P, et al. Update on dialysis: Use of
bioelectrical impedance analysis in dialysis patients (Atualização em
diálise: Uso da Impedãncia Bioelétrica em pacientes em diálise). J Bras
Nefrol. 2000;22(4):249-256.
15. Maggiore Q, Nigrelli S, Ciccarelli C, Grimaldi C, Rossi GA, Michelassi
C. Nutritional and prognostic correlates of bioimpedance indexes in
hemodialysis patients. Kidney Int. 1996;50(6):2103-2108.
16. Segall L, Moscalu M, Hogaş S, et al. Protein-energy wasting, as
well as overweight and obesity, is a long-term risk factor for mor-
tality in chronic hemodialysis patients. Int Urol Nephrol. 2014;46(3):
615-621.
841
17. Piccoli A, Piazza P, Noventa D, Pillon L, Zaccaria M. A new method
for monitoring body fluid variation by bioimpedance analysis: the RXC
graph. Kidney Int. 1994;46(2):534-539.
18. Kuchnia AJ, Teigen LM, Cole AJ, et al. Phase angle and impedance
ratio: reference cut-points from the United States National Health
and Nutrition Examination Survey 1999–2004 from bioimpedance
spectroscopy data (published online ahead of print September
26, 2016). JPEN J Parenter Enteral Nutr. 2016. https://www.
ncbi.nlm.nih.gov/pubmed/27670250. Accessed April 17, 2017.
19. Bosy-Westphal A, Danielzik S, Dorhofer R-P, Later W, Wiese S,
Muller M. Phase angle from bioelectrical impedance analysis: popula-
tion reference values by age, sex, and body mass index. JPEN J Parenter
Enteral Nutr. 2006;30(4):309-316.
20. Mulasi U, Kuchnia AJ, Cole AJ, Earthman CP. Bioimpedance at the
bedside: current applications, limitations, and opportunities. Nutr Clin
Pract. 2015;30(2):180-193.
21. Fürstenberg A, Davenport A. Comparison of multifrequency bio-
electrical impedance analysis and dual-energy X-ray absorptiometry
assessments in outpatient hemodialysis patients. Am J Kidney Dis.
2011;57(1):123.
22. Di Iorio BR, Scalfi L, Terracciano V, Bellizzi V. A systematic eval-
uation of bioelectrical impedance measurement after hemodialysis
session. Kidney Int. 2004;65(6):2435-2440.
23. National Institute of Health (NIH). Bioelectrical impedance analysis
in body composition measurement: National Institutes of Health tech-
nology assessment conference statement. Am J Clin Nutr. 1996;64(3
Suppl.):524S-532S.
24. Kushner RF. Bioelectrical impedance analysis: a review of principles
and applications. J Am Coll Nutr. 1992;11(2):199-209.
25. Piccoli A, Pillon L, Dumler F. Impedance vector distribution by sex,
race, body mass index, and age in the United States: standard reference
intervals as bivariate Z scores. Nutrition. 2002;18(2):153-167.
26. Baumgartner RN, Chumlea WC, Roche AF. Bioelectric impedance
phase angle and body composition. Am J Clin Nutr. 1988;48:
6-43.
27. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective
global assessment of nutritional status? JPEN J Parenter Enter Nutr.
1987;11:8-13.
28. Kalantar-Zadeh K, Kopple JD, Block G, Humphreys MH. A
malnutrition-inflammation score is correlated with morbidity and
mortality in maintenance hemodialysis patients. Am J Kidney Dis.
2001;38(6):1251-1263.
29. Yamada K, Furuya R, Takita T, et al. Simplified nutritional screening
tools for patients on maintenance hemodialysis. Am J Clin Nutr.
2008;87:106-113.
30. Kondrup J, Rasmussen HH, Hamberg O, Stanga Z; Ad Hoc
ESPEN Working Group. Nutritional risk screening (NRS 2002): a new
method based on an analysis of controlled clinical trials. Clin Nutr.
2003;22(3):321-336.
31. Lohman, TG. Advances in body composition assessment. IL: Human
Kinetics Publishers; 1992.
32. World Health Organization. Global database on Body Mass Index
http://apps.who.int/bmi/index.jsp?introPage=intro_3.html, 2008.
33. Siri WE. Body composition from fluid spaces and density: analysis
of methods. In: Brozek J, Henschel A. Techniques for Measuring
Body Composition. Washington DC: National Academy of Sciences;
1961:223-224.
34. Durnin JV, Womersley J. Body fat assessed from total body density
and its estimation from skinfold thickness: measurements on 481
men and women aged from 16 to 72 years. Br J Nutr. 1974;32:
77-97.
35. Frisancho AR. Triceps skin-fold and upper arm muscle size norms for
assessment of nutritional status. Am J Clin Nutr. 1974;27:1052-1058.
842
36. Leal VO, Mafra D, Fouque D, Anjos LA. Use of handgrip strength
in the assessment of the muscle function of chronic kidney disease
patients on dialysis: a systematic review. Nephrol Dial Transplant.
2010;26(4):1354-1360.
37. Headley S, Germain M, Mailloux P, et al. Resistance training improves
strength and functional measures in patients with end-stage renal
disease. Am J Kidney Dis. 2002;40(2):355-364.
38. Leal VO, Stockler-Pinto MB, Farage NE, et al. Handgrip strength
and its dialysis determinants in hemodialysis patients. Nutrition.
2011;27(11-12):1125-1129.
39. Schlüssel MM, Anjos LA, Vasconcellos MT, Kac G. Reference values
of handgrip dynamometry of healthy adults: a population-based study.
Clin Nutr. 2008;27(4):601-607.
40. Doumas BT, Watson WA, Biggs HG. Albumin standards and the
measurement of serum albumin with bromcresol green. Clin Chim
Acta. 1971;31(1):87-96.
41. Arango HG. Bioestatı́stica: Teórica e Computacional. 2nd ed. Rio de
Janeiro: Guanabara Koogan; 2005.
42. Swets JA. Measuring the accuracy of diagnostic systems. Science.
1988;240(4857):1285-1293.
43. Charney P. Nutrition screening vs nutrition assessment: how do
they differ? Nutr Clin Pract. 2008;23(4):366-372. https://doi.org/
10.1177/0884533608321131.
44. Correia MITD. Nutrition screening vs nutrition assessment: what’s
the difference? Nutr Clin Pract. 2017;(1):https://doi.org/10.1177/
0884533617719669.
45. Saragat B, Buffa R, Mereu E, et al. Specific bioelectrical impedance
vector reference values for assessing body composition in the Italian
elderly. Exp Gerontol. 2014;50:52-56.
Nutrition in Clinical Practice 33(6)
46. Jha V, Jairam A, Sharma MC, et al. Body composition analysis with
bioelectric impedance in adult Indians with ESRD: comparison with
healthy population. Kidney Int. 2006;69(9):1649-1653.
47. Barbosa-Silva MC, Barros AJ, Wang J, Heymsfield SB, Pierson RN Jr.
Bioelectrical impedance analysis: population reference values for phase
angle by age and sex. Am J Clin Nutr. 2005;82(1):49-52.
48. Camina Martı́n MA, de Mateo Silleras B, Barrera Ortega S,
Domı́nguez Rodrı́guez L, Redondo del Rı́o MP. Specific bioelectrical
impedance vector analysis (BIVA) is more accurate than classic BIVA
to detect changes in body composition and in nutritional status in
institutionalised elderly with dementia. Exp Gerontol. 2014;57:264-271.
49. Gonzalez MC, Barbosa-Silva TG, Bielemann RM, Gallagher D,
Heymsfield SB. Phase angle and its determinants in healthy subjects:
influence of body composition. Am J Clin Nutr. 2016;103:712-716.
50. Mereu E, Buffa R, Lussu P, Marini E. Phase angle, vector
length, and body composition. Am J Clin Nutr. 2016;104(3):845-847.
https://doi.org/10.3945/ajcn.116.137513.
51. Norman K, Smoliner C, Valentini L, Lochs H, Pirlich M. Is bioelectri-
cal impedance vector analysis of value in the elderly with malnutrition
and impaired functionality? Nutrition. 2007;23(7-8):564-569.
52. Marini E, Buffa R, Saragat B, et al. The potential of classic and
specific bioelectrical impedance vector analysis for the assessment of
sarcopenia and sarcopenic obesity. Clin Interv Aging. 2012;7:585-591.
53. National Kidney Foundation. K/DOQI clinical practice guidelines for
nutrition in chronic renal failure. Am J Kidney Dis. 2000;35(2):S1-S140.
54. Fletcher RH, Fletcher SW, Fletcher GS. Diagnóstico. In: Epidemiolo-
gia Clı́nica: Elementos Essenciais. Porto Alegre: Artmed; 2014:116-139.
55. Barbosa-Silva MC. Subjective and objective nutritional assessment
methods. Curr Opin Clin Nutr Metab Care; 2008;11(3):248-254.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Comparison of Two Methods for Estimating the Tip Position December 2018 843–850

C 2018 American Society for

of a Nasogastric Feeding Tube: A Randomized Controlled Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10112
Trial wileyonlinelibrary.com

Tim Torsy, RN, MSc1 ; Renée Saman, RN, CNS, MSc1,2 ; Kurt Boeykens, RN, CNS,
MSc2 ; Ivo Duysburgh, MD2 ; Nele Van Damme, RN, MSc3 ;
and Dimitri Beeckman, RN, MSc, PhD3

Abstract
Background: Several studies have shown that the nose-earlobe-xiphoid distance (NEX) is inadequate to estimate the insertion length
of nasogastric tubes. An alternative approach tested in these studies, which leads to > 90% well-placed nasogastric tubes, used a
corrected calculation of the NEX: (NEX × 0.38696) + 30.37. The primary aim of this study was to determine whether using the
corrected NEX was more successful than the NEX in determining the insertion length. The secondary aim was to investigate the
likelihood to successfully obtain gastric aspirate. Methods: Adult patients in a general hospital (N = 215) needing a nasogastric tube
were randomized to the control (NEX) or intervention group (corrected NEX). Tip positioning was verified using X-ray. Correct
tip positioning was defined as between 3 and 10 cm under the lower esophageal sphincter (LES). Results: In >20% of all patients,
both methods underestimated the required tube length for correct positioning. The NEX showed an overestimation (17.2%) of the
insertion length (>10 cm under the LES) compared with the corrected NEX (4.8%). There was no difference (P = 0.938) between the
2 groups in obtaining gastric aspirate (55.6% vs 56%). However, correct tip positioning resulted in a fourfold increase of obtaining
gastric aspirate. Conclusions: Both methods resulted in a high risk of incorrectly placed tubes due to malposition of the tip near the
LES or distal esophagus. This may increase the risk of reflux and pulmonary aspiration. Based on these results, the development
of more reliable methods requires consideration. (Nutr Clin Pract. 2018;33:843–850)

Keywords
enteral nutrition; gastrointestinal intubation; nasogastric tube; nutritional support; patient safety; X-rays

Introduction data about aspiration are difficult to determine due to vague

Enteral feeding via nasogastric tube is a frequently used
short-term method for patients who cannot eat or whose
oral nutrition intake is insufficient. In adults, the stomach
is a J-shaped organ with a greater curvature of 34 cm, a
maximum transverse diameter of 10 cm, a pyloric sphincter
diameter of 1.2 cm, and a capacity of 0.9 litres.1 A correct
position of the tip of the nasogastric tube inside the stomach
is required to ensure the safe administration of nutrients.2,3
Correct positioning will also prevent reflux and aspiration
of nutrients into the lungs. An incorrect gastric tip position
due to underestimation of the tube may lead to feeding
remaining in the esophagus because most tubes have several
lateral openings proximal of the distal tip opening. Con-
versely, overestimation (leading to a tube segment located
too deep into the stomach) can cause curling of the tube
inside the stomach or even migrating the tube back into the
esophagus or beyond the pylorus into the duodenum. Incor-
rect positioning may also prevent obtaining an aspirate for
pH measurement or gastric residual volume.4,5
Although pulmonary aspiration is 1 of the most common
complications in enterally fed patients, reliable prevalence
definitions, poor assessment methods, and varying levels of
clinical recognition. Pulmonary aspiration of feeding can
be caused by either pulmonary misplacement of the tube
or incorrect positioning of the tube in the gastrointesti-
nal tract. In 2005, incorrectly placed nasogastric tubes in
general were first recognized in England by the National
Patient Safety Agency (NPSA) as a patient safety issue.
Between September 2005 and March 2010, there were 21
deaths and 79 reported cases in England of harm related
From the 1 Department of Nursing, Odisee University College,
Brussels, Belgium; 2 AZ Nikolaas General Hospital, Sint-Niklaas,
Belgium; and 3 University Centre for Nursing and Midwifery, Ghent
University, Ghent, Belgium.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on June 30, 2018.
Corresponding Author:
Torsy Tim, RN, MSc, Odisee University College, Department of
Health (Nursing and Midwifery), Hospitaalstraat 23, 9100
Sint-Niklaas, Belgium.
Email: tim.torsy@odisee.be
844
to feeding through misplaced nasogastric tubes. In 2009,
inadequate feeding related to a misplaced nasogastric tube
became a never event, which is defined as a serious, largely
preventable patient safety incident that should not occur if
the available preventative measures have been implemented
by healthcare providers. Despite all guidance, during 2009–
2010 there were 41 never events reported to the NPSA in
which a misplaced nasogastric tube was not detected prior
to use. Even at this time, misplaced nasogastric tubes are
never events: Between September 2011 and March 2016, a
total of 95 incidents were reported. These incidents show
that, despite the sensitization of all stakeholders, risks to
patient safety persist.6
Most often, the nose–earlobe–xiphoid (NEX) method is
used to determine the insertion length of a nasogastric tube.
This measures the length from the tip of the nose to the
earlobe to the xiphoid.7,8 This simple and popular method
was introduced by Royce, Tepper, Watson, and Day in 1951.
They described a 6-month experience of estimating the
appropriate tube length using external anatomical reference
points in neonates. Subsequently, the NEX method was
recommended as a gold standard for neonates, children,
and adults in nursing textbooks without further research
or reflection. The reliability of the NEX method in both
adults and pediatric populations, however, has been ques-
tioned several times.8-10 Using the NEX, Ziemer and Carroll
noticed that a deeper insertion was often needed to obtain
a gastric aspirate.7
After a study on healthy volunteers and a postmortem
study including 99 cases, Hanson described alternative
formulas to calculate the nasogastric tube length in a more
reliable way.2 Two formulas were proposed as a correction
of the NEX (method 1 and 2), and 1 formula was based on
body length (method 3):
Method 1: ([NEX – 50 cm]/2) + 50 cm
Method 2: (NEX × 0.38696) + 30.37
Method 3: (body length × 0.20239) + 17.07
Method 2 resulted in more nasogastric tubes being
placed correctly inside the stomach (92.3% vs 91.35%), with
a tip position ranging from 1 to 10 cm under the lower
esophageal sphincter (LES) being optimal.2 For that reason,
method 2 was selected as the experimental intervention
in this current randomized controlled trial. Ellett et al
already had reported that the minimum insertion length of
the nasogastric tube corresponds to the distance from the
tip of the nose to the gastroesophageal junction, thereby
adding the distance from the tip of the tube to the most
proximal lateral opening on the tube to ensure that each
lateral opening was located inside the stomach.3 The ac-
tual required length of the nasogastric tube according to
either control or intervention group was determined just
before the placement procedure starting from the distal tip
opening of the tube. To consider a tube as correctly placed
Nutrition in Clinical Practice 33(6)
inside the stomach, the presence of the lateral openings
was taken into account (as described in the interventions
and outcomes section of this article). The target popu-
lation of this study were adults because a method for
newborns and children (using the distance between the tip
of the nose to the earlobe to the mid-umbilicus) already
exists.9,10
The primary aim of this first prospective randomized
trial was to determine whether the use of the Hanson
formula (NEX × 0.38696) + 30.37 was more successful
than using the NEX method. The secondary aim was to
investigate the likelihood to obtain a gastric aspirate to
perform pH measurement.
Methods
Study Design
This study was designed as a blinded prospective random-
ized trial. The data were collected between December 2015
and September 2016 in a Belgian general hospital on both
intensive care units (ICU) and non ICUs.
Participants
Patients were eligible to participate if they met the following
criteria: adult (aged ࣙ 18 years) and a medical indication
for nasogastric tube feeding. Patients were excluded when
the xiphoid was not palpable or if they had documented
surgical or anatomical abnormalities of the esophagus or
stomach (eg, hiatus hernia). Written informed consent was
obtained from all patients or their legal representatives in
case of diminished awareness or coma.
Participants were randomized to either the control or the
intervention group using block randomization.11 Random-
ization was performed by an investigator with no clinical
involvement in the trial. A randomization list with 2 groups
(control and intervention group), block sizes of 6 patients,
and an actual list length of 240 participants (control group
and intervention group) was created.
After assignment to 1 of the study groups, blinding was
performed at the following levels:
Included patients were kept blinded to their allocation.
Data collectors (clinical nutrition nurse specialists) could
not be blinded. They were aware of the allocated arm
to perform the appropriate intervention.
Outcome adjudicators (radiologists) were also kept
blinded to minimize bias and maximize the validity
of the results.
Interventions and Outcomes
Two experienced clinical nutrition nurse specialists indepen-
dently measured the NEX distance (Figure 1) in all patients.
Torsy et al
Figure 1. NEX distance.
NEX, nose-earlobe-xiphoid distance.
The 2 measurements were taken within a short space of
time according to the following procedure: the patient was
in supine position with the chest at an angle of 30° and
facing forward. The chest was uncovered, and the xiphoid
was marked. A measuring tape with an accuracy up to half
a centimeter was directed in a straight line from the tip of
the nose to the earlobe and then to the marked xiphoid
(Figure 1). A nasogastric tube was subsequently inserted.
The nasogastric tube used in the study was a
polyurethane radiopaque feeding tube (Nutricia Flocare
ENFit; CH10 or CH14, total length 110 cm), manufactured
by Danone Trading B.V., Schiphol, The Netherlands and
consisting following characterisitcs: An internal guidewire,
a distal opening in the tip, and lateral openings up to 2 cm
from the distal tip. The guidewire ended 5 cm before the
distal tip opening. Newly placed tubes were inserted by
the 2 clinical nutrition nurse specialists according to the
randomization criteria. If the tubes were already in situ but
not yet used for tube feeding, they were only repositioned
but not replaced, again according to the randomization
criteria. A safety margin of 1 cm was added so a correct
tip position of the nasogastric tube inside the stomach was
defined as somewhere between 3 and 10 cm under the LES.
After insertion or repositioning, the tube was fixed to the
nose with a fixation tape for nasal tubes manufactured by
ConvaTec, Deeside, UK (ConvaTec Naso-Fix).
A gastric aspirate was obtained immediately after inser-
tion or repositioning of the nasogastric tube because pH
measurement is considered an alternative bedside method
to X-ray to check the position of a nasogastric tube.11
A large syringe (60 mL) with an ENFit (Danone Trading
B.V., Schiphol, The Netherlands) connection was used for
aspiration. An initial attempt of obtaining gastric aspirate
occurred before the removal of the guidewire. If no aspirate
could be obtained, an extra attempt was made 1 hour after
845
Figure 2. Flowchart for evaluating the X-ray images.
Cm, centimeter; VOD, vault of the diaphragm.
removal of the guidewire. Gastric aspirate was tested for pH
using pH color-indicator strips (pH 2.0–9.0) with a 0.5 pH
interval scale (Merck KGaA, Darmstadt, Germany). If an
aspirate could be obtained, a pH value ࣘ 5.5 indicated that
it was indeed gastric aspirate.12
A chest X-ray of the patient, positioned in supine posi-
tion as described above, was taken in situ to determine the
tip position of the tube. Preferably, the guidewire was left
inside or put back inside the tube to increase the visibility
of the nasogastric tube on X-ray. The chest X-ray images
covered the area from the apex to the base of the lungs,
including the diaphragm and the stomach below. All of the
images were retrieved from the hospital picture archiving
and communication system, processed and anonymized
using DicomCleaner (PixelMed Publishing, Bangor, PA),
and then evaluated on medical imaging monitors.
Instruments
The chest X-ray images were independently evaluated
by 3 different radiologists using a predefined evaluation
flowchart (Figure 2). The authors and the 3 participating
radiologists used the vault of the diaphragm as the reference
point for the location of the LES in the assessment of
the images because the LES itself is not very clear on
an X-ray.
“Obtainment of gastric aspirate” was considered as neg-
ative after 2 attempts (as described above) without retrieving
gastric content within 1 hour after insertion. A few drops of
the aspirate were placed on a pH indicator strip.
846
Statistical Methods
Statistical analyses were conducted using SPSS 24.0 (IBM
Corporation, Armonk, NY). Categorical variables were
presented as frequencies (percentages); normally distributed
continuous variables were described using means and stan-
dard deviations (SD); and between-group comparisons were
performed using independent samples t test. Comparison
of categorical data between groups was performed using
χ 2 test.
The interobserver agreement between the 3 independent
radiologists was assessed using κ statistics. Observer agree-
ment was defined as 2 or more concurring interpretations
of the X-ray being tip <3 cm under the LES, between 3
and 10 cm under the LES, > 10 cm under the LES, or
tip not visible on X-ray. To obtain an overall κ value of
the assessments, Fleiss κ was calculated. The classification
system used by Coblentz et al was used to define the strength
of agreement for the κ coefficient.13
The association between the study group and the out-
come was controlled for age, gender, body length, and cogni-
tive awareness using a multivariate logistic binary regression
analysis. The variable “gastric aspirate obtained” was added
to the multivariate model to analyze how this variable was
related to the correct positioning of the nasogastric tube.
Hosmer-Lemeshow statistic was calculated as a measure of
model fit. A significance level of P < 0.05 was applied.
Ethical Statement and Trial Registering
All patients (or their legal representative in case of dimin-
ished awareness or coma) gave informed consent. Collecting
the informed consents was done by the clinical nutrition
nurse specialists. Patients or their representatives who were
unable or unwilling to sign the written informed consent
were excluded from the study.
The study protocol was approved by the Review Com-
mittee for Medical Ethics from AZ Nikolaas (East Flan-
ders, Belgium) (EC15053). The study was registered at
ISRCTN.com (ISRCTN18322407).
Results
Patient Characteristics
A total of 215 patients participated in the study. The quality
of the X-ray images of 32 patients was insufficient to
allow conclusive determination of the position of the tip of
the nasogastric tube. Although the nasogastric tubes were
visible on all X-ray images, the images had poor margin
settings and thus the tip was not visible on the image
and could not be evaluated. Therefore, after the procedure
of taking X-ray images, data of only 183 persons were
used for further analysis: 99 patients were included in the
control group and 84 patients in the intervention group.
Nutrition in Clinical Practice 33(6)
Characteristics of the control and intervention group were
comparable (P > 0.05), as shown in Table 1.
Tip Position and Gastric Aspirate
An overall significant difference (P = 0.032) was demon-
strated between the 2 groups regarding the positioning of
the tip. The tip of the nasogastric tube was situated <3 cm
under the LES in 20.2% of the tubes in the control group
and in 22.6% in the intervention group (P = 0.691). A
significant difference (P = 0.009) concerning overestimation
of the tip position was found between the control group
(17.2%) and the intervention group (4.8%). There was
no statistically significant difference (P > 0.05) between
both methods for external length determination and the
chance of successfully obtaining gastric aspirate through
the nasogastric tube. Absolute values, percentages, and P
values are shown in Table 2. The interobserver reliability of
the X-ray evaluation between the 3 radiologists indicated
acceptable agreement (κ = 0.44, 95% confidence interval
[CI], 0.39–0.49).
Figure 3 shows the distribution of the nasogastric tube
insertion lengths using either the NEX method or Hanson’s
formula-corrected NEX distance. Using the NEX method,
the mean insertion length was 51.62 cm (SD 3.79) and the
interquartile range (IQR) was 4 with a minimum insertion
length of 40.5 cm and a maximum insertion length of
59.5 cm. Using (NEX × 0.38696) + 30.37, the mean
insertion length was 50.50 cm (SD 1.62) and an IQR of 1.95
with a minimum insertion length of 43.5 cm and a maximum
insertion length of 53.0 cm. The distribution of the insertion
lengths was significantly lower (P = 0.009) using Hanson’s
formula-corrected NEX distance.
Influencing Factors
Five potential influencing factors in aiming for correct gas-
tric tip positioning were analyzed in a multivariate binary
logistic regression analysis: age, gender (male/female), body
length, method of length determination (“NEX”/“(NEX
× 0.38696) + 30.37”) and cognitive awareness level
(aware/diminished awareness/comatose) (Table 3). The
binary outcome variable was “tip position of the nasogastric
tube inside the stomach” (<3 cm under LES = incorrect
position; ࣙ 3 cm under LES = correct position). The
variables “weight” and “body mass index (BMI)” were not
included in the model due to the large number of missing
values (81 of 183, 44.3%) and their mutual multicollinearity.
The variable “gastric aspirate after placement (yes/no)”
was also included in the model to explore the differences
between the 2 groups concerning this variable. The model
resulted in a significant association (P = 0.001) between
the outcome variable “tip position” and the obtainment
of gastric aspirate (odds ratio 4.109, 95% CI 1.785–9.458).
There is no discrepancy between the model predicted and
Torsy et al 847

Table 1. Characteristics of Participants (N = 183).a

Control Group Intervention Group No Data Available

Characteristics (n = 99) (n = 84) P Value no. (%)

Demographic variables
Age, mean ± SD, years 72.39 ± 12.33 70,62 ± 12.65 0.339
Gender
Male, no. (%)a 52 (52.5) 48 (57.1) 0.532
Female, no. (%)a 47 (47.5) 36 (42.9) 0.532
Clinical variables
Body length, mean ± SD, cm (no.) 165.37 ± 9.09 (97) 164.83 ± 8.22 (84) 0.675 2 (1.1)
Weight, mean ± SD, kg (no.) 70.59 ± 17.99 (61) 66.36 ± 14.52 (41) 0.193 81 (44.2)
BMI, mean ± SD, kg/m² (no.) 25.63 ± 5.74 (61) 24.35 ± 5.01 (41) 0.249 81 (44.2)
Use of antacids, no. (%)b 78 (78.8) 61 (72.6) 0.407 3 (1.6)
Unit
ICU, no. (%)b 50 (50.5) 44 (52.4) 0.800
Non-ICU, no. (%)b 49 (49.5) 40 (47.6) 0.800
Cognitive awareness level
Aware, no. (%)b 43 (43.4) 29 (34.5) 0.219
Diminished awareness, no. (%)b 8 (8.1) 14 (16.7) 0.075
Comatose, no. (%)b 48 (48.5) 41 (48.8) 0.965

BMI, body mass index; ICU, intensive care unit; no., number; SD, standard deviation.
a Data collected by clinical nutrition nurse specialists
b Percentages within group

Table 2. Localization of Nasogastric Tubes in Both Groups and Obtainment of Gastric Aspirate (N = 183).

Control Group (n = 99) Intervention Group (n = 84) P Value

Tip positioning
< 3 cm under LES, no. (%) 20 (20.2) 19 (22.6) 0.691
3–10 cm under LES, no. (%) 62 (62.6) 61 (72.6) 0.151
> 10 cm under LES, no. (%) 17 (17.2) 4 (4.8) 0.009
Obtaining gastric aspirate,a no. (%) 55 (55.6) 47 (56.0) 0.938

LES, lower esophageal sphincter; no., number.

a After placement of the tube.

the effectively observed probabilities (Hosmer-Lemeshow

2.228, degrees of freedom 8, P = 0.973).

Discussion
This study is the first prospective randomized trial in which
the reliability of the NEX method and Hanson’s formula
has been tested. Both methods are inaccurate when aiming
for a correct tip position of 3 or more cm under the LES.
When using the NEX, the range of tip positions is larger
compared with Hanson’s formula (Figure 3). The chance of
obtaining an aspirate was low (around 56%) in both groups
(Table 2).
One of the key findings from this study was that in both
trial groups in approximately 1 of 5 patients (20.2% in the
control and 22.6% in the intervention group), the tip of the
tube was located <3 cm under the LES. This means that at Figure 3. Boxplot describing the distribution of insertion
least 1 lateral opening of the nasogastric tube was located lengths (in centimeters).
in the esophagus or at the same height as the LES. This can NEX, nose-earlobe-xiphoid distance.
848 Nutrition in Clinical Practice 33(6)

Table 3. Multivariate Binary Logistic Regression Between Possible Influencing Factors in Aiming for a Correct Tip Position.

β Coefficient Standard Error Wald Statistic df OR (95% CI) P Value

Age −0.007 0.020 0.135 1 0.993 (0.956 – 1.032) 0.714

Gender 0.614 0.555 1.222 1 1.847 (0.622 – 5.481) 0.269
Malea
Female
Body length 0.031 0.035 0.779 1 1.031 (0.963 – 1.103) 0.377
Gastric aspirate 1.413 0.425 11.037 1 4.109 (1.785 – 9.458) 0.001
Yes
Noa
Method of length determination 0.332 0.402 0.684 1 1.394 (0.634 – 3.063) 0.408
NEX
(NEX × 0.38696) + 30.37a
Awareness levela 0.634 2 0.728
Awareness level(1) 0.191 0.442 0.187 1 1.211 (0.509 – 2.883) 0.665
Awareness level(2) 0.542 0.695 0.609 1 1.720 (0.441 – 6.711) 0.435
Constant −4.400 6.438 0.467 1 0.012 0.494

CI, confidence interval; df, degrees of freedom; NEX, nose-earlobe-xiphoid distance; OR, odds ratio.
a Reference category.

lead to regurgitation and an increased risk of pulmonary
aspiration.14
The NEX method is the most commonly used technique
for measuring the nasogastric tube length but has been
questioned in the literature.8-10 After a postmortem study
including 99 cases, Hanson concluded that the NEX method
resulted in 72% well-placed nasogastric tubes, with the ideal
tip position between 1 and 10 cm under the LES.2 Using
the 2 Hanson formulas adjusting the NEX, the number of
properly placed tubes increased to around 92%; however,
approximately 3% still were located in the distal esophagus,
which was about the same for the NEX (2%). Note that
Hanson’s study did not mention any lateral openings in the
nasogastric tubes, now common in most tubes. Therefore,
the tube must be inserted sufficiently deep into the stomach
to ensure that not only the tip but also all lateral openings
are located inside the stomach to prevent tube feeding
entering the esophagus. In our study, the presence of lateral
openings may explain why the percentage of insufficiently
deep placed nasogastric tubes exceeded Hanson’s. Hanson
also stated that the NEX might lead to an overestimation
of the required length, resulting in a placement >10 cm
under the LES in 25% of cases compared with <6% using
Hanson’s formulas. Our study reached the same conclusion,
with 17.2% of the tubes located >10 cm under the LES in
the control group and 4.8% in the intervention group.
In a study with a cross-sectional design, Ellett et al
tested the reliability of 1 of the formulas proposed by
Hanson (([NEX – 50 cm]/2) + 50 cm) compared with
the NEX.3 A correct tip position was defined as between
3 and 10 cm under the LES. In the NEX group, 9 out
of 21 malpositioned tubes were placed insufficiently deep
vs 11 of 19 tubes in the Hanson group. Additionally,
24 potentially predicting variables were included in the
data collection and analysis in order to develop a new
method for external length determination. This resulted in a
3-variable model of calculation based on gender, weight,
and the distance between nose and umbilicus in supine
position with the head flat (GWNUF), in which gender was
“1” if the participant was male and “0” if the participant
was female: 29.38 + (4.53 × gender) + (0.34 × NUF) –
(0.06 × weight). The GWNUF resulted in 85.3% of correct
positions (vs 72% for the NEX and 84.2% for Hanson)
but here also 5 of 11 malpositioned tubes were in the
esophageal danger zone. In addition to increasing the risk
of feeding in the esophagus due to a malpositioned tip
of the nasogastric tube, the feasibility of the GWNUF in
clinical practice can also be questioned for other reasons: the
difficulty of obtaining the correct weight of bedridden pa-
tients; discomfort for the patients to maintain the required
position; and higher likelihood to make clinical errors when
using different parameters to estimate the insertion length
in clinical practice. The combination of these elements,
together with the large number of included ICU patients
in the study, ensured that this method was not selected as
the preferred intervention in the study. It may not possible
to obtain an accurate weight in critically ill patients who
usually are also ventilated. In addition, the positioning of
ICU patients as described above is not always possible due
to their physical condition.
The study of Boeykens et al. (2014) mainly fo-
cused on pH measurement as an alternative method for
X-ray to confirm the correct position of the nasogastric
tube.12 In Boeykens’ study, the probability of a correctly
placed nasogastric tube was 98.9% when taking pH mea-
surement as criterion. Metheny et al stated that obtaining
Torsy et al
gastric aspirate and pH measurement is not useful to detect
placement of the nasogastric tube in the esophagus because
of possible gastric reflux.15,16 Therefore, caution should
be exercised in interpreting a pH measurement on gastric
aspirate. Our study showed that a correctly positioned tip
inside the stomach increased the possibility to obtain gastric
aspirate by almost 4 times. In our study, >20% of all
nasogastric tubes were placed too shallow whether the NEX
or Hanson’s formula was used to determine the insertion
length. Therefore, it is recommended to combine either
method for external length determination with X-ray to
verify correct placement of the tip.
Taylor et al suggested the use of NEX + 10 cm as
an alternative to the NEX method, which would lead to
a reduction from 16% to 7% of misplaced nasogastric
tubes, with the tip located into the esophagus. This small
study (N = 36) using electromagnetic guidance and not
X-ray was solely conducted in an intensive care setting in
a population of mainly neurosurgery and trauma patients
with gastric emptying problems whose stomachs possibly
dilate more than in other patients.5 This method made
it possible to guide the tip of the tube in the stomach
immediately. In earlier studies2,3 as well as in this current
study, an overestimation of the NEX was observed; thus,
the NEX + 10 cm potentially could result in tubes migrating
postpyloric, curling up/kinking inside the stomach—or even
worse, migrating up and re-entering the esophageal danger
zone.
There are some limitations in this prospective random-
ized trial. Firstly, there was no registration of the number of
excluded patients with previously documented hiatus hernia
during the period of data collection. After including all
patients in the study, there was also a 14.9% patient dropout
due to the poor quality of the chest X-ray images (tube or
tip not visible). Secondly, there was no separate registration
of data obtained by the 2 clinical nutrition nurse specialists
measuring the NEX distances, and thus no conclusions
can be drawn about interrater variability. Finally, the κ
value (κ = 0.44) of the interobserver agreement between the
radiologists can also be questioned. The clinical significance
of this value depends on its context. In a review17 of medical
imaging literature regarding imaging for expressing observer
agreement with regard to categorical data, the conclusion
was made that a κ value ࣙ 0.4 could be considered an
acceptable level of observer variability. For this reason,
the classification system used by Coblentz at al., in which
agreement is quantified as poor, fair, moderate, good, or
excellent, was followed.13
Despite the fact that Hanson’s formula in our study led
to 72.6% correctly placed nasogastric tubes, with the tip
located somewhere between 3 and 10 cm under the LES,
there were still outliers (both over- and underestimation
of the insertion length) that could not be associated with
any of the variables of the data collection. Despite all,
849
this study again emphasizes the importance of strict and
clear guidance on preventing potential incidents that may
lead to harm or death. Therefore, further research should
focus on determining other possible influencing factors
(e.g.k anatomical characteristics) for a more correct gastric
tip positioning in order to develop an easy, safe, and
clinically feasible method for estimating the insertion length
of nasogastric tubes.
Conclusions
Both the NEX-method and Hanson’s formula are not reli-
able to determine the appropriate length of the nasogastric
tube in order to obtain a correct tip position between 3
and 10 cm under the LES. At the same time, differences
in the likelihood of obtaining gastric aspirate could not
be demonstrated between the 2 methods. In >20% of
patients, the tip of the nasogastric tube was located in
the esophageal danger zone. To reduce the risk of mis-
placement, and consequently also reduce the inconvenience
caused to patients due to repositioning of the tube, a more
adequate method of determining the appropriate length of
the nasogastric tube should be developed. Meanwhile, it is
strongly recommended to confirm placement of the tube tip
through X-ray and reposition the tube when necessary. This
will minimize the risk of feeding remaining in the esophagus
and of possible reflux.
Statement of Authorship
T. Torsy, R. Saman, K. Boeykens and I. Duysburgh equally
contributed to the conception and design of the research; D.
Beeckman and N. Van Damme contributed to the design of the
research; T. Torsy, R. Saman, K. Boeykens and N. Van Damme
contributed to the acquisition and analysis of the data; T. Torsy,
D. Beeckman and N. Van Damme contributed to the inter-
pretation of the data; and T. Torsy and R. Saman drafted the
manuscript. All authors critically revised the manuscript, agree
to be fully accountable for ensuring the integrity and accuracy
of the work, and read and approved the final manuscript.
Aknowledgements
The authors would like to thank Leo Verguts, Karline
Schutyser, Dirk De Backer, Kurt Van Belle and Patrick Palmans
for their valuable contribution to this study and Leen Trommel-
mans for assisting with the final draft. They also would like to
thank Odisee University College and the general hospital AZ
Nikolaas for facilitating this study.
References
1. Ferrua MJ, Singh RP. Modeling the fluid dynamics in a human
stomach to gain insight of food digestion. J Food Sci 2010;75:
151–162.
2. Hanson RL. Predictive criteria for length of nasogastric tube insertion
for tube feeding. JPEN J Parenter Enteral Nutr 1979;3:160–163.
850
3. Ellett MLC, Beckstrand J, Flueckiger J, Perkins SM, Johnson CS.
Predicting the insertion distance for placing gastric tubes. Clin Nurs
Res 2005;14:11–27.
4. Chen Y, Wang L, Chang Y et al. Potential risk of malposition
of nasogastric tube using nose-ear-xiphoid measurement. PLoS One
2014;9:e88046.
5. Taylor SJ, Allan K, McWilliam H, Toher D. The ‘NEX’ guideline is
incorrect. Br J Nurs 2014;23:641–644.
6. National Patient Safety Agency (NPSA). Patient Safety Alert:
Nasogastric tube misplacement: continuing risk of death and severe
harm. Updated July 22, 2016. Available at: https://improvement.nhs.uk/
news-alerts/nasogastric-tube-misplacement-continuing-risk-of-death-severe-harm.
Accessed March 2, 2018.
7. Fitzpatrick JJ, Wallace M. Encyclopaedia of Nursing Research 3rd ed.
New York, NY: Springer Publishing; 2011.
8. Santos SC, Woith W, Freitas MI, Zeferino EB. Methods to determine
the internal length of nasogastric feeding tubes: an integrative review.
Int J Nurs Stud 2016;61:95–103.
9. Beckstrand J, Ellett MLC, McDaniel A. Predicting the internal dis-
tance to the stomach for positioning nasogastric and orogastric feeding
tubes in children. J Adv Nurs 2007;59:274–289.
Nutrition in Clinical Practice 33(6)
10. Ellett MLC, Cohen MD, Perkins SM, Smith CE, Lane KA, Austin JK.
Predicting the insertion length for gastric tube placement in neonates.
J Obstet Gynecol Neonatal Nurs 2011;40:412–421.
11. Sealed Envelope Ltd. Create a blocked randomi-
sation list. Updated January 1, 2015. Available at:
https://www.sealedenvelope.com/simple-randomiser/v1/lists. Accessed
October 30, 2015.
12. Boeykens K, Steeman E, Duysburgh I. Reliability of pH measurement
and the auscultatory method to confirm the position of a nasogastric
tube. Int J Nurs Stud 2014;51:1427–1433.
13. Coblentz CL, Babcook CJ, Alton D, Riley BJ, Norman G. Observer
variation in detecting the radiologic features associated with bronchi-
olitis. Invest Radiol 1991;26:115–118.
14. Metheny NA, Titler MG. Assessing placement of feeding tubes. Am
J Nurs 2001;101:36–45.
15. Metheny NA, Reed L, Wiersema L, McSeeney M, Wehrle MA,
Clark J. Effectiveness of pH measurements in predicting feeding tube
placement: an update. Nurs Res 1993;42:324–331.
16. Metheny NA, Clouse RA, Clarck JM, Reed L, Wehrle MA, Wiersema
L. pH testing of feeding-tube aspirates to determine placement. Nutr
Clin Pract 1994;9:185–190.
17. Kundel HL, Polansky M. Measurement of observer agreement. Radi-
ology 2003;228:303–308.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Long-Term Use of Mixed-Oil Lipid Emulsion in Adult Home December 2018 851–857

C 2018 American Society for

Parenteral Nutrition Patients: A Case Series Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10014
wileyonlinelibrary.com

Manpreet S. Mundi, MD1 ; Megan T. McMahon, PA-C2 ;

Jennifer J. Carnell, PharmD2 ; and Ryan T. Hurt, MD, PhD1,2,3,4

Abstract
Background: Despite providing significant benefits, home parenteral nutrition (HPN) can be associated with complications such
as infections, intestinal failure–associated liver disease, and metabolic abnormalities. Soybean oil (SO)–based intravenous lipid
emulsion (ILE) has been noted to contribute to some of these complications, leading to the development of alternative sources
of ILE. Mixed-oil (MO) ILE has recently been approved for use in adults with short-term studies revealing a benefit over SO
ILE. Currently there is a paucity of data regarding long-term use in the HPN population. Methods: The current study reports our
experience with MO ILE in 9 HPN patients. Results: A total of 9 patients (3 women and 6 men) with average age of 48.1 ± 15.1 years
and a median HPN use of 9.9 years (9.0 months–30.7 years) were transitioned from SO ILE to MO ILE as a result of intolerance.
The 9 patients tolerated MO ILE well for 140.7 ± 29.7 days. The percentage of calories provided through ILE increased from 7.6 ±
6.5% to 18.4 ± 8.2% (P = .003), whereas the dextrose decreased from 66.9 ± 8.4% to 56.9 ± 5.5% (P = .0007). Although statistical
significance was not reached, there was a trend toward improvement in alkaline phosphatase from 138.0 (52–884) to 106 (47–512;
P = .09). Conclusion: MO ILE was well tolerated in this small cohort and led to improvement in the macronutrient composition
of HPN while providing a trend toward improvement in liver studies. These results are promising; however, additional randomized
control trials are needed to delineate the true benefit. (Nutr Clin Pract. 2018;33:851–857)

Keywords
fatty acids; home parenteral nutrition; intravenous fat emulsions; intravenous lipid emulsions; parenteral nutrition; parenteral
nutrition solutions

Introduction liest signs of IFALD can be lab abnormalities, including

Studies have estimated between 25,000–35,000 patients use
parenteral nutrition (PN) at home as a result of intestinal
failure.1,2 With appropriate training, home PN (HPN) can
be provided safely by patients or family members, allowing
them to receive adequate nutrition and electrolytes outside
the hospital setting. However, with long-term use, compli-
cations such as infection, lipid abnormalities, and intesti-
nal failure–associated liver disease (IFALD) can arise.3,4
Lipid profile changes such as a decrease in high-density
lipoprotein and an increase in low-density lipoprotein and
triglyceride levels can also be seen after the initiation of
HPN.5 Liposomes, which are created from excess phospho-
lipid emulsifier, as well as triglyceride content may play a
significant role given that 10% intravenous lipid emulsions
(ILEs) tend to have greater lipid profile changes when
compared with 20% ILE.3,6
IFALD may present as hepatic steatosis, cholestasis,
cholelithiasis, and hepatic fibrosis and can be seen in as high
as 30%–60% of children and 15%–40% of adults requiring
long-term PN.7-11 Especially in children, some of the ear-
elevations in bilirubin and alkaline phosphatase, followed by
elevations in aspartate aminotransferase (AST) and alanine
aminotransferase (ALT).9,12 Although the exact etiology
is not known, factors such as infections, inflammation,
From the 1 Division of Endocrinology, Diabetes, Metabolism and
Nutrition, Mayo Clinic, Rochester, Minnesota, USA; 2 Division of
General Internal Medicine, Mayo Clinic, Rochester, Minnesota,
USA; 3 Division of Gastroenterology and Hepatology, Mayo Clinic,
Rochester, Minnesota, USA; and the 4 Division of Gastroenterology,
Hepatology and Nutrition, University of Louisville, Louisville,
Kentucky, USA.
Financial disclosure: None declared.
Conflicts of interest: Ryan T. Hurt is a consultant for Nestlé.
Manpreet S. Mundi, Megan T. McMahon, and Jennifer Carnell have
no financial disclosure or conflict of interest to report.
This article originally appeared online on February 13, 2018.
Corresponding Author:
Manpreet S. Mundi, MD, Division of Endocrinology, Diabetes,
Metabolism, and Nutrition, Mayo Clinic, 200 First St. SW, Rochester,
MN 55905.
Email: mundi.manpreet@mayo.edu
852
decreased bile flow, and excess calories especially from high
dextrose and lipids have been felt to be contributory.5,9,13
In addition to excess calories, the high n-6 fatty acid and
phytosterol content of soybean oil (SO)–based ILE are
also thought be associated with IFLAD.3 Based on this,
strategies for treatment have included use of ursodiol,
cycling of HPN, and a reduction or discontinuation of SO
ILE. In fact, many programs including ours provide SO ILE
3 times per week as a preventive measure.3,13
In addition, attempts have been made with subsequent
generations of ILE to reduce the SO lipid content. Novel
ILEs have also focused on increasing the n-3 to n-6 fatty
acid ratio through the use of alternative fatty acid sources
such as medium-chain triglycerides, olive oil, and most
recently fish oil (FO). In terms of FO containing ILE, a
formulation composed entirely of FO (Omegaven, Frese-
nius Kabi, Bad Homburg, Germany) is available in Europe
but is not approved for use or commercially available in the
United States. This ILE is available under a compassionate
use allowance through the Food and Drug Administration
(FDA) for the treatment of IFLAD. The FDA has recently
approved a mixed-oil (MO) ILE for adults (Smoflipid,
Fresenius Kabi, Bad Homburg, Germany), which is com-
posed of a combination of fatty acid sources (30% SO,
30% medium-chain triglycerides, 25% olive oil, and 15%
FO). Short-term trials (14–28 days) in adults and pediatric
patients from various cohorts have shown benefit over SO
ILE in terms of inflammatory markers, fatty acid profile,
and liver function studies as well as bilirubin levels.14-18
Despite these promising results, the impact of FO ILE
beyond 28 days is not fully known. To assess tolerability
and impact on liver function studies and total bilirubin, we
prospectively followed adult patients placed on MO ILE
as a result of an intolerance of SO ILE. In addition, we
assessed the impact of transitioning to MO ILE on the
macronutrient composition of PN, with the hypothesis that
transitioning to MO ILE containing PN would be better
tolerated based on previous studies and would allow for
improved macronutrient composition.
Methods
Our program initiates HPN in 90–100 adult patients annu-
ally and closely follows 110–140 patients on HPN per year.
Patients are usually inpatients at the time of training and
initiation of HPN. On discharge, patients typically receive
PN with total daily calories between 20–35 kcal/kg depend-
ing on current body mass index based on major guidelines.19
Amino acids range between 0.8–1.0 g/kg for unstressed
patients to as high as 1.5–2.0 g/kg for patients with signifi-
cant fistulae, wounds, or renal dialysis dependence. Overall
composition of macronutrients typically leads to 50%–60%
of calories from dextrose, 15%–20% from amino acids, and
20%–25% from ILE. Per our current protocol, laboratory
Nutrition in Clinical Practice 33(6)
tests are obtained on a routine basis (weekly initially, fol-
lowed by biweekly, then monthly, and depending on stabil-
ity, potentially quarterly). If indications of ILE intolerance
such as an elevation of liver enzymes or bilirubin level
1.5 times baseline is noted, the frequency of ILE infusion
is reduced. Typically, reduction proceeds from 3 times per
week ILE to once per week with the reassessment of labs.
If there is no improvement or stabilization in liver enzymes
or bilirubin levels, the ILE is reduced further to once every
other week, and then once monthly. The reduction in ILE
calories is compensated for with an increase in calories from
dextrose or potentially amino acids if not optimized.
With FDA approval and commercial availability of MO
ILE in the United States, our program reviewed our adult
HPN cohort to identify patients who were intolerant to
SO ILE and require a reduction in the frequency of ILE
administration. These patients were contacted and placed
on MO ILE if they approved and had no contraindications
for use such as hypersensitivity to egg, soybean proteins,
fish, or peanut proteins. Patients were started on the same
frequency and grams of MO ILE as their last SO ILE dose.
As an example, if patient was receiving 50 g of SO ILE once
a week, they would be started on 50 g once a week of MO
ILE. Labs including liver enzymes and bilirubin levels were
assessed frequently (every 2–4 weeks) to ensure tolerance. If
patients were tolerating MO ILE indicated by stability or
improvement in labs as well as a lack of adverse reactions,
then the dose was increased with a corresponding reduction
in dextrose dose every 2 weeks. Dose escalation depended
on specific clinical scenario, however, typically if a patient
was on ILE once every other week, ILE frequency would be
increased to every week while the dose in grams was kept the
same. If they tolerated once a week, they would be increased
to 2–3 times per week followed by further increase to 4–7
times per week.
The results from this case series are presented below.
All data are presented as mean ± standard deviation for
normally distributed data and median (range) for nonpara-
metric data. The statistical analysis was performed using
JMP, Version 10 (SAS Institute Inc., Cary, NC).
Results
A total of 9 adult patients (3 women and 6 men) with an
average age of 48.1 ± 15.1 years were transitioned from
SO ILE to MO ILE as a result of intolerance (Table 1).
They had been on HPN for a median of 9.9 years with
a range of 9.0 months to 30.7 years. The most common
indications for HPN were short bowel syndrome (44%) and
bowel obstruction or dysmotilty (33%) as well as radiation
enteritis (11%) and enterocutaneous fistula (11%). A major-
ity of the patients were transitioned to MO ILE because
of an intolerance of SO ILE that presented in the form of
elevated liver studies. One patient subsequently developed
Mundi et al. 853

Table 1. Baseline Demographics. a catheter-related blood stream infection 30 days prior to

Variable
Age, y
Gender, women/men
Duration of HPN, y
Duration of MO ILE at analysis, d
Indication for HPN, n (%)
- Short bowel syndrome
- Bowel obstruction/motility
disorder
- Radiation enteritis
- Entrocutaneous fistula
and during MO ILE use. Although some patients such as
Results, n = 9
patient 1 (Table 3) did have a significant decline in alkaline
48.1 ± 15.1 phosphatase, the overall group revealed only a trend toward
3/6 improvement with median value changing from 138.0 (52–
9.9 (0.9–30.7) 884) to 106 (47–512; P = .09). Similarly, AST, ALT, and total
140.7 ± 29.7 (range of bilirubin values also showed trends toward improvement,
75–180 days) again with some individuals showing clinically significant
improvements (Table 3). As expected, vitamin E levels im-
4 (44)
proved significantly with MO ILE administration increasing
3 (33)
1 (11) from 6.6 ± 1.9 to 10.3 ± 4.3 (P = .02). In addition, there was
1 (11) no significant change in triglyceride (117.9 ± 45.3 to 107.3
± 46.5, P = .8) or glucose levels (93.0 ± 14.6 to 98.6 ± 18.9,
P = .37).

Data are presented as mean ± standard deviation if normal

distribution or as median (range) for nonparametric data. HPN, home
parenteral nutrition; MO ILE, mixed-oil intravenous lipid emulsion.
intolerance of the high dextrose formula with hyperglycemia
requiring insulin in HPN.
At the time of analysis, the 9 patients were on MO ILE
for an average of 140.7 ± 29.7 days. Their initial body mass
index was 23.6 ± 4.6 kg/m2 , and they were receiving an
average 25.7 ± 8.0 kcal per kg of body weight per day in
their HPN formula (Table 2). All 9 patients tolerated MO
ILE well, and we were able to increase the percentage of
calories provided through ILE from 7.6 ± 6.5% to 18.4
± 8.2% (P = .003) while decreasing the dextrose from
66.9 ± 8.4% to 56.9 ± 5.5% (P = .0007; Figure 1). The
increase in lipids corresponded to an increase from 14.7
± 13.5 g/day (0.2 ± 0.2 g/kg/day) to 34.0 ± 21.3 g/day
(0.6 ± 0.4 g/kg/day; P = .01). Only 1 patient developed
Discussion
The present case series outlines our HPN program’s experi-
ence with our first 10 patients transitioned from 100% SO
ILE to MO ILE. These patients were transitioned to MO
ILE because of an intolerance of SO ILE that presented
predominantly as liver function test abnormalities. One
patient with neuroendocrine tumor in the abdomen com-
plained of pain with SO ILE and thus was transitioned to
MO ILE to assess if there would be an improvement in
symptoms given the lack of alternative options for nutrition.
Unfortunately, despite 2 weeks of MO ILE, she did not
experience an improvement and thus was removed from
further analysis. The remaining patients tolerated MO ILE
well without adverse reactions and with trends toward
improvement in liver studies. In addition, the percent of
calories delivered as fats were significantly increased and
dextrose decreased when compared with baseline.

Table 2. Pre–MO ILE and Post–MO ILE Values From 9 Patients.

Variable Pre–MO ILE Latest Value on MO ILE P Value

BMI 23.6 ± 4.6 22.9 ± 4.1 .39

Total calories per day, kcal 1751.2 ± 491.4 1762.9 ± 444.2 .92
Total calories per kg per day, kcal 25.7 ± 8.0 27.0 ± 11.1 .43
% Calories from amino acid, % 25.5 ± 5.5 24.8 ± 7.6 .63
% Calories from dextrose, % 66.9 ± 8.4 56.9 ± 5.5 .0007
% Calories from fats, % 7.6 ± 6.5 18.4 ± 8.2 .003
ILE g/day, g/kg/day 14.7 ± 13.5 (0.2 ± 0.2) 34.0 ± 21.3 (0.6 ± 0.4) .01
Frequency of ILE, days per week 1.4 ± 1.2 3.4 ± 1.8 .01
Alkaline phosphatase, U/L 138.0 (52–884) 106 (47–512) .09
AST, U/L 64.3 ± 33.9 46.2 ± 19.9 .18
ALT, U/L 108.9 ± 105.0 63.8 ± 35.9 .31
Total bilirubin, mg/dL 1.4 ± 0.7 1.2 ± 0.9 .07
Vitamin E, mg/L, n = 8 6.6 ± 1.9 10.3 ± 4.3 .02
Triglyceride, mg/dL, n = 8 117.9 ± 45.3 107.3 ± 46.5 .8
Glucose, mg/dL 93.0 ± 14.6 98.6 ± 18.9 .37

Data are presented as mean ± standard deviation if normal distribution or as median (range) for nonparametric data. ALT, alanine
aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; ILE intravenous lipid emulsion; MO ILE, mixed-oil intravenous lipid
emulsion. p-value that reached significance indicated by bolding.
854 Nutrition in Clinical Practice 33(6)

Figure 1. Change in macronutrient composition after initiation of mixed-oil intravenous lipid emulsion (MO ILE).

Table 3. Liver Studies for Each Patient Before and After MO ILE.

Alkaline Phosphatase, U/L AST, U/L ALT, U/L Total Bilirubin, mg/dL

Pre– Post– Pre– Post– Pre– Post– Pre– Post–

Patient MO ILE MO ILE MO ILE MO ILE MO ILE MO ILE MO ILE MO ILE

1 884 512 116 62 91 47 1.1 0.7

2 125 103 41 45 61 66 1.6 1.2
3 125 127 27 29 52 63 0.9 1.0
4 139 131 22 25 46 42 2.3 2.3
5 138 63 59 27 66 29 0.7 0.4
6 52 47 79 46 82 59 1.7 1.9
7 200 195 55 92 70 155 1.1 0.3
8 104 79 56 53 111 80 2.8 2.8
9 384 131 124 44 401 55 0.8 0.4

ALT, alanine aminotransferase; AST, aspartate aminotransferase; MO ILE, mixed-oil intravenous lipid emulsion.

IFALD, which can present as hepatic steatosis, cholesta-
sis, cholelithiasis, and potentially hepatic fibrosis, is a
common problem affecting patients on HPN.7-11 In fact,
Cavicchi et al10 followed 90 patients supported with HPN
in the setting of intestinal failure and noted that 65%
had developed chronic cholestasis after a median of 6
months. A total of 41.5% developed some form of IFALD
after median of 17 months, and 18.9% developed fibrosis
with 5.6% progressing to cirrhosis with long-term HPN
use. Although the precise etiology is not fully understood,
nutrient excess and toxicity along with infection are felt to
be risk factors.3,9,20 Treatment options are limited, with case
series and small trials showing some benefit with use of
antibiotics and ursodeoxycholic acid.21-24 In addition, cyclic
HPN (12-hour infusion) along with decreased frequency of
ILE (3 days per week) have also shown some promise to help
improve IFALD.25,26
Unfortunately, despite these treatments and preventive
options, patients can still progress to IFALD requiring
further reduction in ILE frequency (<3 days per week). This
Mundi et al.
often leads to a significant mismatch in composition of
HPN macronutrients, with the majority of calories being
provided as dextrose to compensate for the reduction in
ILE. This was evident in our case series as >65% of calories
on average were being provided from dextrose with only
7.6% being provided from ILE prior to initiating MO ILE.
This strategy may not be sufficient in many cases as the
high dextrose formula can lead to complications of its own,
including hyperglycemia requiring insulin to be added to
PN, as was seen in 1 of our patients. In addition to hyper-
glycemia, high-dextrose PN formulas that were commonly
used prior to SO ILE being available in the United States
were also associated with hepatic enzyme elevations, steato-
sis, and essential fatty acid deficiency (EFAD).3 In addition,
our group has also noted that these metabolic complications
from high-dextrose formulas limit the number of overall
calories that can be provided, resulting in the underfeeding
of often malnourished patients.
These issues have led to the exploration of the use of
alternative ILE with small case series or short-term (5–
29 days) trials showing promising results.3 Piper et al15
randomized 44 postoperative patients who required PN to
either MO ILE (n = 22) or olive oil/SO containing ILE
(20% ClinOleic) for 5 days. They noted that at days 2 and 5,
the MO ILE group had lower AST and ALT levels despite
there being no difference at day 0. Goulet et al17 randomized
28 children on HPN to either MO ILE or SO ILE (20%
intralipid) and followed them for 29 days. Although they
found no difference in liver enzymes between the groups,
they did note that the MO ILE group had a reduction in
total bilirubin values, whereas the SO ILE group had an
increase. Similarly, Klek et al18 randomized 73 adult patients
on HPN as a result of intestinal failure to either MO ILE
or SO ILE (20% intralipid) and followed them for 4 weeks.
They noted that mean concentrations of ALT, AST, and
total bilirubin were significantly lower in the MO ILE group
when compared with the control group.
Our case series also found similar results with all liver
studies (alkaline phosphatase, AST, ALT, and total biliru-
bin) showing overall improvement although statistical sig-
nificance was not reached because of the small number of
patients. However, as opposed to the previous trials, which
tended to enroll patients who were being initiated on HPN
and excluding those with liver disease, our cohort consisted
of patients who had previously been on HPN for a median
duration of 9.9 years and had developed an intolerance
of SO ILE that presented as abnormal liver studies in the
majority of cases. Based on our protocol, ILE frequency
was reduced to improve or stabilize liver studies, resulting in
a significant increase in calories from dextrose. At the time
that MO ILE was approved for use in adults, 5 patients had
been on HPN for close to 10 years or greater and during
this time multiple attempts had been made to increase SO
ILE after the initial reduction led to an improvement in
855
liver studies. In some cases, this increase would work for a
few months, but given sufficient time liver studies became
abnormal once again, leading to a reduction in SO ILE use.
With the transition to MO ILE, the calories from ILE were
increased from 7.6% to 18.4%, with the frequency of ILE
increasing to 3.4 days per week from the initial 1.4 days per
week. This allowed the reduction in dextrose calories from
66.9% to 56.9%.
Our cohort also experienced an increase in vitamin E
levels from 6.6 to 10.4, similar to other trials.16-18 This is
expected given that most FO ILE contain approximately
200 mg/L of α-tocopherol.3 α-tocopherol is an antioxidant
from the vitamin E family that can scavenge free radicals
that bind to deoxyribonucleic acid and proteins resulting in
cell damage and death. In addition to raising α-tocopherol
levels, FO ILE also has higher ratio of n-3 poly-unsaturated
fatty acids (PUFAs) when compared with n-6 PUFAs. The
n-6 PUFAs can be further metabolized to give rise to
proinflammatory eicosanoids, with ILEs that have a high
ratio of n-6 to n-3 PUFA being felt to be proinflammatory
and thus contributing to some of the long-term complica-
tions associated with HPN.27-30 Although this was beyond
the capability of this current study, other randomized trials
have shown that FO ILE use can lead to a more beneficial
n-3 to n-6 PUFA ratio as well as decreased inflammatory
markers such as interleukin-6 and tumor necrosis factor-
α.14,17,18
The current study did have a number of limitations,
starting with the fact that it was a prospective case series
and not a randomized control trial. As a result, all patients
were not blinded to the treatment, which could have resulted
in some bias occurring in terms of tolerance. In fact, some
patients self-requested initiation of MO ILE as soon as it
was available in the United States. These patients may have
been less likely to report any adverse tolerance issues. The
study was also a case series and thus patients served as their
own control with comparisons being performed between
values gathered prior to initiation of MO ILE and at the
time of analysis. Certainly other clinical factors such as
infection or improvement in clinical status could contribute
to the findings. In addition, most of the patients in the case
series had suffered from elevations in liver studies for many
years with recurrent reduction in SO ILE. As a result, labs
that led to the initial reduction in SO ILE were not attainable
and perhaps could have revealed a significant improvement
with transition to MO ILE.
Another limitation pertains to the development of
EFAD. Because SO has the highest concentration of es-
sential fatty acids (linoleic acid and α-linolenic acid), a
reduction in SO content in MO ILE when compared with
SO ILE does raise increased concern for the develop-
ment of EFAD.3 In our present case series, no patients
reported symptoms consistent with EFAD and thus fatty
acid profile was not obtained. Literature has estimated that
856
0.05–0.14 g/kg/day of SO ILE should be sufficient to prevent
EFAD.31 In our case series, on average, patients received
0.6 ± 0.4 g/kg/day of MO ILE. As a result, we felt less
concern regarding the development of EFAD. In addition,
most of the patients in the case series were consuming
and possibly absorbing some nutrition from an oral intake
of variable amounts. Although it was very difficult for us
to obtain and estimate how much nutrition was absorbed
through oral intake given the anatomic variability such as
short bowel syndrome, some essential fatty acids could have
been obtained through oral intake. As a result, a long-term
randomized trial with the assessment of fatty acid profile
will be needed to assess whether MO ILE use is associated
with a higher risk of EFAD when compared with SO ILE.
Until these results are available, caution should be taken
in those who are at high risk for EFAD such as severely
malnourished adults or children. In fact, MO ILE does
not currently have FDA indication for use in the pediatric
population.
Conclusions
Overall, our case series did reveal that initiation of MO
ILE in patients with intolerance to SO ILE predominantly
because of elevated liver studies allowed for the increase
in percentage of calories obtained from ILE along with
a significant reduction in the percentage of calories from
dextrose, bringing the overall macronutrient composition
closer to the desired amounts. This change was associated
with no worsening of liver enzyme and function tests, which
instead showed a slight improvement. Long-term prospec-
tive randomized control trials are necessary to elucidate
the full impact of MO ILE in HPN patients in terms of
inflammation, fat metabolism, and liver function.
Statement of Authorship
Manpreet S. Mundi and Ryan T. Hurt contributed to the
research; Megan T. McMahon, Jennifer Carnell, and Manpreet
S. Mundi contributed to acquisition, analysis, or interpretation
of data. Manpreet S. Mundi drafted the manuscript; Megan T.
McMahon, Jennifer Carnell, and Ryan T. Hurt critically revised
the manuscript; and all authors agree to be fully accountable for
ensuring the integrity and accuracy of the work. All authors
read and approved the final manuscript.
References
1. Howard L, Ament M, Fleming CR, Shike M, Steiger E. Current use and
clinical outcome of home parenteral and enteral nutrition therapies in
the United States. Gastroenterology. 1995;109(2):355-365.
2. Mundi MS, Pattinson A, McMahon MT, Davidson J, Hurt RT.
Prevalence of home parenteral and enteral nutrition in the United
States. Nutr Clin Pract. 2017;32(6):799-805. https://doi.org/10.1177/
0884533617718472
3. Mundi MS, Salonen BR, Bonnes S. Home parenteral nutrition:
Fat emulsions and potential complications. Nutr Clin Pract.
2016;31(5):629-641. https://doi.org/10.1177/0884533616663635
Nutrition in Clinical Practice 33(6)
4. Varayil JE, Whitaker JA, Okano A, et al. Catheter salvage af-
ter catheter-related bloodstream infection during home parenteral
nutrition. J Parenter Enter Nutr. May 2015;41(3):481-488. https://
doi.org/10.1177/0148607115587018
5. Meguid MM, Kurzer M, Hayashi RJ, Akahoshi MP. Short-term effects
of fat emulsion on serum lipids in postoperative patients. JPEN J
Parenter Enteral Nutr. 1989;13(1):77-80.
6. Tashiro T, Mashima Y, Yamamori H, Sanada M, Nishizawa M, Okui
K. Intravenous intralipid 10% vs. 20%, hyperlipidemia, and increase in
lipoprotein X in humans. Nutrition. 1992;8(3):155-160.
7. Xu Z-W, Li Y-S. Pathogenesis and treatment of parenteral nutrition-
associated liver disease. Hepatobiliary Pancreat Dis Int. 2012;11(6):586-
593.
8. Burns DL, Gill BM. Reversal of parenteral nutrition–associated liver
disease with a fish oil–based lipid emulsion (omegaven) in an adult
dependent on home parenteral nutrition. J Parenter Enter Nutr.
2013;37(2):274-280. https://doi.org/10.1177/0148607112450301
9. Kelly DA. Intestinal failure–associated liver disease: what do
we know today? Gastroenterology. 2006;130(2, suppl):S70-S77.
https://doi.org/10.1053/j.gastro.2005.10.066
10. Cavicchi M, Beau P, Crenn P, Degott C, Messing B. Prevalence
of Liver Disease and Contributing Factors in Patients Receiving
Home Parenteral Nutrition for Permanent Intestinal Failure. Ann
Intern Med. 2000;132(7):525-532. https://doi.org/10.7326/0003-4819-
132-7-200004040-00003.
11. Chan S, McCowen KC, Bistrian BR, et al. Incidence, prognosis,
and etiology of end-stage liver disease in patients receiving home
total parenteral nutrition. Surgery. 1999;126(1):28-34. https://doi.org/
10.1067/msy.1999.98925
12. Bishay M, Pichler J, Horn V, et al. Intestinal failure-associated
liver disease in surgical infants requiring long-term parenteral
nutrition. J Pediatr Surg. 2012;47(2):359-362. https://doi.org/
10.1016/j.jpedsurg.2011.11.032
13. Tillman EM. Review and clinical update on parenteral nutrition–
associated liver disease. Nutr Clin Pract. 2013;28(1):30-39.
https://doi.org/10.1177/0884533612462900
14. Metry AA, Abdelaal W, Ragaei M, Refaat M, Nakhla G. SMOFlipid
versus intralipid in postoperative ICU patients. Enliven J Anesth Crit
Care Med. 2014;1(6):15.
15. Piper SN, Schade I, Beschmann RB, Maleck WH, Boldt J, Röhm KD.
Hepatocellular integrity after parenteral nutrition: comparison of a
fish-oil-containing lipid emulsion with an olive-soybean oil-based lipid
emulsion. Eur J Anaesthesiol. 2009;26(12):1076-1082.
16. Tomsits E, Pataki M, Tölgyesi A, Fekete G, Rischak K, Szollár
L. Safety and efficacy of a lipid emulsion containing a mixture of
soybean oil, medium-chain triglycerides, olive oil, and fish oil: a
randomised, double-blind clinical trial in premature infants requiring
parenteral nutrition. J Pediatr Gastroenterol Nutr. 2010;51(4):514-521.
https://doi.org/10.1097/MPG.0b013e3181de210c
17. Goulet O, Antébi H, Wolf C, et al. A new intravenous fat emulsion con-
taining soybean oil, medium-chain triglycerides, olive oil, and fish oil a
single-center, double-blind randomized study on efficacy and safety in
pediatric patients receiving home parenteral nutrition. J Parenter Enter
Nutr. 2010;34(5):485-495. https://doi.org/10.1177/0148607110363614
18. Klek S, Chambrier C, Singer P, et al. Four-week parenteral nutrition
using a third generation lipid emulsion (SMOFlipid)—a double-blind,
randomised, multicentre study in adults. Clin Nutr. 2013;32(2):224-231.
https://doi.org/10.1016/j.clnu.2012.06.011
19. Staun M, Pironi L, Bozzetti F, et al. ESPEN Guidelines on parenteral
nutrition: home parenteral nutrition (HPN) in adult patients. Clin Nutr.
2009;28(4):467-479. https://doi.org/10.1016/j.clnu.2009.04.001
20. Carter BA, Shulman RJ. Mechanisms of disease: update on the
molecular etiology and fundamentals of parenteral nutrition associated
cholestasis. Nat Clin Pract Gastroenterol Hepatol. 2007;4(5):277-287.
https://doi.org/10.1038/ncpgasthep0796
Mundi et al.
21. Luis VAS, Btaiche IF. Ursodiol in patients with parenteral nutrition–
associated cholestasis. Ann Pharmacother. 2007;41(11):1867-1872.
https://doi.org/10.1345/aph.1K229
22. Lindor KD, Burnes J. Ursodeoxycholic acid for the treatment of home
parenteral nutrition-associated cholestasis. A case report. Gastroen-
terology. 1991;101(1):250-253.
23. Beau P, Labat-Labourdette J, Ingrand P, Beauchant M. Is ursodeoxy-
cholic acid an effective therapy for total parenteral nutrition-related
liver disease? J Hepatol. 1994;20(2):240-244.
24. Günşar C, Vatansever S, Var A, et al. Antibiotic treatment is
superior to ursodeoxycholic acid on total parenteral nutrition as-
sociated hepatic dysfunction. Pediatr Surg Int. 2010;26(5):479-486.
https://doi.org/10.1007/s00383-010-2578-5
25. Maini B, Blackburn GL, Bistrian BR, et al. Cyclic hyperalimentation:
An optimal technique for preservation of visceral protein. J Surg Res.
1976;20(6):515-525. https://doi.org/10.1016/0022-4804(76)90085-8
26. Hwang TL, Lue MC, Chen LL. Early use of cyclic TPN prevents fur-
857
ther deterioration of liver functions for the TPN patients with impaired
liver function. Hepatogastroenterology. 2000;47(35):1347-1350.
27. Wanten GJ, Calder PC. Immune modulation by parenteral lipid
emulsions. Am J Clin Nutr. 2007;85(5):1171-1184.
28. Tilley SL, Coffman TM, Koller BH. Mixed messages: modulation of
inflammation and immune responses by prostaglandins and thrombox-
anes. J Clin Invest. 2001;108(1):15-23.
29. Lewis RA. Interactions of eicosanoids and cytokines in immune
regulation. Adv Prostaglandin Thromboxane Leukot Res. 1990;20:170-
178.
30. Vanek VW, Seidner DL, Allen P, et al. A.S.P.E.N. position paper
clinical role for alternative intravenous fat emulsions. Nutr Clin Pract.
2012;27(2):150-192. https://doi.org/10.1177/0884533612439896
31. Gramlich L, Meddings L, Alberda C, et al. Essential fatty acid
deficiency in 2015: the impact of novel intravenous lipid emul-
sions. J Parenter Enter Nutr. 2015;39(1 suppl):61S-66S. https://doi.
org/10.1177/0148607115595977
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Nutrition Status Among HIV-Positive and HIV-Negative December 2018 858–864

C 2018 American Society for

Inpatients with Pulmonary Tuberculosis Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10006
wileyonlinelibrary.com

Tássia Kirchmann Lazzari, MSc1 ; Gabriele Carra Forte, PhD2 ;

and Denise Rossato Silva, MD, PhD3

Abstract
Background: The association between tuberculosis (TB) and malnutrition is well recognized. Considering the risk of mortality
due to malnutrition in patients with TB, it is necessary to conduct a thorough nutrition assessment to identify individuals at
nutrition risk. The study objective was to assess the nutrition status of hospitalized patients with TB, co-infected or not by human
immunodeficiency virus (HIV). Methods: Patients with confirmed diagnosis of TB were included using a cross-sectional design.
Nutrition assessment parameters included: body mass index (BMI), triceps skin-fold thickness (TSF), bioelectrical impedance
analysis (BIA), mid-upper-arm circumference (MUAC), mid-arm muscle circumference (MAMC), food frequency questionnaire,
Malnutrition Screening Tool (MST), Subjective Global Assessment (SGA), and serum levels of hemoglobin. Results: A total 108
patients completed the study. Forty-four patients (40.7%) were HIV positive. Considering the BMI, 36.1% of the patients met the
criteria for nutrition deficiency. Body fat percentage was low in 27.8% of patients. In addition, more than half of the participants met
criteria for malnutrition according to MUAC, MAMC, TSF, SGA, or MST. Malnutrition measured by MAMC was more frequent
in HIV-positive patients (n=33, 75.0%) than in HIV-negative patients (n=31, 48.4%) (P = 0.010). Regarding the components of diet,
selenium and vitamin C intake among HIV-positive patients was significantly lower than in HIV-negative patients. Conclusions: We
identified a high prevalence of malnutrition in hospitalized patients with pulmonary TB, regardless of the method used to assess
nutrition status. In HIV-positive patients, malnutrition measured by MAMC was more frequent than in HIV-negative patients.
(Nutr Clin Pract. 2018;33:858–864)

Keywords
ascorbic acid; body mass index; electrical impedance; nutritional status; selenium; skin fold thickness; tuberculosis

Tuberculosis (TB) is a major public health problem world-
wide, particularly in low and middle income populations.
It is estimated that one-third of the world population is
infected with Mycobacterium tuberculosis. Brazil is in 16th
place among the 22 countries that collectively account for
80% of TB cases globally, with a reported incidence of 30.9
cases/100,000 inhabitants/year in 2015.1,2 Porto Alegre is
the second Brazilian capital with the highest number of TB
cases, with an incidence of 88.8 cases/100,000 inhabitants
and a high percentage of TB-human immunodeficiency
virus (HIV) (25.2%).1
The association between TB and malnutrition is well
recognized; TB can lead to malnutrition and malnutrition
may predispose to TB. The protein-energy malnutrition
observed in patients with TB with or without infection
by HIV is characterized by reduction of visceral protein,
lean mass and fat, and a significant decrease of some
anthropometric and biochemical changes (Hb).3 Moreover,
malnutrition is associated with fatigue of the respiratory
muscles, leading to worsening disease prognosis.4,5
The systemic response to infection, comprising neuro-
humoral and immunologic changes, is characterized by
increased production of proinflammatory cytokines, such
as interleukin-6 (IL-6) and tumor necrosis factor (TNF)-
α, that are anorectic and also responsible for metabolic
changes which in most cases result in increased mortality
From the 1 Programa de Pós-Graduação em Ciências Pneumológicas
da Universidade Federal do Rio Grande do Sul, Rio Grande do Sul,
Brazil; 2 Programa de Pós-Graduação em Ciências Pneumológicas da
Universidade Federal do Rio Grande do Sul, Rio Grande do Sul,
Brazil; 3 Faculdade de Medicina, Universidade Federal do Rio Grande
do Sul, Serviço de Pneumologia, Hospital de Clı́nicas de Porto Alegre,
Programa de Pós-Graduação em Ciências Pneumológicas da
Universidade Federal do Rio Grande do Sul, Rio Grande do Sul,
Brazil.
Financial disclosure: FIPE-HCPA (Fundo de Incentivo à Pesquisa –
Hospital de Clı́nicas de Porto Alegre).
Conflicts of interest: None declared.
Received for publication June 22, 2017; accepted for publication
September 9, 2017.
This article originally appeared online on February 3, 2018.
Corresponding author:
Denise Rossato Silva, MD, PhD, Serviço de Pneumologia, Hospital
de Clı́nicas de Porto Alegre, Rua Ramiro Barcelos, 2350, 2° andar,
Porto Alegre, RS, CEP 90.035-003, Brazil.
Email: denise.rossato@terra.com.br
Kirchmann Lazzari et al
in TB patients.6-9 Moreover, in TB there is reduction of
liver production of albumin, apolipoprotein, transferrin,
and thus, Hb. Simultaneously, the synthesis of acute-phase
proteins increases, including C-reactive protein (CRP).6
The lower the lymphocyte count, the more susceptible to
opportunistic infections the patient becomes and greater
will be the energy expenditure, progressing to severe, and
in some cases intractable, weight loss.10,11 Another study
showed that low Hb levels are related to malnutrition and
poor prognosis.12
The effects of malnutrition on hospitalized patients with
TB are not well known. Many studies are retrospective and
the results are inconsistent.13-17 There are several limitations
in many studies regarding the body composition of patients
with TB, with or without HIV. There is lack of information
on food intake. Another limitation is that the lymphocyte
cell count is not carried out to evaluate the effect of HIV
on disease severity.18,19 Therefore, the aim of this study is to
assess the nutrition status of hospitalized patients with TB
co-infected or not by HIV.
Methods
Study Design and Location
We conducted a cross-sectional study in a general, tertiary
care, university-affiliated hospital with 750 beds. The study
was approved by the Ethics Committee at the hospital on
December 19, 2012 (number 13-0002).
Patients
Adult patients (age ࣙ18 years) with pulmonary TB who
were hospitalized during the study period were identified
and invited to participate. Patients with extrapulmonary TB,
those who were receiving treatment for >1 week, those who
were unable to comply with study procedures, and those
who refused to sign the consent form were excluded from
this study. We also excluded patients with renal or hepatic
insufficiency and those with peripheral edema. Pulmonary
TB was diagnosed according to the Brazilian Guidelines for
Tuberculosis.20
Data Collection
The following data were collected from patient records using
a standardized data extraction tool: demographic data (sex,
age, race, years of schooling), behavioral data (smoking
status, alcohol abuse, injection drug use), and medical
history (clinical form of TB, symptoms at admission,
methods of diagnosis, presence of comorbidities, prior TB
treatment, drug regimen, interval from hospital admission
until diagnosis, length of hospital stay, intensive care unit
[ICU] admission, length of mechanical ventilation, and
hospitalization outcome [death or discharge]). A current
859
smoker was defined as smoking at least 100 cigarettes in
their lifetime and at the time of the survey, smoking at least
1 day a week. A former smoker was defined as smoking
at least 100 cigarettes in their lifetime but at the time of
the survey, did not smoke at all. Never smoked reported
having smoked <100 cigarettes in their lifetime. Alcohol
abuse was defined as daily consumption of at least 30
grams (equivalent to 24 ounces of 4% beer) for men and 24
grams (equivalent to a 175-mL glass of wine) for women.
An independent physician analyzed the chest X-rays and
classified them as typical or compatible with active TB
according to previously described guidelines.21
Nutrition Assessment
Nutrition assessment was performed by body mass index
(BMI), triceps skin-fold thickness (TSF), mid-arm circum-
ference, mid-arm muscle circumference (MAMC), bioelec-
tric impedance analysis (BIA), food frequency question-
naire (FFQ), Malnutrition Screening Tool (MST), Subjec-
tive Global Assessment (SGA), and laboratory tests. The
same researcher made all of the measurements.
BMI. BMI (kg/m2 ) was calculated and the patients were
classified into categories based on the BMI cutoffs for
weight categories as recommended by the WHO.22 Weight
was measured using a standing scale. No patient had fluid
overload that would affect BMI. Malnutrition was defined
as a BMI <18.5 kg/m2 .22
TSF. TSF thickness was measured with a skin caliper on
the posterior upper arm midway between the acromion and
olecranon processes.23 The values obtained were compared
with reference values.24
Mid-upper-arm circumference (MUAC). MUAC was mea-
sured using a non-stretch plastic tape midway between the
acromion and olecranon of the non-dominant arm.25 The
values obtained were compared with reference values.26
MAMC. MAMC was calculated by using MUAC and TSF
measurements (MUAC [cm] – 3.142 x TSF [cm]). The values
obtained were compared with reference values.27
BIA. BIA is a simple, non-invasive technique that has been
recommended for clinical investigation of body composi-
tion analysis. A single-frequency BIA was performed at 50
kHz and 800 mA with standard tetrapolar lead placement
to measure fat and lean tissue mass. All BIA measurements
were performed by 1 trained observer using the same
equipment and recommended standard conditions.28 Fat
mass and fat-free mass (FFM) were estimated from the
values of resistance and reactance obtained through BIA,
using sex-specific equations validated by Kotler et al29 in a
860 Nutrition in Clinical Practice 33(6)

sample of white, black, and Hispanic patients that included Table 1. Characteristics of Hospitalized Patients With
HIV-infected individuals. Tuberculosis.

Characteristics n=108
FFQ. To assess daily food intake we used a FFQ that
is reproducible and validated.30 The study’s nutritionists Male sex, n (%) 77 (71.3)
reviewed the diet records and resolved any questions with Age (years), mean ± SD 44.9 ± 15.5
the participant. We calculated the mean daily intake of White race, n (%) 60 (55.6)
macronutrients and micronutrients using NutWin - Nu- Current smoking, n (%) 45 (41.7)
Alcoholism, n (%) 42 (38.9)
trition Support Program software, version 1.5 (Health’s
Drug use, n (%) 37 (34.3)
Informatics Department, Escola Paulista de Medicina, Uni- HIV, n (%) 44 (40.7)
versidade Federal de São Paulo, São Paulo, Brazil). The Smear positive, n (%) 74 (68.5)
FFQ was used to assess usual intake before hospitalization Chest X-ray typicala of TB, n (%) 67 (62.0)
(food intake in the past month) because our idea was to Chest X-ray compatibleb with TB, n (%) 41 (38.0)
see the effects of the disease in nutrition status and not the In-hospital mortality, n (%) 9 (8.3)
effects of hospitalization in nutrition status. Hb (g/dL), mean ± SD 10.4 ± 2.1
BMI (kg/m2 ), mean ± SD 20.7 ± 4.5
Body fat (%, BIA), mean ± SD 24.1 ± 10.6
SGA. SGA is a valid and reliable tool which assesses Body fat (%,TSF), mean ± SD 22.8 ± 9.2
nutrition status based on the features of a medical history SGA A, n (%) 5 (4.6)
and physical examination. Subjects were rated as being well SGA B, n (%) 63 (58.3)
nourished (SGA A); moderately, or suspected of being, SGA C, n (%) 40 (37.0)
malnourished (SGA B); or severely malnourished (SGA C). MST 1, n (%) 7 (6.5)
MST 2, n (%) 101 (93.5)
It is widely used both in hospital care and in studies to assess
Malnutrition by MUAC, n (%) 69 (63.9)
nutrition status.31 Malnutrition by MAMC, n (%) 64 (59.3)
Malnutrition by TSF, n (%) 64 (59.3)
MST. The nutrition risk assessment was carried out using
a Nodular, alveolar, or interstitial infiltrates predominantly affecting
the MST. It incorporates 3 components: the presence of
the zones above the clavicles or upper zones; cavitation affecting the
weight loss (score 0 or 2), the amount of weight lost (score upper zones or the apical segment of the lower lobe. b Enlarged hilar
1–4), and poor food intake or poor appetite (score 0 or 1). nodes, pneumonic lesion, atelectasis, mass lesion, miliary. BMI, body
The total score was calculated for each patient. A score ࣘ2 mass index. BIA, bioelectrical impedance analysis. Hb, hemoglobina.
means that the patient is at risk for malnutrition. It has been HIV, human immunodeficiency virus. MAMC, mid-arm muscle
circumference. MST, malnutrition screeening tool (1: not at risk of
used in patients with TB as a tool capable of predicting malnutrition; 2: at risk of malnutrition). MUAC, mid upperarm
prognosis due to the nutrition status of these patients. It has circumference. SD, standard deviation. SGA, subjective global
the advantage of being easy, fast, and simple to use and does assessment (A: well nourished; B: moderately, or suspected of being,
not require calculations. Its sensitivity and specificity is 93% malnourished; C: severely malnourished). TB, Tuberculosis. TSF,
triceps skin-fold thickness.
compared with the SGA, but both can and should be used
together for optimal nutrition assessment.32-34
confidence level, we estimated a sample size of at least 21
Laboratory testing. Blood samples were collected for Hb patients per group.
dosage. Anemia was determined when the Hb levels were
<13 g/dL (men) and <12 g/dL (women).35 Results
During the study period, 108 patients met the inclusion
Statistical Analysis criteria and were included in the analysis. The characteristics
Data analysis was performed using SPSS 18.0 (Statistical of the study population are shown in Table 1. The mean
Package for the Social Sciences, Chicago, Illinois). Data age of all patients was 44.9 ± 15.5 years, 71.3% were males,
were presented as number of cases, mean ± standard devi- and 55.6% were white. Forty-four patients (40.7%) were HIV
ation (SD), or median with interquartile range (IQR). Cat- positive. BMI was ࣘ18.5 kg/m2 in 39 (36.1%) patients. The
egoric comparisons were performed by χ 2 test using Yates’ majority of patients were considered malnourished or at
correction if indicated or by Fisher’s exact test. Continuous risk of malnutrition by SGA, MST, MUAC, MAMC, and
variables were compared using the t-test or Wilcoxon test. TSF evaluations.
A 2-sided P value <0.05 was considered significant for Table 2 shows the comparisons between HIV-positive
all analyses. Sample size calculation for the comparison and HIV-negative patients. Malnutrition measured by
between HIV-positive and HIV-negative groups was based MAMC was more frequent in HIV-positive patients (n =
on a previous study.36 Considering 90% power and a 95% 33, 75.0%) than in HIV-negative patients (n = 31, 48.4%)
Kirchmann Lazzari et al 861

Table 2. Comparison Between HIV Positive and HIV Negative Patients With Tuberculosis.

Characteristics HIV positive n=44 HIV negative n=64 p value

Male sex, n (%) 27 (61.4) 50 (78.1) 0.094

Age (years), mean ± SD 39.2 ± 10.3 48.9 ± 17.2 <0.0001
White race, n (%) 15 (34.1) 45 (70.3) <0.0001
Current smoking, n (%) 22 (50.0) 23 (35.9) 0.208
Alcoholism, n (%) 19 (43.2) 23 (35.9) 0.577
Drug use, n (%) 23 (52.3) 14 (21.9) 0.002
Smear positive, n (%) 29 (65.9) 45 (70.3) 0.785
Chest X-ray typicala of TB, n (%) 24 (54.5) 43 (67.2) 0.259
Chest X-ray compatibleb with TB, n (%) 21 (47.7) 20 (31.3) 0.126
In-hospital mortality, n (%) 3 (6.8) 6 (9.4) 0.906
Hb (g/dL), mean ± SD 9.6 ± 1.7 10.9 ± 2.2 0.001
Selenium (μg), mean ± SD 0.011± 0.053 0.027 ± 0.069 0.02
Vitamin C (mg), median (P25-P75) 17.9 (6.3-59.0) 38.5 (24.4-107.7) 0.005
BMI (kg/m2 ), mean ± SD 19.87 ± 3.8 21.3 ± 4.9 0.079
Body fat (%, BIA), mean ± SD 24.2 ± 11.4 24.1 ± 10.1 0.951
Body fat (%,TSF), mean ± SD 21.2 ± 8.2 23.8 ± 9.7 0.143
SGA A, n (%) 2 (4.5) 3 (4.7) 0.999
SGA B, n (%) 23 (52.3) 40 (62.5) 0.389
SGA C, n (%) 19 (43.2) 21 (32.8) 0.371
MST 1, n (%) 4 (9.1) 3 (4.7) 0.440
MST 2, n (%) 40 (90.9) 61 (95.3) 0.440
Malnutrition by MUAC, n (%) 32 (72.7) 37 (57.8) 0.167
Malnutrition by MAMC, n (%) 33 (75.0) 31 (48.4) 0.010
Malnutrition by TSF, n (%) 27 (61.4) 37 (57.8) 0.865
a Nodular, alveolar, or interstitial infiltrates predominantly affecting the zones above the clavicles or upper zones; cavitation affecting the upper

zones or the apical segment of the lower lobe. b Enlarged hilar nodes, pneumonic lesion, atelectasis, mass lesion, miliary. BIA, bioelectrical
impedance analysis. BMI, body mass index. CRP, C-reactive protein. Hb, hemoglobina. HIV, human immunodeficiency virus. MAMC, mid-arm
muscle circumference. MST, malnutrition screeening tool (1: not at risk of malnutrition; 2: at risk of malnutrition). MUAC, mid upper arm
circumference. P25-P75: percentile 25-percentile 75. SD, standard deviation. SGA, subjective global assessment (A: well nourished; B: moderately,
or suspected of being, malnourished; C: severely malnourished). TB, Tuberculosis. TSF, triceps skin-fold thickness.

(P = 0.010). Regarding diet components, only selenium
and vitamin C intake were lower in HIV-positive patients
compared with HIV-negative patients (P = 0.02 and P =
0.005, respectively). The other components of the diet were
not statistically different between groups (data not shown).
Discussion
In this study, we found a high prevalence of malnutrition in
hospitalized patients with newly diagnosed pulmonary TB,
regardless of the parameter used to assess nutrition status.
Considering the BMI, more often used in studies, 36.1% of
the patients met the criteria for malnutrition. In addition,
27.8% of patients had a low body fat percentage. More than
half of the individuals studied had criteria for malnutrition
according to MUAC, MAMC, TSF, SGA, or MST.
The prevalence of malnutrition varies across studies. Two
other studies also conducted in Brazil, 1 in São Paulo37
and another in Rio de Janeiro,8 showed results similar to
those found in the present study, with a prevalence of 34.9%
and 32.0%, respectively, for malnutrition according to BMI.
However, in a study conducted in a rural population of
India, the prevalence of malnutrition by BMI was ˃85%.38
There is much discussion about the use of BMI as the only
method for nutritionl assessment due to its limitations.39.40
Thus, a variety of objective and subjective measures have
been used to diagnose malnutrition or identify patients at
risk of malnutrition. Using some of these measures, such
as BIA, TSF, MUAC, and MAMC, we found a higher
prevalence of malnutrition than by using only BMI. Still,
using the SGA, previously evaluated in patients with TB,41
95.7% of patients in our sample had moderate or severe
malnutrition. We also used the MST to assess nutrition
status, a tool that already was shown to be useful in the
evaluation of TB patients,33 and it showed that 93.5%
of patients in the present study were malnourished or at
risk of malnutrition. Therefore, many patients considered
having normal weight, overweight, or obesity by BMI were
classified as malnourished or at risk of malnutrition by
other nutrition assessment methods. We then consider that
BMI has limitations for nutrition assessment also in patients
with TB and that other measures should be used together.
Among the other measures used in our study, it is difficult
to say which ones are the best or if all of them are needed in
862
future studies. We don’t think it will be possible to establish
a gold standard for the evaluation of malnutrition in TB
patients. Unfortunately, the present study was not designed
to calculate sensitivity and specificity of each tool. In spite
of that, a single parameter probably will not be able to reflect
nutrition deficits among inpatients with TB.
Malnutrition and TB are both problems of developing
countries. TB can lead to malnutrition and malnutrition
may predispose to TB. Poor nutrition leads to protein-
energy malnutrition and micronutrient deficiencies that lead
to immunodeficiency. This secondary immunodeficiency
increases the susceptibility of the host to infections and
thus increases the risk of developing TB. TB alone leads
to reduced appetite, malabsorption of nutrients and mi-
cronutrients, and altered metabolism causing poor nutrition
status.42
Nutrition status appears to be an important determinant
of clinical outcomes during TB treatment. Malnutrition
can adversely affect treatment outcomes, being linked to
excess deaths43-45 and increased risk of relapse.44,46 Severe
malnutrition at diagnosis was associated with a 2-fold higher
risk of death.38 In a study in Malawi, a high degree of
malnutrition was a risk factor for early mortality among
TB patients.45 One study suggests that malnutrition inter-
vention could have a substantial impact on TB mortality
in areas with high prevalence of malnutrition, even if only
small gains in malnutrition can be achieved.47 In addition,
restoration of nutrition could also lead to immunologic
changes that could enhance the clearance of mycobacteria
and reduce infectiousness of patients.43 One study showed
that nutrition support can improve both sputum smears or
culture-negative conversion rate and BMI, shortening the
time of sputum conversion.48 In another study, patients re-
ceiving better nutrition tended to show more rapid clearance
of bacteria and radiographic changes in addition to greater
weight gain.49 We collected data on mortality in our study;
however, we had only 9 deaths and could not evaluate risk
factors for mortality, including malnutrition.
High prevalence of HIV infection in underdeveloped
countries further aggravates the problems of malnutrition
and TB. In a study that compared nutrition status of
individuals with HIV alone, TB-HIV, and HIV-negative,
the authors found that malnutrition was most pronounced
among patients with TB-HIV coinfection.50 Net protein
anabolism is impaired in patients in HIV-positive adults
with TB, and this impairment is significantly greater than
that seen in patients with HIV or TB alone.43 We found
differences in malnutrition between HIV-positive and HIV-
negative patients only by MAMC. A study conducted in
Guinea-Bissau also showed a low MAMC in HIV-infected
TB patients, and this variable was associated with greater
mortality.51
The interaction between nutrition deficiencies and infec-
tion is well known. In studies evaluating the relationship
Nutrition in Clinical Practice 33(6)
between diet and TB, nutrients examined vary, but include
proteins; vitamins A, C, D, and E; zinc; selenium; iron;
copper; cholesterol; and polyunsaturated fatty acids.43,52
Micronutrient deficiency is considered the most frequent
cause of morbidity and secondary immunodeficiency asso-
ciated with infection, including TB.43 In our study, HIV-
positive patients had a lower intake of selenium and vita-
min C than HIV-negative patients, and both were below
the RDA. The essential trace element selenium plays an
important role in maintaining the immune process and
thus may play a critical role in the removal of mycobacte-
ria. Deficiencies in micronutrients, including selenium, are
widespread in HIV-infected individuals, most notably under
resource-limited conditions.53 Selenium has been described
as a significant risk factor in the development of disease
by mycobacteria in HIV-positive patients.54 Vitamin C also
has a key function in pulmonary antioxidant defense,55 and
studies have reported low concentrations of this vitamin in
TB patients.55,56 Orally administered vitamin C has been de-
scribed to have beneficial effects in TB,57,58 even increasing
the antibacterial activity of rifampin and isoniazid against
M. tuberculosis.59
This study has some methodologic limitations. First, this
was a single-center study and included only hospitalized
patients, probably the most severe cases of TB. Second, the
reduced number of deaths does not allow the evaluation
of nutrition parameters as predictors of mortality. Also,
single-frequency BIA is less accurate than multifrequency
BIA; however, it was used and also validated previously in
hospitalized patients with other diseases.60,61 In addition,
we used FFQ to determine micronutrients; however, FFQ is
a reasonable instrument for assessing the intake of most mi-
cronutrients when compared with nutrition biomarkers. The
advantages of this method are that the burden on examiners
and examinees is small, the cost is low, and self-recording
is feasible. On the other hand, it is limited because of the
limited food categories and intake amount per meal; thus,
the measurement of the intake amount is less accurate than
with the dietary record method or the recall method.62-64
Therefore, considering the advantages and shortcomings of
these methods, we used the FFQ method in the initial period
to assess the routine general intake pattern over a long
period of time because it is considered to be a useful method
for examining the relationship between diet and diseases.
Nevertheless, despite these limitations, our study reaches its
goal to show nutrition deficiencies in patients hospitalized
with pulmonary TB. These results are applicable to other
regions with similar TB and HIV incidences. Considering
that the complications of malnutrition increase the length
of hospital stay and costs,65 it is important to conduct
an adequate nutrition assessment of patients at hospital
admission. Nutrition supplementation may be an approach
to rapid recovery and improve outcomes in patients with
TB.66,67
Kirchmann Lazzari et al
Conclusion
This study identified a high prevalence of malnutrition in
patients with pulmonary TB admitted in a tertiary hospital,
regardless of the method used to assess nutrition status. In
HIV-positive patients, malnutrition measured by MAMC
was more frequent than in HIV-negative patients. Further-
more, we demonstrated that selenium and vitamin C intake
were significantly lower in HIV-positive patients than in
HIV-negative patients. Further studies with a larger sample
size are needed to clarify these findings.
Acknowledgments
We would like to acknowledge the support from the Interna-
tional Clinical Operational Health Services Research Training
Award (ICOHRTA/Fogarty International Center/National In-
stitutes for Health-NIH) and Johns Hopkins University (Johns
Hopkins Bloomberg School of Public Health).
Statement of Authorship
T. Kirchmann Lazzari is responsible for literature search, study
design, data collection, and review of the manuscript. G. Carra
Forte did literature search, study design, analysis of data, and
review of the manuscript. D. Rossato Silva did literature search,
study design, analysis of data, manuscript preparation, and
review of the manuscript.
References
1. Secretariat of Health Surveillance – Brazilian Ministry of Health. Epi-
demiological Bulletin 2016. Available at: www.saude.gov.br. Accessed:
June 2017.
2. World Health Organization. Global tuberculosis control: WHO report
2015. Available at: www.who.int. Accessed: June 2017.
3. Brazilian Ministry of Health. National Tuberculosis Control Program
2011. Available at: www.saude.gov.br. Accessed: June 2017.
4. Range N, Andersen AB, Magnussen P, et al. The effect of micronutrient
supplementation on treatment outcome in patients with pulmonary
tuberculosis: a randomized controlled trial in Mwanza, Tanzania. Trop
Med Int Health. 2005; 10(9):826-832.
5. Visser ME, Grewal HM, Swart EC, et al. The effect of vitamin A and
zinc supplementation on treatment outcomes in pulmonary tuberculo-
sis: a randomized controlled trial. Am J Clin Nutr. 2011;93(1):93-100.
6. Costa AE. Resposta imune humoral ao antı́geno registros do my-
cobacterium tuberculosis em pacientes com tuberculose e seus contatos
domiciliares. Rev Patol Trop. 2011;40(1):23-34.
7. Cruz RC, De Albuquerque MF, Campelo AR, et al. Pulmonary
tuberculosis: association between extent of the residual pulmonary
lesion and alteration in the lung function [in Portuguese]. Rev Assoc
Med Bras. 2008;54(5):406-410.
8. Ramalho RA. Avaliação Nutricional de pacientes com tuberculose
pulmonar antendidos na UISHL. Bol Pneumol Sanitaria. 2000;8(2):13-
19.
9. Teixeira HC, Abramo C, Munk ME. Immunological diagnosis of
tuberculosis: problems and strategies for success. J Bras Pneumol.
2007;33(3):323-334.
10. Nogueira CR. Aspectos antropométricos, bioquı́micos e sintoma-
tológicos em mulheres com tuberculose pulmonar. Rev Ciênc Med.
2006;15(4):281-288.
863
11. Ramos LM, Sulmonett N, Ferreira CS et al. Functional profile of
patients with tuberculosis sequelae in a university hospital. J Bras
Pneumol. 2006;32(1):43-47.
12. Van LM, Kumwenda JJ, Harries AD, et al. Malnutrition and the
severity of lung disease in adults with pulmonary tuberculosis in
Malawi. Int J Tuberc Lung Dis. 2004;8(2):211-217.
13. Cakir B, Yonem A, Guler S, et al. Relation of leptin and tumor necrosis
factor alpha to body weight changes in patients with pulmonary
tuberculosis. Horm Res. 1999;52(6):279-283.
14. Kim JH, Lee CT, Yoon HI, et al. Relation of ghrelin, leptin and inflam-
matory markers to nutritional status in active pulmonary tuberculosis.
Clin Nutr. 2010; 29(4):512-518.
15. Langhans W. Anorexia of infection: current prospects. Nutrition. 2000;
16(10):996-1005.
16. Schwenk A, Hodgson L, Rayner CF et al. Leptin and energy
metabolism in pulmonary tuberculosis. Am J Clin Nutr. 2003;77(2):392-
398.
17. van Crevel R, Karyadi E, Netea MG, et al. Decreased plasma
leptin concentrations in tuberculosis patients are associated with
wasting and inflammation. J Clin Endocrinol Metab. 2002;87(2):758-
763.
18. Ando M, Mori A, Esaki H, et al. The effect of pulmonary re-
habilitation in patients with post-tuberculosis lung disorder. Chest.
2003;123(6):1988-1995.
19. Maguire GP, Anstey NM, Ardian M, et al. Pulmonary tuberculosis,
impaired lung function, disability and quality of life in a high-burden
setting. Int J Tuberc Lung Dis. 2009;13(12):1500-1506.
20. Conde MB, Melo FA, Marques AM, et al. III Brazilian Thoracic Asso-
ciation Guidelines on tuberculosis. J Bras Pneumol. 2009;35(10):1018-
1048.
21. Diagnostic Standards and Classification of Tuberculosis in Adults and
Children. This official statement of the American Thoracic Society
and the Centers for Disease Control and Prevention was adopted
by the ATS Board of Directors, July 1999. This statement was en-
dorsed by the Council of the Infectious Disease Society of America,
September 1999. Am J Respir Crit Care Med. 2000;161(4 Pt 1):1376-
1395.
22. World Health Organization. BMI classification – Global Database on
Body Mass Index. 2015. Available at: www.who.int. Accessed: June
2017.
23. Heymsfield SB, Baugmgartier RN, Sheau-Fang R. Avaliação nutri-
cional da desnutrição por métodos antropométricos. In: Shils ME,
OJSMRAC, ed. Tratado de nutrição moderna na saúde e na doença. São
Paulo: Manoel; 2003;1168-1190.
24. de Onis M, Habicht JP. Anthropometric reference data for interna-
tional use: recommendations from a World Health Organization Expert
Committee. Am J Clin Nutr. 1996;64(4):650-658.
25. National Health and Nutrition Examination Survey (NHANES).
Anthropometric procedures manual. 2007. Available at: http://
www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf.
Accessed: June 2017.
26. Tang AM, Dong K, Deitcher M, Chung M, Maalouf-Manasseh Z,
Tumilowicz A, et al. Food and Nutrition Technical Assistance III Project
(FANTA) Washington, DC: United States Agency for International
Development (USAID); 2013. Use of cutoffs for Mid-Upper Arm Cir-
cumference (MUAC) as an Indicator or Predictor of Nutritional and
Health-Related Outcomes in Adolescents and Adults: A Systematic
Review. 2016:1-39.
27. Weber JKJ. Assessing nutrition. In: Nieginski E, ed. Health As-
sessment in Nursing. Philadelphia: Lippincott Williams & Wilkins;
2003.
28. Kyle UG, Genton L, Karsegard L, et al. Single prediction equation for
bioelectrical impedance analysis in adults aged 20–94 years. Nutrition.
2001; 17(3):248-253.
864
29. Kotler DP, Burastero S, Wang J, et al. Prediction of body cell mass, fat-
free mass, and total body water with bioelectrical impedance analysis:
effects of race, sex, and disease. Am J Clin Nutr. 1996;64(3 Suppl):489S-
497S.
30. Molina MC, Bensenor IM, Cardoso LO, et al. Reproducibility and
relative validity of the Food Frequency Questionnaire used in the
ELSA-Brasil [in Portuguese]. Cad Saude Publica. 2013;29(2):379-389.
31. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective
global assessment of nutritional status? JPEN J Parenter Enteral Nutr.
1987;11(1):8-13.
32. Ferguson M, Capra S, Bauer J, et al. Development of a valid and
reliable malnutrition screening tool for adult acute hospital patients.
Nutrition. 1999; 15(6):458-464.
33. Miyata S, Tanaka M, Ihaku D. Usefulness of the Malnutrition
Screening Tool in patients with pulmonary tuberculosis. Nutrition.
2012;28(3):271-274.
34. Reilly HM, Martineau JK, Moran A, et al. Nutritional screening–
evaluation and implementation of a simple Nutrition Risk Score. Clin
Nutr. 1995;14(5):269-273.
35. Bottoni A, Oliveira GPC, Ferrini MT, Waitzberg DL, Avaliação
nutricional: exames laboratoriais . In: Waitzberg DL, ed. Nutrição oral,
enteral e parenteral na prática clı́nica. 3rd. Edition. São Paulo: Atheneu,
2000:279-294.
36. Ramakrishnan K, Shenbagarathai R, Kavitha K, et al. Serum zinc and
albumin levels in pulmonary tuberculosis patients with and without
HIV. Jpn J Infect Dis. 2008;61(3):202-204.
37. Souza ARA, Tudisco ES. Avaliação nutricional de pacientes tubercu-
losos em tratamento ambulatorial. J Pneumol. 1992;18(4):167-170.
38. Bhargava A, Chatterjee M, Jain Y, et al. Nutritional status of adult
patients with pulmonary tuberculosis in rural central India and its
association with mortality. PLoS One. 2013;8(10):e77979.
39. Bhurosy T JR. Pitfalls Of Using Body Mass Index (BMI) In Assess-
ment Of Obesity Risk. Curr Res Nutr Food Sci. 2013;1(1):71-76.
40. Garn SM, Leonard WR, Hawthorne VM. Three limitations of the body
mass index. Am J Clin Nutr. 1986;44(6):996-997.
41. Miyata S, Tanaka M, Ihaku D. Subjective global assessment in patients
with pulmonary tuberculosis. Nutr Clin Pract. 2011;26(1):55-60.
42. Kant S, Gupta H, Ahluwalia S. Significance of nutrition in pulmonary
tuberculosis. Crit Rev Food Sci Nutr. 2015;55(7):955-963.
43. Gupta KB, Gupta R, Atreja A, et al. Tuberculosis and nutrition. Lung
India. 2009; 26(1):9-16.
44. Koethe JR, von Reyn CF. Protein-calorie malnutrition, macronutrient
supplements, and tuberculosis. Int J Tuberc Lung Dis. 2016;20(7):857-
863.
45. Zachariah R, Spielmann MP, Harries AD, et al. Moderate to severe
malnutrition in patients with tuberculosis is a risk factor associated with
early death. Trans R Soc Trop Med Hyg. 2002; 96(3):291-294.
46. Khan A, Sterling TR, Reves R, et al. Lack of weight gain and relapse
risk in a large tuberculosis treatment trial. Am J Respir Crit Care Med.
2006;174(3):344-348.
47. Oxlade O, Huang CC, Murray M. Estimating the Impact of Re-
ducing Under-Nutrition on the Tuberculosis Epidemic in the Central
Eastern States of India: A Dynamic Modeling Study. PLoS One.
2015;10(6):e0128187.
48. Si ZL, Kang LL, Shen XB et al. Adjuvant Efficacy of Nutrition
Support During Pulmonary Tuberculosis Treating Course: Systematic
Review and Meta-analysis. Chin Med J (Engl). 2015;128(23):3219-
3230.
49. Ramakrishnan CV, Rajendran K, Jacob PG, et al. The role of diet in
the treatment of pulmonary tuberculosis. An evaluation in a controlled
chemotherapy study in home and sanatorium patients in South India.
Bull World Health Organ. 1961;25:339-359.
Nutrition in Clinical Practice 33(6)
50. Swaminathan S, Padmapriyadarsini C, Sukumar B, et al. Nutritional
status of persons with HIV infection, persons with HIV infection and
tuberculosis, and HIV-negative individuals from southern India. Clin
Infect Dis 2008;46(6):946-949.
51. Gustafson P, Gomes VF, Vieira CS, et al. Clinical predictors for death
in HIV-positive and HIV-negative tuberculosis patients in Guinea-
Bissau. Infection. 2007;35(2):69-80.
52. Cegielski JP, McMurray DN. The relationship between
malnutrition and tuberculosis: evidence from studies in humans
and experimental animals. Int J Tuberc Lung Dis. 2004;8(3):286-
298.
53. Irlam JH, Siegfried N, Visser ME, et al. Micronutrient supplemen-
tation for children with HIV infection. Cochrane Database Syst Rev.
2013;(10):CD010666.
54. Shor-Posner G, Miguez MJ, Pineda LM, et al. Impact of selenium
status on the pathogenesis of mycobacterial disease in HIV-1-infected
drug users during the era of highly active antiretroviral therapy.
J Acquir Immune Defic Syndr. 2002; 29(2):169-173.
55. Bakaev VV, Duntau AP. Ascorbic acid in blood serum of patients
with pulmonary tuberculosis and pneumonia. Int J Tuberc Lung Dis.
2004;8(2):263-266.
56. Andosca JB, Foley JA. Calcium ribonate and vitamin C (Nu 240-10) in
the treatment of tuberculosis. Dis Chest. 1948;14(1):107-114.
57. Hemila H, Kaprio J, Pietinen P, et al. Vitamin C and other compounds
in vitamin C rich food in relation to risk of tuberculosis in male
smokers. Am J Epidemiol.1999;150(6):632-641.
58. McConkey M, Smith DT. The Relation Of Vitamin C Deficiency To
Intestinal Tuberculosis In The Guinea Pig. J Exp Med. 1933;58(4):503-
512.
59. Khameneh B, Fazly Bazzaz BS, Amani A, et al. Combination of
anti-tuberculosis drugs with vitamin C or NAC against different
Staphylococcus aureus and Mycobacterium tuberculosis strains. Microb
Pathog. 2016;93:83-87.
60. Selberg O, Selberg D. Norms and correlates of bioimpedance
phase angle in healthy human subjects, hospitalized patients, and
patients with liver cirrhosis. Eur J Appl Physiol. 2002;86(6):509-
516.
61. Yalin SF, Gulcicek S, Avci S, et al. Single-frequency and multi-
frequency bioimpedance analysis: What is the difference? Nephrol-
ogy. (Carlton) 2017. htts://doi.org/10.1111/nep.13042. [Epub ahead of
print].
62. Hong S, Choi Y, Lee HJ, et al. Development and validation of
a semi-quantitative food frequency questionnaire to assess diets of
Korean type 2 diabetic patients. Korean Diabetes J. 2010;34(1):
32-39.
63. Roman-Vinas B, Ortiz-Andrellucchi A, Mendez M, et al. Is the
food frequency questionnaire suitable to assess micronutrient intake
adequacy for infants, children and adolescents? Matern Child Nutr.
2010;6 Suppl 2:112-121.
64. Senekal M, Steyn NP, Nel J. A questionnaire for screening the
micronutrient intake of economically active South African adults.
Public Health Nutr. 2009; 12(11):2159-2167.
65. Linn BS. Outcomes of older and younger malnourished and well-
nourished patients one year after hospitalization. Am J Clin Nutr.
1984;39(1):66-73.
66. Mehta JB, Fields CL, Byrd RP Jr, et al. Nutritional status and
mortality in respiratory failure caused by tuberculosis. Tenn Med.
1996;89(10):369-371.
67. Volosevich GV. Blood T-lymphocyte functional activity in pulmonary
tuberculosis patients undergoing chemotherapeutic agent and vitamin
treatment [in Russian]. Probl Tuberk. 1982;(3):47-50.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Associations Between Enteral Nutrition and Acute December 2018 865–871

C 2018 American Society for

Respiratory Infection Among Patients in New York Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10017
Metropolitan Region Pediatric Long-Term Care Facilities wileyonlinelibrary.com

Marissa Burgermaster, PhD1 ; Meghan Murray, MPH, RN2 ;

Lisa Saiman, MD, MPH ; David S. Seres, MD, ScM, PNS, FASPEN5
3,4
;
and Elaine L. Larson, RN, PhD, CIC, FAAN6,7

Abstract
Background: Pediatric long-term care facilities (pLTCF) serve a complicated and resource-intensive patient population with high
usage of nutrition support. However, the relationship between nutrition support and outcomes among pLTCF residents is not well
understood. We described this relationship in three metropolitan New York pLTCF and a subsample of infants from one of these
facilities with a feeding disorders unit. Methods: In this prospective cohort study, we used logistic regression to assess relationships
between enteral nutrition (EN), and acute respiratory infections (ARI) among residents (n = 720, 50% male, mean age = 5.5 years,
mean number comorbidities = 2.1) and infant subsample (<1 year, n = 208, 50% male, mean number comorbidities = 2.0). We
tested these associations in multivariable models controlling for numbers of comorbidities and infections. Results: Many residents
received nutrition via percutaneous (59%) or nasogastric (15%) feeding tubes. In univariate analyses, residents receiving EN had
more comorbidities. In multivariable analyses, EN was associated with ARI (incidence rate ratio = 1.65, p < .001). Among infants
in the specialized unit, greater risk of ARI was associated only with percutaneous (incidence rate ratio = 1.94, p < .01) feeding. EN
was associated with lower odds of being discharged home (OR = 0.45, p < .01). Conclusion: The prevalence of EN, complexity of
cases, and necessity of long-term EN make nutrition support important in pLTCFs. Differences in EN types and adverse outcomes
in the infant subsample suggest different care is necessary for this subpopulation. Results provide context for improving quality of
care and clinician/caregiver education for this population. (Nutr Clin Pract. 2018;33:865–871)

Keywords
enteral nutrition; long-term care; pediatrics; respiratory tract infections; skilled nursing facilities

Pediatric long-term care facilities (pLTCFs) primarily serve
children who need extensive assistance with activities of
daily living.1 Residents of pLTCFs have substantial, spe-
cialized healthcare needs; chronic comorbid conditions
associated with medical fragility; severe functional limita-
tions; and require frequent, ongoing healthcare use.2 They
often have respiratory, nutrition, and/or neuromuscular
disorders.1 With ongoing advances in critical care for in-
fants and children, survival rates of premature infants and
children with birth defects and genetic disorders continue to

From the 1 Division of Preventive Medicine and Nutrition, Department of Medicine, Columbia University Medical Center, New York, New
York, USA; 2 School of Nursing, Columbia University Medical Center, New York, New York, USA; 3 Department of Pediatrics, Columbia
University Medical Center, New York, New York, USA; 4 Department of Infection Prevention & Control, NewYork-Presbyterian Hospital, New
York, New York, USA; 5 Division of Preventive Medicine and Nutrition, Department of Medicine and Institute of Human Nutrition, Columbia
University Medical Center, New York, New York, USA; 6 Associate Dean for Research, Anna Maxwell Professor of Nursing Research, School of
Nursing, Columbia University, New York, New York, USA; and 7 Department of Epidemiology, Mailman School of Public Health, Columbia
University, New York, New York, USA.
Financial disclosure: This work was supported by the Agency for Healthcare Research and Quality, U.S. Department of Health and Human
Services (Keep It Clean for Kids: The KICK Project, R01HS021470). M.B.’s involvement in the preparation of this manuscript was supported by
National Institutes of Health training grants T32HL007343 and T15LM007079.
Conflicts of interest: None.
This article originally appeared online on February 15, 2018.
Corresponding Author:
David S. Seres, MD, ScM, PNS, FASPEN, Department of Medicine, Columbia University Irving Medical Center, 630 W 168th Street, P&S 9-501,
New York, NY 10032, USA.
Email: dseres@columbia.edu
866
improve.3 Likewise, improved treatment of acute illnesses
has increased survival, as well as disability.4 Although
estimates suggest that children with chronic comorbid con-
ditions represent <1% of the population and those who
reside in pLTCF make up a smaller fraction, this growing
and resource-intense healthcare sector has been found to
account for disproportionate healthcare costs.2,5-8
Many children with chronic comorbid conditions re-
quire the support of medical technology to maximize their
functioning.2 A 2011 retrospective analysis of data from 4
U.S. children’s hospitals indicated that 56% of children who
received complex care coordination services had gastros-
tomy tubes.9 Children with chronic comorbid conditions
who require technology assistance, including feeding tubes,
incur 3.5 times the costs of other children with chronic
comorbid conditions.5 Enteral nutrition (EN; ie, nasal or
percutaneous tube feeding) is, therefore, an important com-
ponent of pLTCF care; however, the prevalence of EN and
potential adverse impacts of EN in this unique population
has not yet been well described in the literature.
Some pLTCFs also admit children who need close ob-
servation, medication administration, rehabilitation, and/or
family education for shorter-term stays; many residents are
premature infants with acute nutrition needs.1 Some of
these infants cannot feed safely or efficiently because of
inability to suck, swallow, or coordinate breathing with
swallowing.10 Clinical indications for EN in both very
premature as well as medically complex infants11 and re-
ported relationships between extrauterine growth restriction
and neurodevelopmental impairment have driven recent
interest in optimizing EN in this population,12 but there
is wide variation and lack of consensus for nourishment
strategies.13
A recent retrospective cohort study found that of 15,642
children having a gastrostomy tube placed, 8.6% had an
emergency department visit and 3.9% were readmitted
within 30 days of discharge; infection, mechanical com-
plication, and replacement were the primary gastrostomy-
related reasons for these visits.14 Children in pLTCFs
are at particularly high risk for healthcare-associated in-
fections due to their underlying conditions; mechanical
supports; and multiple contacts with medical providers,
family members, and other children with chronic comorbid
conditions.15 We performed an epidemiological study of
infections in residents in 3 pLTCFs in the New York City
metropolitan area and found use of EN ranged from 77%
to 85% of residents.16 Percutaneous feeding tubes (PFTs),
but not nasogastric feeding tubes (NGFTs), were an inde-
pendent predictor of acute respiratory infection (ARI).16
In this exploratory study, we present a more detailed
analysis focused on nutrition support to provide a bet-
ter understanding of the relationship between nutrition
support and adverse outcomes among residents in pLTCFs.
In addition to a more detailed description of the clinical
Nutrition in Clinical Practice 33(6)
characteristics and outcomes for residents receiving EN,
we also separately examined a subsample of infants, many
of whom received nutrition support for different reasons
than the full sample. We aimed to: (1) describe the use of
EN and clinical characteristics of children at pLTCFs who
are receiving EN; (2) describe the use of EN and clinical
characteristics among residents younger than 1 year in 1 of
the facilities with a specialized unit for infants with feeding
difficulties; and (3) identify relationships between feeding
type and outcomes including infections, hospitalization,
and discharge home in both the whole sample and the infant
subsample.
Methods
We conducted a prospective cohort study of residents at 3
pLTCFs from September 2012 to December 2015 as part
of an intervention study to reduce healthcare-associated
infections, the Keep It Clean for Kids (KICK) Project
(R01HS021470).15,17 The 3 pLTCFs were independent fa-
cilities located in the New York City metropolitan area,
ranging in size from 54 to 137 beds, with therapeutic, medi-
cal, and school services on-site. This study was approved by
the BLINDED Institutional Review Board and the ethics
boards of the pLTCFs with approval for waiver of written
documentation of consent.
Sample
Analyses included all residents during 2012–2015 except for
those admitted for respite care only. Subanalyses comparing
PFT, NGFT, and no EN in infants included residents
younger than 1 year at the time of enrollment from a single
facility.
Residents were enrolled from 1 of 3 pLTCFs in the
New York metropolitan region, including 54-bed, 97-bed,
and 137-bed facilities. Each of these pLTCFs is a stand-
alone facility and provides care for only pediatric residents.
On average, 30% (range 22%–44%) of residents had tra-
cheostomies and 7% (range 5%–10%) were mechanically
ventilated. About 25% of residents are hospitalized at
least once each year. Overall, 70% (range 58%–86%) of
residents in these facilities have a neurological disorder and
50% (range 36%–65%) have a respiratory disorder. The 3
pLTCFs provide similar care, each with in-house physicians,
nursing staff, dietitians, and therapists, including physical,
occupational, music, and art therapies.
Variables of Interest
Dependent variables. We measured 3 dependent variables:
ARI, infection-related hospitalization, and discharge home.
Acute respiratory infection. Infections were defined based
on clinical diagnosis by a physician or nurse practitioner.
Burgermaster et al
Study staff conducted ongoing medical record reviews to
collect data related to clinician diagnosis, signs and symp-
toms, and diagnostic testing related to infections. ARIs
included both viral and nonviral acute upper respiratory
infections, as well as pneumonia, pharyngitis, and sinusitis.
Infection-related hospitalization. Hospital admissions re-
lated to infection were defined as transfers to acute care for
>24 hours with the indication of an infection diagnosed at
the pLTCF. This information was collected from medical
record reviews conducted by study staff.
Discharge home. Discharge disposition was obtained from
the medical records by the study staff. Home was defined as
the legal guardian’s residence. Discharge home was based
on a resident’s last discharge and his or her status at the
study’s conclusion; that is, if a resident was discharged home
and then readmitted, the resident was not considered to be
discharged home.
Independent variable. In these analyses, nutrition support
was the independent variable. We defined EN to include any
tube feeding (ie, nasal or percutaneous tube feeding). Few
residents enrolled during the study period had parenteral
nutrition (n = 2); these residents were excluded from analy-
ses. Residents rarely received oral feeding in addition to EN;
therefore, it was not considered in analyses. For subanalyses,
we compared EN via NGFT (including nasogastric and
nasojejunal feeding tubes) and PFT (including all types of
gastrostomy, jejunostomy, or other feeding enterostomy).
Eight residents were documented as having both PFT and
NGFT at admission. For this study, they were included with
residents having PFT.
Covariates. We included patient characteristics and comor-
bidities as covariates in our analyses. Patient demographics
and comorbidities were collected from medical records at
enrollment, including information about medical devices.
Comorbidities were based on the admitting diagnoses for
each resident and included the presence of ࣙ1 neurological
(eg, cerebral palsy, epilepsy), respiratory (eg, bronchopul-
monary dysplasia, respiratory failure), cardiovascular (eg,
congenital heart disease, hypertension), gastrointestinal (eg,
short bowel disease, feeding disorder), metabolic (eg, mi-
tochondrial disease), or immune disorder (eg, HIV/AIDS).
In previous research, our team found that the number of
chronic comorbid conditions rather than the type of condi-
tion was predictive of ARI.16 Comorbidities were summed
by type for each resident (ie, a resident with a neurological
disorder and a respiratory disorder has 2 comorbidities
independent of how many individual neurological and
respiratory disorders the resident may have); this variable
was called number of comorbidity types throughout the
manuscript. Number of infections was a count of infections
867
each patient was diagnosed with, including infections of the
respiratory tract, conjunctivitis/otitis, skin and soft tissue,
urinary tract, gastrointestinal, and bloodstream.
Analyses
Descriptive statistics were used to characterize the study
sample of all residents and subsample of infants. We
assessed relationships between EN, clinical characteristics,
and infections in the entire study sample using binomial
logistic regression. We assessed relationships among NGFT,
PFT, clinical characteristics, and infections among the
subsample of infants using multinomial logistic regression.
For all analyses involving rates, Poisson regression was
used. To better understand the independent influence of
EN among residents and types of EN among infants, we
also tested these associations while controlling for clinically
relevant covariates, including number of comorbidities and
number of infections, in multivariable models. Statistical
significance was set a priori at .05. Analyses were conducted
with SAS version 9.3 (SAS Institute, Cary, NC).
Results
Of 720 pLTCF residents enrolled in the study, 59% had a
PFT and 15% had a NGFT. Among residents older than
1 year, only 25 (5%) received feeding via NGFT. Residents
receiving any EN were older and had resided in the facilities
for longer; they also had more comorbidity types, more
gastrointestinal disorders, and, specifically, more feeding
disorders. Residents receiving EN had more infections,
and fewer residents receiving EN were discharged home.
Nutrition, comorbidity types, discharge disposition, and
infections in residents who did and did not receive EN are
described in Table 1.
Among the 208 infants in the subsample, frequency
of feeding type varied for different disorders. Specifically,
infants with feeding disorders had higher odds of being fed
via NGFT. Infants fed via PFT had more infections, and
fewer infants with PFT were discharged home. Nutrition,
comorbidity types, discharge disposition, and infections in
infants with NGFT, PFT, or no feeding tube are described
in Table 2.
Associations between feeding type and outcomes of
interest were measured in all residents and the subsample
of infants, controlling for number of comorbidity types.
As gender and age were not significant in the bivariable
models, they were not included in the multivariable models.
These results are described in Table 3. Residents with
EN had a higher rate of ARI compared with residents
without EN when controlling for number of comorbidity
types. Residents with EN were significantly less likely to
be discharged home as compared with residents without
EN when controlling for number of comorbidity types.
However, residents with EN were not significantly more
868 Nutrition in Clinical Practice 33(6)

Table 1. Characteristics of Residents With and Without Enteral Feeding Tubes at 3 Pediatric Long-Term Care Facilities,
2012–2015, and Unadjusted Comparisons.

Residents Residents Not

Receiving ENa Receiving EN
(n = 526, 73%) (n = 194, 27%) OR/IRRb P Value

Mean age at enrollment, y (range) 5.2 (0.02–22.3) 5.8 (0.02–26.2) 0.98 .23

Male sex, n (%) 254 (48) 107 (55) 0.76 .10

Mean no. of comorbidities (range) 2.3 (0–5) 1.6 (0–4) 2.01 <.001
Respiratory disorder 291 (55) 70 (36) 2.19 <.001
Neurological disorder 398 (76) 107 (55) 2.53 <.001
Gastrointestinal disorder 350 (67) 76 (39) 3.09 <.001
Feeding disorder 139 (26) 28 (14) 2.13 <.001
Dysfunctional suck/swallow 12 (2) 0 (0) — —
Short bowel disease 17 (3) 4 (2) 1.57 .42
Failure to thrive 57 (11) 8 (4) 2.80 .008

Mean length of enrollment, d (range) 606.2 (1–1217) 398.1 (4–1217) 1.001 <.001

Rate of infection-related hospital admissions 1.01 0.67 1.50c .007

per 1000 resident-days

Discharged home,d n (%) 154 (29) 99 (51) 0.40 <.001

Mean no. of infections (range) 3.42 (0–32) 1.33 (0–17) 1.21 <.001
ARI 298 (57) 55 (28) 3.30 <.001
Rate of ARI per 1000 resident-days 3.57 2.02 1.77c <.001
Pneumonia 109 (21) 13 (7) 3.64 <.001
Mean no. of pneumonia infections (range) 0.33 (0–8) 0.11 (0–4) 2.01 <.001

ARI, acute respiratory infection; EN, enteral nutrition; IRR, incidence rate ratio; OR, odds ratio.
a EN includes both nasal (including nasogastric and nasojejunal feeding tubes) and percutaneous feeding tubes (including all types of

gastrostomy, jejunostomy, or other feeding enterostomy).

b OR was determined by logistic regression. IRR was determined by Poisson regression.
c IRR reported for Poisson regression with continuous outcome.
d Based on last discharge only, does not include residents who were discharged home and subsequently returned to the same facility.

likely to be hospitalized for an infection, controlling for
number of comorbidity types.
Among the infants, only those with a PFT had an
increased rate of acute respiratory tract infections compared
with those without EN when adjusting for number of co-
morbidity types. Neither infants with an NGFT nor infants
with a PFT had a statistically significantly increased rate of
hospitalizations for infection when compared with infants
without EN, nor were they significantly less likely to be dis-
charged home, controlling for number of comorbidity types.
Discussion
Seventy percent of pLTCF residents in our sample received
EN, and the frequency of EN via PFT (59%) among our
sample is consistent with the frequency of gastrostomy
(56%) reported in a nationwide study of health resource
use among children with chronic comorbid conditions.9
That residents receiving EN had, on average, more types
of comorbidities suggests that these residents were more
complicated, which is consistent with their lower odds
of being discharged home when controlling for number
of comorbidity types. The prevalence of EN, complexity
of cases, and necessity of long-term EN underscore the
importance of nutrition support and clinical experts to
manage it properly for this population.
Our finding that the pLTCF residents in this study who
received EN had more ARIs and pneumonia is consistent
with a study of respiratory infection among a sample of
children with profound intellectual and multiple disability
in day care centers, 42% with gastrostomy, which found that
feeding difficulty and gastrostomy both increased risk for
ARI.18 The greater number of mean number of infections
per patient among the patients receiving EN may be par-
tially explained by other types of infections, including skin
infections at the insertion site. Superficial skin infections
have been related to gastrostomy in pediatric acute care
settings, with rates ranging from 7% to 25%.14,19,20 Future
Burgermaster et al 869

Table 2. Characteristics of Infants Younger Than 1 Year With Nasogastric, Percutaneous, and No Feeding Tubes at a Single
Pediatric Long-Term Care Facility With Specialized Unit for Infants, 2012–2015, and Unadjusted Comparisons.

Infants With Infants With Infants Without a

NGFTsa PFTsb Feeding Tube
(n = 65, 31%) (n = 76, 37%) (n = 67, 32%) P Valuec

Mean age at enrollment, y (range) 0.32 (0.05–0.90) 0.38 (0.02–0.99) 0.32 (0.02–0.92) .17

Male sex, n (%) 27 (42) 38 (50) 38 (56) .22

Mean no. of comorbidities (range) 1.95 (0–4) 2.14 (1–4) 1.82 (0–4) .12
Respiratory disorder 28 (43) 45 (59) 36 (54) .16
Neurological disorder 22 (34) 35 (46) 24 (36) .27
Gastrointestinal disorder 50 (77) 52 (68) 36 (54) .02
Feeding disorder 42 (65) 31 (41) 21 (31) <.001
Dysfunctional suck/swallow 2 (3) 1 (1) 0 (0) —
Short bowel disease 2 (3) 6 (8) 3 (4) .44
Failure to thrive 8 (12) 8 (11) 2 (3) .17

Mean length of enrollment, d (range) 78 (4–1190) 240 (1–1217) 168 (4–1217) .009

Rate of infection-related hospital admissions per 1.18 1.15 0.89d .77

1000 resident-days

Discharged homee 44 (68) 36 (47) 44 (66) .02

Mean no. of infections (range) 0.46 (0–4) 1.58 (0–9) 0.67 (0–6) .002
ARI 11 (17) 24 (32) 14 (21) .11
Rate of ARI per 1000 resident-days 2.56 4.17 2.05d .004
Pneumonia 2 (3) 7 (9) 2 (3) .19
Mean no. of pneumonia infections (range) 0.03 (0–1) 0.11 (0–2) 0.04 (0–2) .28

ARI, acute respiratory infection; NGFT, nasogastric feeding tube; PFT, percutaneous feeding tube.
a NGFTs (including NGFT and nasojejunal feeding tubes) and PFTs.
b PFTs (including all types of gastrostomy, jejunostomy, or other feeding enterostomy).
c Based on logistic regression for all variables except rate of infection-related hospital admissions and rate of ARI, which were based on Poisson

regression.
d Incidence rate ratio reported for Poisson regression with continuous outcome.
e Based on last discharge only, does not include residents who were discharged home and subsequently returned to the same facility.

research should examine the prevalence of EN-related skin
infections in pLTCF.
Although 68% of the infants in our subsample received
any type of EN, PFT was less frequent than in our overall
sample (37% vs 59%). This may be because some infants are
admitted to a pLTCF for short-term intensive feeding. Our
findings that many more infants with NGFT had feeding
disorder–related gastrointestinal comorbidities is consistent
with our observation that infants with different feeding tube
types are admitted to pLTCFs for different reasons and,
therefore, need different care.
This is the first study, to our knowledge, to report
respiratory infection rates in relation to EN among in-
fants in a pLTCF. The finding that even when controlling
for number of comorbidities, the incidence rate of ARI,
including pneumonia, was significantly higher for infants
with PFT contradicts common wisdom associating NGFT
with an increased incidence of pneumonia relative to PFTs.
The number of infants with NGFT and no EN who were
experiencing pneumonia was very similar (2 cases in each
condition) compared with 7 cases in infants with PFT, sug-
gesting that NGFTs may not be a risk factor for pneumonia
compared with PFT in this population. There is limited
research comparing ARI among patients with NGFT and
PFT in pediatric population. A retrospective cohort study
of children 0–14 years old at a single children’s hospital
revealed similar rates of pneumonia among patients with
long-term (>3 months) use of PFT or NGFT, but that
complications (excluding pneumonia and food refusal) were
almost twice as common among patients with PFT (80%)
than NGFT (46%).21
Limitations of this study include its observational na-
ture: causality cannot be inferred from these analyses. Fur-
ther, the results may not be generalizable to pLTCF in other
regions. We only collected information about NGFT or PFT
at enrollment into the study. Therefore, it is possible that
some residents who received EN at enrollment had tubes
removed during the study period, and others who initially
870 Nutrition in Clinical Practice 33(6)

Table 3. Comparison of Acute Respiratory Infection, and feeding tube problems have been cited as reasons
Hospitalization, and Home Discharge Outcomes by Type of for cessation or holding of tube feeds.24-26 Due to the
Feeding in All Residents and Infants at a Single Pediatric complexity of nutrition support required by this population,
Long-Term Care Facility With Specialized Unit.
future research should also describe the role of registered
IRRa /ORb P dietitians and other nutrition experts in pLTCFs.
(95% CI) Value Our findings suggest that care providers in pLTCFs
should monitor residents who are receiving EN for respira-
All patients, comparison between ENc and no EN tory infections and take appropriate precaution to prevent
Rate of ARId 1.65a,e (1.39, 1.97) <.001e them. Additional testing is necessary to identify pathophys-
Discharged home 0.61b,e (0.42, 0.89) <.01e
iological pathways that explain the link between EN and
Rate of hospitalizationf 1.17a (0.68, 1.27) .65
Infants, comparison among NGFT, PFT,h and no EN
g respiratory infection, as well as specific risk factors and
Rate of ARI intervention targets; however, because EN and PFT are
NGFT vs no EN 1.19a (0.60, 2.36) .621 typically nonmodifiable risk factors, prevention is crucial.
PFT vs no EN 1.94a,e (1.21, 3.11) .006e Nutrition support care providers can play an important
Discharged home role in reducing infection by engaging in evidence-based
NGFT vs no EN 1.04b (0.50, 2.18) .90 practices and by educating key stakeholders. Residents
PFT vs no EN 0.57b (0.28, 1.14) .11 fed via EN, as well as staff and visitors, should receive
Rate of hospitalization
vaccinations for respiratory viruses. Additional preventive
NGFT vs no EN 1.47a (0.53, 4.09) .46
PFT vs no EN 0.75a (0.32, 1.77) .52 clinical practices (eg, attention to secretions, engagement
in movement and activity) and policies (eg, prohibiting ill
ARI, acute respiratory infection; EN, enteral nutrition; IRR, staff or visitors, enacting transmission reduction protocols)
incidence rate ratio; NGFT, nasogastric feeding tube; OR, odds ratio; should be followed when caring for residents with EN.
PFT, percutaneous feeding tube.
a IRR reported for Poisson regression with continuous outcome.
b OR reported for binomial logistic regression.
c EN includes both nasal (including NGFTs and nasojejunal feeding
Conclusions
tubes) and PFTs (including all types of gastrostomy, jejunostomy, or Our description of nutrition support in pLTCFs indicates
other feeding enterostomy). that most residents, including infants, receive EN, and that
d ARI, including pneumonia.
e Statistically significant result. Multivariable regression was EN is associated with ARI. Among infants, PFTs were
conducted while controlling for number of comorbidity types. For associated with greater risk for ARI. Furthermore, EN was
discharged home and rate of hospitalization outcomes, we also associated with lower odds of being discharged home in the
controlled for number of infections. whole sample. The results presented in this article provide
f Based on last discharge only, does not include residents who were

discharged home and subsequently returned to the same facility.
g NGFT (including NGFTs and nasojejunal feeding tubes).
h PFT (including all types of gastrostomy, jejunostomy, or other
feeding enterostomy).
a context for improving quality of care, as well as clinician
and caregiver education, for this challenging and growing
patient population.

did not receive EN had feeding tubes inserted during the
study period. Further, this precluded any assessment of
nutrition adequacy. We did not have information about
indications for EN or, more specifically, for the choice of
PFT or NGFT for individual residents. Finally, because of
the small number of residents older than 1 year who were fed
via NGFT, we were not able to compare adverse outcomes
between NGFT and PFT in the full sample.
Future research should examine nutrition adequacy and
mechanical EN complications in pLTCFs. Nutrition sup-
port care is especially important for children with chronic
comorbid conditions because frequent reassessment of
estimated energy requirement is necessary due to growth
and disease progression,22 and because many experience
feeding difficulties, malnutrition, and inadequate micronu-
trient intake.23 In pediatric intensive care units, perceived
intolerance, high gastric residual volumes, nausea and vom-
iting, fasting around procedures, hemodynamic instability,
Acknowledgments
The authors would like to acknowledge the administration and
staff at the 3 contributing facilities, Rebecca Rudel for her
contribution to the literature review, and Bevin Cohen for her
consultation on statistical analyses.
Statement of Authorship
M. Burgermaster, L. Saiman, D.S. Seres, and E.L. Larson
contributed to the conception or design; M. Burgermaster, M.
Murray, D.S. Seres, and E.L. Larson contributed to the ac-
quisition, analysis, or interpretation; M. Burgermaster and M.
Murray drafted the manuscript; M. Burgermaster, M. Murray,
L. Saiman, D.S. Seres, and E.L. Larson critically revised the
manuscript, gave final approval, and agree to be accountable
for all aspects of work ensuring integrity and accuracy.
References
1. O’Malley M, Hutcheon RG. Genetic disorders and congenital mal-
formations in pediatric long-term care. J Am Med Dir Assoc.
2007;8(5):332-334.
Burgermaster et al
2. Cohen E, Kuo DZ, Agrawal R, et al. Children with medical complexity:
an emerging population for clinical and research initiatives. Pediatrics.
2011;127(3):529-538.
3. Stoll BJ, Hansen NI, Bell EF, et al. Trends in care practices, morbidity,
and mortality of extremely preterm neonates, 1993–2012. JAMA.
2015;314(10):1039-1051.
4. Hjern A, Lindblad F, Boman KK. Disability in adult survivors of
childhood cancer: a Swedish national cohort study. J Clin Oncol.
2007;25(33):5262-5266.
5. Cohen E, Berry JG, Camacho X, Anderson G, Wodchis W, Guttmann
A. Patterns and costs of health care use of children with medical
complexity. Pediatrics. 2012;130(6):e1463-e1470.
6. Glendinning C, Kirk S, Guiffrida A, Lawton D. Technology-dependent
children in the community: definitions, numbers and costs. Child Care
Health Dev. 2001;27(4):321-334.
7. Simon TD, Berry J, Feudtner C, et al. Children with complex chronic
conditions in inpatient hospital settings in the United States. Pediatrics.
2010;126(4):647-655.
8. Berry JG, Hall DE, Kuo DZ, et al. Hospital utilization and character-
istics of patients experiencing recurrent readmissions within children’s
hospitals. JAMA. 2011;305(7):682-690.
9. Berry JG, Agrawal R, Kuo DZ, et al. Characteristics of hospitalizations
for patients who use a structured clinical care program for children with
medical complexity. J Pediatr. 2011;159(2):284-290.
10. Lau C. Development of infant oral feeding skills: what do we know?
Am J Clin Nutr. 2016;103(2):616S-621S.
11. Khan Z, Marinschek S, Pahsini K, et al. Nutritional/growth status in
a large cohort of medically fragile children receiving long-term enteral
nutrition support. J Pediatr Gastroenterol Nutr. 2016;62(1):157-160.
12. Stevens TP, Shields E, Campbell D, et al. Variation in enteral feeding
practices and growth outcomes among very premature infants: a
report from the New York State Perinatal Quality Collaborative. Am
J Perinatol. 2016;33(1):9-19.
13. Mills L, Modi N. Clinician enteral feeding preferences for very preterm
babies in the UK. Arch Dis Child Fetal Neonatal Ed. 2015;100(4):F372-
F373.
14. Goldin AB, Heiss KF, Hall M, et al. Emergency department visits
and readmissions among children after gastrostomy tube placement.
J Pediatr. 2016;174:139-145.e132.
871
15. Murray MT, Pavia M, Jackson O, et al. Health care–associated
infection outbreaks in pediatric long-term care facilities. Am J Infect
Control. 2015;43(7):756-758.
16. Saiman L, Maykowski P, Murray M, et al. Epidemiology of infections
in pediatric long-term care facilities. JAMA Pediatr. 2017;171(9):872-
878. https://doi.org/10.1001/jamapediatrics.2017.1482
17. Larson EL, Cohen B, Murray M, Saiman L. Challenges in conducting
research in pediatric long-term care facilities. Clin Pediatr. 53(11):1041-
1046.
18. Proesmans M, Vreys M, Huenaerts E, et al. Respiratory morbidity
in children with profound intellectual and multiple disability. Pediatr
Pulmonol. 2015;50(10):1033-1038.
19. Jacob A, Delesalle D, Coopman S, et al. Safety of the one-step
percutaneous endoscopic gastrostomy button in children. J Pediatr.
2015;166(6):1526-1528.
20. Viktorsdóttir MB, Óskarsson K, Gunnarsdóttir A, Sigurdsson L.
Percutaneous endoscopic gastrostomy in children: a population-based
study from iceland, 1999–2010. J Laparoendosc Adv Surg Tech.
2015;25(3):248-251.
21. Ricciuto A, Baird R, Sant’Anna A. A retrospective review of enteral
nutrition support practices at a tertiary pediatric hospital: a compari-
son of prolonged nasogastric and gastrostomy tube feeding. Clin Nutr.
2015;34(4):652-658.
22. Mahan L, Escott-Stump S, Raymond J, Krause M. Krause’s Food &
the Nutrition Care Process, 13th ed. St. Louis, MO: Elsevier/Saunders;
2012.
23. Wolff J, Wiltzer Karim J, Ronis S, Grossberg R. Effects of nutritional
care coordination on children with medical complexity. Dev Med Child
Neurol. 2015;57:107.
24. Skillman HE, Mehta NM. Nutrition therapy in the critically ill child.
Curr Opin Crit Care. 2012;18(2):192-198.
25. Leong AY, Cartwright KR, Guerra GG, Joffe AR, Mazurak VC,
Larsen BM. A Canadian survey of perceived barriers to initiation
and continuation of enteral feeding in PICUs. Pediatr Crit Care Med.
2014;15(2):e49-e55.
26. Dokken M, Rustøen T, Stubhaug A. Indirect calorimetry reveals
that better monitoring of nutrition therapy in pediatric intensive
care is needed. JPEN J Parenter Enteral Nutr. 2015;39(3):344-
352.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Reducing Blood Glucose Testing Is Safe in Patients Receiving December 2018 872–878

C 2018 American Society for

Parenteral Nutrition Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10020
wileyonlinelibrary.com

Ashley Ratliff, MS, RD, LD, CNSC1 ; Amy Nishnick, RD, LD, CNSC1 ;
Robert DeChicco, MS, RD, LD, CNSC1 ; and Rocio Lopez, MS, MPH2

Abstract
Background: Hyperglycemia is a common adverse event associated with parenteral nutrition (PN); however, there is no consensus
on optimal glucose monitoring. Unnecessary point-of-care (POC) blood glucose (BG) testing can result in additional healthcare
cost and discomfort to the patient. This study’s aim was to determine whether decreasing the frequency of POC glucose testing in
patients receiving PN can reduce costs without increasing the frequency of glycemic events. Methods: This study examined adult,
noncritically ill patients who require PN and are managed by a nutrition support team. Guidelines were developed for the initiation
and discontinuation of POC BG monitoring. Pre-intervention and post-intervention data were collected for each patient, including
daily POC glucose and glucose from the basic metabolic panel (BMP). Equivalency testing was completed for each group. Results:
Data were collected on 171 subjects (86 preguideline, 85 postguideline). The total number of POC tests per patient was significantly
less in the postintervention period (median 23.0 vs 25.5; P = .011), as well as the average number of POC tests per patient per
day (2.9 ± 1.4 vs 3.6 ± 1.4; P < .001). The average BMP glucose (126 mg/dL preintervention vs 129 mg/dL postintervention; P
= .005) and the number of hypoglycemic and hyperglycemic events per 100 tests (both P < .001) were equivalent. Conclusions:
Implementation of new guidelines resulted in decreased frequency of POC BG tests by 20%, which may impact cost savings and
patient comfort in hospitalized patients receiving PN without increasing average BG level or glycemic events. (Nutr Clin Pract.
2018;33:872–878)

Keywords
blood glucose; cost savings; glucose monitoring; hyperglycemia; nutritional support; parenteral nutrition; quality improvement

Hyperglycemia is a common adverse event associated with
parenteral nutrition (PN), occurring in up to 50% of hos-
pitalized patients.1 Hyperglycemia-induced complications
in patients who are receiving PN increase morbidity and
mortality.2-5 The optimal blood glucose (BG) level in hos-
pitalized patients who are receiving PN is not clear. Tight
BG control (ie, <110 mg/dL) in critically ill patients in a
surgical intensive care unit (ICU) demonstrated improved
outcomes,6 but these results were not duplicated in patients
in a medical ICU.7 One potential factor for the lack of
improved outcomes with tight BG control in medical ICU
patients was increased episodes of hypoglycemia. For this
reason, the American Society for Parenteral and Enteral
Nutrition (ASPEN) recommends a target BG range of 140–
180 mg/dL in hospitalized patients who are receiving PN
nutrition support.8
Point-of-care (POC) capillary testing is the standard for
monitoring BG because it is fast, easy, and relatively precise.
However, there is no consensus on the optimal frequency,
timing, and duration of POC testing, or when it is appro-
priate to decrease or discontinue monitoring when it is no
longer necessary. Lin et al9 described monitoring BG based
on the presence of hyperglycemia. Patients were divided
into quartiles based on their BG levels. Lin et al9 found
an increased risk for complications (eg, death, infection,
cardiac complications, or acute renal failure) between the
lowest BG levels (<114 mg/dL) and the highest BG levels
(>180 mg/dL). All patients who were receiving PN had
BG measurements twice weekly via serum glucose, whereas
patients with hyperglycemia had BG monitoring every
6 hours via capillary finger sticks. Cheung et al10 reported
hyperglycemia increased the risk for complications in pa-
tients who were receiving PN. Monitoring was described
From the 1 Nutrition Support Team, Center for Human Nutrition
Cleveland Clinic, Cleveland, Ohio, USA; and 2 Quantitative Health
Sciences JJN3-01, Cleveland Clinic, Cleveland, Ohio, USA.
Financial disclosure: None.
Conflicts of interest: None.
This article originally appeared online on March 12, 2018.
Corresponding Author:
Ashley Ratliff, MS, RD, LD, CNSC, Center for Human Nutrition
TT-22, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195,
USA.
Email: ratlifa@ccf.org
Ratliff et al
as once-daily BG monitoring, with increased frequency to
every 4 hours if the patient was hyperglycemic. Neither
study commented on the effect of the frequency of BG
monitoring on outcomes. Sarkisian et al2 evaluated BG
using glucometer readings and central blood draws as part
of standard PN laboratory results. Frequency of glucometer
readings ranged from 2 to 7 times in the first 48 hours and
3 to 18 times in the following 3–9 days on PN. There was
no association with the frequency of BG monitoring and
mortality. Pleva et al1 reported hyperglycemia in noncrit-
ically ill patients using daily laboratory tests for the first
3 days, which transitioned to every Monday, Wednesday,
and Friday thereafter. POC glucose testing was done every
6 hours. Once the BG was well controlled and the goal
rate of PN was reached, the POC testing was decreased or
discontinued. ASPEN recommends POC BG monitoring
every 6 hours, or more frequently based on the degree of
hyperglycemia, for patients starting on PN, then 3 times
daily until BG is controlled.11 During cyclic infusion of
PN, POC BG testing should be done 30 minutes to 1 hour
postinfusion to assess for rebound hypoglycemia.11
The current practice for monitoring patients initiated
on continuous PN at a large medical center followed by
a nutrition support team (NST) includes a daily basic
metabolic panel (BMP) and POC BG tests every 6 hours
at 6:00 AM, 12:00 PM, 6:00 PM, and 12:00 AM. Patients who
are receiving cyclic PN receive POC BG tests 3 times daily: 2
hours after starting the cycle, halfway through the cycle, and
1 hour postinfusion. POC BG monitoring can be decreased
or discontinued at the discretion of the clinician once a
patient is deemed stable at full calorie PN, with or without
cycling, but this is not consistently practiced. As such, there
may be opportunities to reduce BG testing in stable patients
without affecting patient safety or quality of care.
The purpose of this study was to determine the effects of
implementing guidelines aimed at decreasing the frequency
of POC glucose testing in stable patients who are receiving
PN on frequency of adverse glycemic events and healthcare
costs.
Methods
Study Design
This quality improvement project examined hospitalized
adult, noncritically ill patients admitted to a quaternary care
1200-bed urban hospital, who received PN, defined as an
intravenous (IV) formula containing protein and dextrose,
initiated and managed by the NST inpatient service. The
institutional review board determined the study protocol
to be exempt. Patients were excluded from the study if
they were followed by the endocrinology service for BG
management, had PN initiated in an ICU, or received PN
873
for <5 days. Eligible patients were identified from daily work
lists.
Two study groups were defined: preintervention and
postintervention or preguideline and postguideline
implementation. Data for each patient were collected
retrospectively from November 2014–May 2015 during the
preintervention period and August–October 2015 during
the postintervention period. Data were collected from
initiation of PN for a minimum of 5 days to a maximum
of 10 days or until PN was discontinued for any reason (eg,
transitioned to oral or enteral nutrition, discharged, patient
refused, withdrawal of care), PN was held for >24 hours
for any reason (eg, IV access problems, surgery, orders not
written), patient was transferred to an ICU, or the patient
was discharged on home PN.
The intervention consisted of guidelines for the initi-
ation and discontinuation of POC BG monitoring (Ta-
ble 1). POC BGs were obtained using glucometers that
were calibrated monthly to ensure accuracy. A BG <70
Table 1. Guidelines for Point-of-Care Blood Glucose
Monitoring.
1. Starting POC testing
a. New central PN
i. Continuous infusion: order POC glucose
monitoring every 6 hours at 6:00 AM,
12:00 PM, 6:00 PM, and 12:00 AM
ii. Cyclic infusion: order POC glucose
monitoring 3 times daily 2 hours after
beginning cycle, midcycle, and 1 hour after
completing cycle
b. New peripheral PN: POC blood glucose
monitoring not required unless patient has
insulin-dependent diabetes and/or evidence of
recent hyperglycemia (ie, blood glucose
>200 mg/dL)
c. Existing home PN: POC blood glucose
monitoring is not required if patient’s home PN
formula does not contain insulin and there is no
evidence of recent hyperglycemia (ie, blood
glucose >200 mg/dL)
2. Discontinuing POC testing
a. Discontinue POC blood glucose monitoring
when blood glucose levels <200 mg/dL for >24
hours and PN at goal calories (continuous
infusion) or goal calories and cycle hours (cyclic
infusion)
b. Discontinue POC blood glucose test if ordered
by another service if patient does not meet
criteria for POC blood glucose monitoring
3. Restarting POC testing
a. Start/restart POC blood glucose monitoring
every 6 hours for any patient with BMP blood
glucose <70 or >200 mg/dL
BMP, basic metabolic panel; PN, parenteral nutrition; POC, point of
care.
874
mg/dL was defined as hypoglycemia and >200 mg/dL as
hyperglycemia. The guidelines were based on a review
of published standards along with expert opinion. The
guidelines were implemented after the preintervention data
were collected, and a training program targeting NST
registered dietitians, staff physicians, and nurse practition-
ers/physician assistants was completed. Postintervention
data were not collected until 2 weeks after the new guidelines
were implemented to allow the staff to become accustomed
to the new practice. Nurse managers were informed about
the new guidelines and educated their staff 1 time at their
weekly meeting. Continuing education was provided by
NST clinicians to bedside nursing when requested. Per
guidelines, NST clinicians had the discretion to discontinue,
reduce, start, restart, or otherwise alter BG monitoring
based on clinical judgment. The NST clinicians received
weekly continued education of the new guidelines for
1 month.
Data Collection
Data were collected retrospectively from the electronic med-
ical record. Daily BMPs and POC BG tests were obtained
during the specified time range. Preintervention and postin-
tervention data were collected and analyzed separately.
Morning BMP glucose levels were used as standard for
benchmarking BG levels for each patient. Morning BMP
glucose levels were used to monitor glucose levels, given
each patient who was receiving PN had these drawn daily
even when POC glucose tests were discontinued. Other
data collected included PN indication (eg, ileus/obstruction,
leak/high-output enterocutaneous fistula, malabsorption,
other), PN route (ie, central, peripheral) at end of data
collection for each patient, PN infusion schedule (ie, con-
tinuous, cyclic) at end of study period, demographics (eg,
age, gender), diet order (eg, NPO, clear liquid/full liquid,
other) at end of study period, enteral tube feeding (eg, yes,
no) at end of study period, diseases/conditions affecting BG
control such as history of type 1 or 2 diabetes mellitus re-
quiring medication before admission, actively being treated
for infection (ie, current medications including at least 1 an-
tibiotic), medications potentially contributing to BG abnor-
malities (ie, steroids, thiazide diuretics, fluoroquinolones,
β-blockers, hormones, anticonvulsants, antipsychotics, and
miscellaneous medications such as octreotide, isoniazid, and
clonidine), and nutrition status (eg, no malnutrition, mild
malnutrition, moderate malnutrition, severe protein-calorie
malnutrition).
PN was individualized for each patient based on clinician
judgments using ASPEN guidelines. Patients were typically
started on 150–200 g dextrose on day 1 of PN adminis-
tration and advanced to goal over the next 3 days. Protein
was generally dosed at 1.5–2.0 g/kg/d and fats at 250 mL
of 20% lipid injectable emulsion 3 times per week. Goal
Nutrition in Clinical Practice 33(6)
calories were generally 25–35 kcal/kg/d but were adjusted
for underweight or obese patients.
Statistical Analysis
Pilot data were collected to determine sample size. A total
of 86 subjects per period was needed to provide 95% power.
Univariate analysis was performed to compare patient char-
acteristics between the 2 time periods. Analysis of variance
or the nonparametric Kruskal-Wallis test was used for
continuous or ordinal variables. The Pearson’s χ 2 tests or
Fisher’s exact tests were used for categorical factors.
In addition, an equivalence test was used to assess
whether the average BMP glucose and the observed
proportion of subjects with adverse glycemic events during
preintervention and postintervention were equivalent.
Equivalency testing was used to determine whether the
preintervention and postintervention groups were similar.
Equivalency test uses a null hypothesis that assumes the
differences between the 2 groups are greater than the
noted amount, also known as the “interval of tolerable
differences.” This threshold represents a difference that
has clinically insignificant outcomes. For this study, this
difference was noted as postintervention was equivalent
to preintervention if there was a difference no >15 mg/dL
for average glucose level, as determined by the average
variability in BG samples, 5% for low glucose (<70) or 10%
for high glucose (>200). In equivalency testing, P < .05
means the 2 groups are equivalent, not different.
Also, the adverse glycemic event rate per 100 BMP tests
for each time period was calculated by dividing the total of
observed events by the total number of tests performed; 2
1-sided tests were used to assess equivalence between the
groups.
All analyses were performed using SAS version 9.4
(SAS Institute, Cary, NC), and P < .05 was considered
statistically significant. Data are presented as median or
mean ± standard deviation.
Results
Data were collected on 171 subjects (86 preintervention,
85 postintervention). Average age was 56 ± 17 years, and
53% were male. Table 2 presents a summary of patient
demographics and clinical characteristics for each time
period. There were no significant differences in PN route,
PN infusion schedule, history of diabetes, antibiotic ad-
ministration, or administration of medications that could
affect BG between the preintervention and postintervention
groups.
Table 3 demonstrates the total number of POC tests
per patient was significantly fewer in the postintervention
group (median 23.0 vs 25.5; P = .011). The average number
of POC tests per patient per day was significantly fewer
in the postintervention than the preintervention group
Ratliff et al 875

Table 2. Demographic and Clinical Characteristics of Patients Who Are Receiving Parenteral Nutrition.

Preintervention (n = 86) Postintervention (n = 85)

Factors n (%) n(%) P Value

Age, y 54.8 ± 16.3 56.5 ± 16.9 .52a

Gender, n (%) .49b
Male 43 (50.0) 47 (55.3)
Female 43 (50.0) 38 (44.7)
Nutrition status, n (%) .77b
None 23 (26.7) 20 (23.5)
Mild 10 (11.6) 13 (15.3)
Moderate 12 (14.0) 15 (17.6)
Severe 41 (47.7) 37 (43.5)
PN indication, n (%) .026b,c
Ileus/obstruction 35 (40.7) 41 (48.2)
Leak/ECF 13 (15.1) 23 (27.1)
Malabsorption 16 (18.6) 12 (14.1)
Other 22 (25.6) 9 (10.6)
PN route, n (%) .95b
Central 81 (94.2) 78 (94.0)
Peripheral 5 (5.8) 5 (6.0)
PN infusion schedule, n (%) .79b
Continuous 34 (39.5) 32 (41.6)
Cyclic 52 (60.5) 45 (58.4)
Diet, n (%) .24b
NPO/Clear liquid 38 (44.2) 26 (35.1)
Full liquid/other 48 (55.8) 48 (64.9)
History of diabetes mellitus, n (%) 16 (18.6) 17 (20.0) .82b
Antibiotics when started, n (%) 38 (44.2) 31 (36.5) .30b
Glucose-increasing medications, n (%) 25 (29.1) 26 (30.6) .83b
Median LOS (Q1, Q3), d 8.0 (5.0, 9.0) 9.0 (6.0, 9.0) .80d

ECF, enterocutaneous fistula; LOS, length of stay; PN, parenteral nutrition; Q1, 25th percentile; Q3,75th percentile.
a Analysis of variance.
b Pearson’s χ 2 test.
c Represents those P values <.05 and signifies the 2 groups are significantly different.
d Kruskal-Wallis test.

(2.9 ± 1.4 vs 3.6 ± 1.4; P < .001). The number of POC tests
per patient per day was similar between the 2 periods until
day 3. Afterward, the number of POC tests was lower in
the postintervention period, and this difference continued
until the end of the study. There was no evidence of a
significant difference in average POC glucose levels between
the 2 periods (135 ± 35.2 vs 135 ± 30.9; P = .94) (Table 3).
Table 4 presents the results of the equivalence tests.
The average BMP glucose was equivalent between the 2
time periods (ie, 126 mg/dL preintervention vs 129 mg/dL
postintervention; P = .005). Similarly, the number of hy-
poglycemic and hyperglycemic events per 100 tests was
equivalent (P < .001).
Discussion
This project provides evidence that at a large medical center,
with a dedicated NST managing PN daily, it is safe to
discontinue POC BG monitoring once patients are stable,
or forgo testing entirely in patients who are at low risk for
development of hyperglycemia.
The patients were similar in both the preintervention
and postintervention groups in regard to all demographics
except classification of PN indication, which was signifi-
cantly different (P = .026). This could have been caused by
differences in clinical judgment due to separate investigators
collecting data for preintervention and postintervention
periods.
POC BG monitoring is the cornerstone of glycemic
management in hospitalized patients. This is especially true
for patients who are receiving PN because hyperglycemia is
a common metabolic complication. There are guidelines for
the initiation of POC BG monitoring in patients who are
receiving PN,11 but discontinuing testing is not addressed.
Reducing the frequency of diagnostic tests may improve
the value of healthcare providing the quality of patient
care can be maintained. In particular, laboratory tests
are scrutinized because they are routinely performed in
876 Nutrition in Clinical Practice 33(6)

Table 3. Frequency and Average Blood Glucose Testing.

Preintervention Postintervention
Factors (n = 86) (n = 85) P Value

Total no. of BMP tests, median (min, max) 86 9 (3, 12) 85 9 (5, 11) .84a
Total no. of POC tests, median (min, max) 86 25 (0, 59) 85 23 (0, 47) .011a,b
Average no. of POC tests/day, mean ± SD 86 3.6 ± 1.4 85 2.9 ± 1.4 <.001b,c
Overall average POC, mean ± SD 83 135 ± 35 81 136 ± 31 .94c
Overall minimum POC, mean ± SD 83 84 ± 18 81 95 ± 23 <.001b,c
Overall maximum POC, mean ± SD 83 203 ± 69 81 194 ± 67 .38c

BMP, basic metabolic panel; max, maximum; min, minimum; POC, point of care.
a Kruskal-Wallis test.
b Represents those P values <.05 and signifies the 2 groups are significantly different.
c Analysis of variance.

Table 4. Effect of Point-of-Care Glucose Monitoring on Glycemic Events: Equivalence Testing.

Preintervention Postintervention Equivalence

Factors (n = 86) (n = 85) Test P Value

Average BMP glucose (mg/dL), mean ± SD 86 126.4 ± 30.9 85 128.8 ± 32.6 .005
Hypoglycemia on BMP
Total no. of events, median (min, max) 86 0 (0, 1) 85 0 (0, 3) —
Any event, n (%) 86 3 (3.5) 85 4 (4.7) .1
No. of events/100 tests, rate (95% CI) 86 0.43 (0.14, 1.3) 85 0.99 (0.47, 2.1) <.001a
Hyperglycemia on BMP
Total no. of events, median (min, max) 86 0 (0, 6) 85 0 (0, 6) —
Any event, n (%) 86 18 (20.9) 85 21 (24.7) .16
No. of events/100 tests, rate (95% CI) 86 7.0 (5.3, 9.2) 85 7.5 (5.8, 9.9) <.001a

Two 1-sided tests were used to assess equivalence. Margin of equivalence was defined as 5% for hypoglycemia and 10% for hyperglycemia.
BMP, basic metabolic panel.
a Represents those P values <.05 and signifies the 2 groups are equivalent.

hospitalized patients, many times without a clear indication.
Efforts to reduce the number of unnecessary laboratory tests
have involved a variety of approaches with varying degrees
of success ranging from a 98% decrease to a 28% increase in
testing.12
Reducing laboratory tests is meaningful only if clinical
outcomes are maintained or improved. If there is an increase
in adverse events attributable to a reduction in testing,
any potential cost savings may be negated. Unfortunately,
the effect of reducing the frequency of laboratory test-
ing on clinical outcomes, such as rate of complications
or hospital length of stay, is difficult to determine due
to the number of confounding factors. Most studies in
this area assumed clinical equivalence rather than tracking
outcomes.13,14
Results of this quality improvement project demonstrate
the frequency of POC BG monitoring can be reduced
without affecting the average BG level or the number of
episodes of hypoglycemia or hyperglycemia in hospitalized
patients who are receiving PN. Adverse glycemic events
were defined as blood sugars <70 or >200 mg/dL be-
cause readings at those levels usually require intervention,
typically oral or IV carbohydrate for hypoglycemia, and
subcutaneous sliding scale insulin for hyperglycemia, even in
the absence of symptoms. Although serum BG levels are not
a clinical outcome, they were used as a surrogate measure
of blood sugar management because they are considered
more precise than capillary BG levels and were available
daily throughout the study period.
The average number of POC BG tests was reduced ap-
proximately 20% in the current project. Potential benefits of
reducing the frequency of POC BG tests include decreases
in costs and patient discomfort. This can benefit the patient,
the healthcare team, and the institution. Reducing POC BG
tests can decrease disruption of sleep cycles and patient
discomfort from the finger sticks. With the time savings re-
alized from performing fewer POC tests, healthcare workers
can engage in other activities to benefit patients, and the cost
savings from decreased testing can be passed along to the
patient and institution.
Although a single POC test is relatively inexpensive, it
is typically ordered every 6 hours for patients who are
Ratliff et al
receiving PN compared with once daily for most laboratory
tests. Hospital charges are proprietary, making it difficult to
calculate actual savings due to decreased POC testing. For
this project, we estimated the cost of a single POC BG test
at $6.00. Based on the patient volume of the NST inpatient
service at our institution, a 20% reduction in POC BG tests
would equal 7700 fewer tests per year and an annual savings
of $46,000.
The savings in labor due to nursing assistants performing
fewer tests is less clear. In theory, if nursing assistants
perform fewer unnecessary POC tests, they would have more
time to perform other, more vital patient care activities. This
would likely result in improved patient care, but overall labor
costs would be unchanged.
A limitation of the project was the variation among clin-
icians in application of the new POC BG testing guidelines.
Because guidelines are simply suggestions for practice rather
than policies, they allowed clinicians to vary treatment
based on their clinical judgment, and although the NST is
very experienced overall, this was a new practice for most
of the clinicians, so not everyone had the same level of
comfort reducing the number of POC tests. It is expected,
with time, that clinicians will become more accustomed
to the new guidelines and follow them more uniformly.
This lack of comfort with reducing POC tests extended to
the patients’ primary service. Because the primary service
is primarily responsible for submitting orders for their
patients, they reordered POC BG monitoring multiple times
in study patients in good faith, thinking it was inadvertently
discontinued.
Another limitation of the project was the lack of a
measure of patient satisfaction specifically related to the re-
duction in POC BG tests. We assumed reducing POC testing
would improve satisfaction because of fewer interruptions
in the sleep cycle, particularly at midnight and 6:00 AM,
when testing is typically done, and less discomfort because
of the fewer finger sticks. Unfortunately, POC testing is
only one of many factors that comprise patient satisfaction
scores, and it was beyond the scope of this project to tease
out these data.
Lastly, a potential weakness of this project is the number
of patients. The goal was to enroll 86 patients in each group
based on the expectation there would be a 25% decrease
in the frequency of POC testing. The actual decrease
was closer to 20%, and 1 patient in the postintervention
group was an outlier because she had repeated episodes of
hypoglycemia and hyperglycemia. In spite of this, the data
demonstrated equivalence between the groups in terms of
the number of adverse events.
A strength of this project is that it was conducted
at a single institution by members of a single team of
experienced clinicians who are primarily responsible for
managing blood sugars in patients who are receiving PN.
877
This simplified training of the clinicians increased the
likelihood that guidelines were followed.
Another strength of this project was the availability of
daily serum BG measurements. Serum BG was considered
the gold standard to determine overall blood sugar man-
agement for this project compared with POC BG tests.
Although daily serum BG measurements were probably un-
necessary, at least in a portion of the patients, it gave us the
opportunity to serially monitor blood sugar management in
all patients without gaps in the data.
Conclusions
This project demonstrated that the implementation of new
guidelines reduced the frequency of POC BG tests by 20%
in hospitalized patients who were receiving PN without
increasing average BG level or adverse glycemic events. This
was achieved by discontinuing monitoring in patients once
they were stable at goal calories and infusion schedule,
and forgoing testing in patients who were at low risk for
development of hyperglycemia. Although these interven-
tions were successful on a small scale with a limited group
of clinicians for a limited time in patients receiving PN,
it is not clear whether they would be as successful on a
larger scale over a larger period with patients who are not
receiving PN. In addition, the subjects were on regular
nursing floors, so it is not clear whether the same guidelines
can be used in ICUs. Regardless, reducing unnecessary
POC BG tests should be a goal for all clinicians who
are managing patients receiving PN and warrants fur-
ther investigation because it can potentially reduce health-
care costs and patient discomfort, and improve healthcare
value.
Statement of Authorship
A. Ratliff, B. DeChicco, A. Nishnick contributed to concep-
tion/design of the research, acquisition, R. Lopez contributed
to the analysis or the data; All authors contributed to the
interpretation of the data, drafted the manuscript, A. Ratliff,
B. DeChicco, A. Nishnick critically revised the manuscript;
and All authors agree to be fully accountable for ensuring
the integrity and accuracy of the work. All authors read and
approved the final manuscript.
References
1. Pleva M, Mirtallo JM, Steinberg SM. Hyperglycemic events in non-
intensive care unit patients receiving parenteral nutrition. Nutr Clin
Pract. 2009;24:626-634.
2. Sarkisian S, Fenton TR, Shaheen AA, Raman M. Parenteral nutrition-
associated hyperglycemia in noncritically ill inpatients is associated
with higher mortality. Can J Gastroenterol. 2010;24:453-457.
3. Kumar PR, Crotty P, Raman M. Hyperglycemia in hospitalized
patients receiving parenteral nutrition is associated with
increased morbidity and mortality: a review. Gastroenterol Res
Pract. 2011;2011:760720.
878
4. Pasquel FJ, Spiegelman R, McCauley M, et al. Hyperglycemia during
total parenteral nutrition. Diabetes Care. 2010;33:739-741.
5. Oliveira G, Tapia MJ, Ocon J, et al. Parenteral nutrition-associated
hyperglycemia in non-critically ill patients increases the risk of in-
hospital mortality (multicenter study). Diabetes Care. 2013;36:1061-
1066.
6. Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin
therapy in critically ill patients. N Engl J Med. 2001;345:1359-1367.
7. Van den Berghe G, Wilmer A, Hermans G, et al. Intensive insulin
therapy in the medical ICU. N Engl J Med. 2006;354:449-461.
8. McMahon MM, Nystrom R, Braunschweig C, Miles J, Compher
C; American Society for Parenteral and Enteral Nutrition Board of
Directors. A.S.P.E.N. clinical guidelines: nutrition support of adult
patients with hyperglycemia. JPEN J Parenter Enteral Nutr. 2013;37:
23-36.
9. Lin LY, Lin HC, Lee PC, Ma WY, Lin HD. Hyperglycemia correlates
with outcomes in patients receiving total parenteral nutrition. Am J
Med Sci. 2007;333:261-265.
Nutrition in Clinical Practice 33(6)
10. Cheung NW, Zaccaria C, Napier B, Fletcher JP. Hyperglycemia is
associated with adverse outcomes in patients receiving total parenteral
nutrition. Diabetes Care. 2005;28:2367-2371.
11. Sacks GS, Mayhew S, Peterson C. Parenteral nutrition implementation
and management. In: Merrit R, ed. ASPEN Nutrition Support Practice
Manual. 2nd ed. Silver Spring, MD: American Society for Parenteral
and Enteral Nutrition; 2009:114-115.
12. Kobewka D, Ronksley P, McKay J, Forster A, van Walraven
C. Influence of educational, audit and feedback, system based,
and incentive and penalty interventions to reduce laboratory test
utilization: a systematic review. Clin Chem Lab Med. 2015;53:
157-183.
13. Feldman LS, Shihab HM, Thiemann D, et al. Impact of providing
fee data on laboratory test ordering. JAMA Intern Med. 2013;173:
903-908.
14. Thakkar R, Kim D, Knight AM, et al. Impact of an educational
intervention on the frequency of daily blood test orders for hospitalized
patients. Am J Clin Pathol. 2015;143:393-397.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Risk of Malnutrition Evaluated by Mini Nutritional December 2018 879–886

C 2018 American Society for

Assessment and Sarcopenia in Noninstitutionalized Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10022
Elderly People wileyonlinelibrary.com

Ilaria Liguori, MD1 ; Francesco Curcio, MD1 ; Gennaro Russo, MD1 ;

Michele Cellurale, MD1 ; Luisa Aran, MD1 ; Giulia Bulli, MD1 ;
David Della-Morte, MD, PhD2,3 ; Gaetano Gargiulo, MD4 ; Gianluca Testa, MD,
PhD1,5 ; Francesco Cacciatore, MD, PhD1,6 ; Domenico Bonaduce, MD1 ;
and Pasquale Abete, MD, PhD1

Abstract
Background: Malnutrition indices and muscle mass and strength in the elderly are poorly investigated. Moreover, malnutrition seems
to be 1 of the more important factors in the cause of sarcopenia. The presence of sarcopenia and its relationship with malnutrition
indices were studied in noninstitutionalized elderly people who underwent Comprehensive Geriatric Assessment (CGA). Methods:
A total of 473 elderly subjects (mean age, 80.9 ± 6.6 years) admitted to CGA were studied. Malnutrition risk was evaluated with
Mini Nutritional Assessment (MNA) score, whereas muscle mass and muscle strength were evaluated by bioimpedentiometry and
hand grip, respectively. Sarcopenia was assessed as indicated in the European Working Group on Sarcopenia in Older People
(EWGSOP) consensus. Results: Overall prevalence of sarcopenia was 13.1%, and it increased from 6.1% to 31.4% as MNA
decreased (P < .001). MNA score was lower in elderly subjects with sarcopenia (15.4 ± 4.2) than without sarcopenia (22.0 ±
4.0) (P = .024). Linear regression analysis showed that MNA score is linearly related both with muscle mass (r = 0.72; P < .001)
and strength (r = 0.42; P < .001). Multivariate analysis, adjusted for several confounding variables including comorbidity and
disability, confirmed these results. Conclusions: MNA score is low in noninstitutionalized elderly subjects with sarcopenia, and it
is linearly related to muscle mass and muscle strength. These data indicate that MNA score, when evaluated with muscle mass and
strength, may recognize elderly subjects with sarcopenia. (Nutr Clin Pract. 2018;33:879–886)

Keywords
aged; elderly; malnutrition; muscle strength; nutrition assessment; nutritional status; sarcopenia

Sarcopenia has been defined as a phenomenon character-
ized by an age-related loss of muscle mass and strength
particularly evident in older people aged >75 years.1-3
Sarcopenia is related to an increased risk for morbidity
and mortality in elderly people,4-6 especially in institution-
alized elderly nursing home residents.7 This pathological
condition is defined as a geriatric syndrome characterized
by a loss of independence, a higher risk for falls, a reduc-
tion of the quality of life, and an increase in healthcare
cost.1-3 From the pathophysiological point of view, several
factors may be involved in the pathogenesis of sarcope-
nia including hormonal, metabolic, inflammatory state,
physical inactivity, and more importantly, malnutrition
state.8
Muscle protein loss at advancing age depends on a
discrepancy between muscle protein synthesis and muscle
protein breakdown; in fact, muscle protein synthesis rates
are reduced in the older population with age-related decline
at a rate of 3.5% per decade from age 20–90 years.9-11 It
is well known that protein-energy-malnutrition prevalence
in the acute care setting ranges between 23% and 60% of
elderly patients, and 22%–28% in this setting are at nutrition
risk.12 However, data on the prevalence of malnutrition in
From the 1 Department of Translational Medical Sciences, University
of Naples “Federico II”, Naples, Italy; 2 Department of Systems
Medicine, University of Rome Tor Vergata, Rome, Italy; 3 IRCCS San
Raffaele Pisana, Rome, Italy; 4 Division of Internal Medicine, AOU
San Giovanni di Dio e Ruggi di Aragona, Salerno, Italy; 5 Department
of Medicine and Health Sciences, University of Molise, Campobasso,
Italy; and 6 Azienda Ospedaliera dei Colli, Monaldi Hospital, Heart
Transplantation Unit, Naples, Italy.
Financial disclosure: None declared.
Conflicts of interest: None declared.
[This article was modified on 2018-06-05 after initial online
publication to correct reference 17.]
This article originally appeared online on February 13, 2018.
Corresponding Author:
Pasquale Abete, MD, PhD, Associate Professor of Geriatrics,
Dipartimento di Scienze Mediche Traslazionali, Università di Napoli
Federico II, Via S. Pansini, 80131 Naples, Italy.
Email: p.abete@unina.it
880
community-dwelling older adults are poor, although some
reports indicate a range of 5%–30%.12
Moreover, malnutrition and sarcopenia often coexist in
a clinical syndrome characterized by a combination of
decreased nutrient intake and decreased body weight, along
with a decrease in muscle mass, strength, and/or physical
function (ie, malnutrition-sarcopenia syndrome [MSS]).13
Malnutrition and sarcopenia, taken together, are associated
with negative health outcomes. Thus, it is extremely impor-
tant to assess malnutrition status, and more importantly,
nutrition assessment should be integrated with sarcopenia
screening.13
Several malnutrition tools have been used with different
results.14 Mini Nutritional Assessment (MNA) represents
1 of the more specific tool to assess malnutrition state in
geriatric settings.15 However, this approach, together with
sarcopenia screening, is not well investigated in outpatient
older adults.
Thus, in this study, we aimed to investigate the rela-
tionship between muscle mass and strength and malnutri-
tion evaluated by MNA in outpatients older adults who
underwent Comprehensive Geriatric Assessment (CGA),
including several clinical and biochemical parameters.
Methods
Study Population
The study enrolled 473 consecutive elderly outpatients (ࣙ65
years) admitted to a CGA center. The study received
full ethical approval from research ethics committees in
accordance with the ethical standards laid down in the
1964 Declaration of Helsinki and its later amendments. All
participants signed an informed consent form, and the insti-
tutional review boards approved the study. Anthropometric
measurements including age, sex, body mass index (BMI),
and waist circumference (WC) were performed.16,17
Comprehensive Geriatric Assessment
Elderly participants underwent a comprehensive geri-
atric multidimensional evaluation that included cognitive
function evaluation with Mini-Mental State Examination
(MMSE)18 ; depressive symptoms with Geriatric Depression
Scale (GDS)19 ; comorbidity and comorbidity severity with
a Cumulative Illness Rating Scale (CIRS-Comorbidity and
CIRS-Severity)20 and drug number; disability with basic
and instrumental activities of daily living (BADLs and
IADLs, respectively); physical performance with 4-m gait
speed (cutoff < 0.8 m/s)21 ; physical activity with Physical
Activity Scale for the Elderly (PASE)22 ; social support eval-
uation with Social Support Assessment (SSA) scored from
17 (participants with the lowest support) to 0 (participants
with the highest support)23 and Tinetti Mobility Test or
Performance-Oriented Mobility Assessment (TMT).
Nutrition in Clinical Practice 33(6)
Frailty status was assessed by the Italian Frailty Index
derived by Rockwood’s frailty methods, recently validated
in an Italian cohort of elderly subjects.24 Fried’s frailty
methods were also assessed according to the Cardiovascular
Health Study Collaborative Research Group.25
Assessment of Nutrition Status
Nutrition status was assessed by MNA. The MNA test
is composed of simple measurements and brief questions
that can be completed in <10 minutes: (1) anthropometric
measurements (weight, height, and weight loss); (2) global
assessment (6 questions related to lifestyle, medication, and
mobility); (3) dietary questionnaire (8 questions, related to
number of meals, food and fluid intake, and autonomy of
feeding); and (4) subjective assessment (self-perception of
health and nutrition). The sum of the MNA score distin-
guishes between elderly patients: (1) with adequate nutrition
status (MNA ࣙ 24), (2) with protein-calorie malnutrition
(MNA >17), and (3) at risk for malnutrition (MNA between
17 and 23.5). With this scoring, sensitivity was found to be
96%, specificity 98%, and predictive value 97%.26
Assessment of Muscle Mass and Strength
Muscle mass was measured by bioelectrical impedance
analysis (BIA) using a Quantum/S Bioelectrical Body Com-
position Analyzer (Akern Srl, Florence, Italy). Whole-body
BIA measurements were taken between the right wrist and
ankle with the subject in a supine position. Muscle mass
was calculated using the following BIA equation of Janssen
and colleagues.27 Skeletal muscle mass (kg) = ([height2 /BIA
resistance × 0.401] + [gender × 3.825] + [age × −0.71])
+ 5.102, where height is measured in centimeters; BIA
resistance is measured in ohms; for gender, men = 1 and
women = 0; and age is measured in years. Absolute skeletal
muscle mass (kg) was converted to skeletal muscle index
standardizing by meters squared (kg/m2 ). Muscle strength
was assessed by grip strength, measured using a hand-grip
dynamometer (Mecmesin Advanced Force Gauge 500N;
GDM, Corsico (MI), Italy). The participant squeezed the
dynamometer with maximum effort for at least 5 seconds.
Hand-grip strength was tested 3 times with a minimum of
60 seconds rest provided for recovery between each attempt.
Maximum grip strength was recorded as the highest value of
3 trials. The maximum value was used rather than the mean
of 3 trials allowing for a measure that was independent
of fatigue, which may have been possible in some of the
participants. BMI-adjusted values were used as a cutoff
point to classify low muscle strength (BMI ࣘ24, 24.1–28,
<28 was 29, ࣘ30, and ࣘ32 kg for men; BMI ࣘ23, 23.1–
26, 26.1–29, and <29 was 17, ࣘ17.3, ࣘ18, and ࣘ21 kg for
women, respectively).25 Using the cutoff points indicated
in the European Working Group on Sarcopenia in Older
Liguori et al
People (EWGSOP) consensus, subjects with a low grip
strength presenting low muscle mass (skeletal muscle index
<8.87 and 6.42 kg/m2 in men and women, respectively) were
classified as “sarcopenia.”28
Biochemical Parameters
Hemoglobin (normal values: 12.0–17.5 g/dL in men and
12.0–16.0 g/dL in women), lymphocytes (normal values:
1.0–4.8 × 103 ), serum albumin level (normal values: 3.6–
5.2 g/dL), total iron binding capacity (normal values: 218–
411 μg/dL), C-reactive protein (CRP; normal values: <5
mg/dL), and butyryl-cholinesterase (b-CHE; normal values:
5.400–13.200 U/L) were routinely obtained.
Statistical Analysis
Baseline characteristics of the sample are expressed as
mean ± standard deviation. Participants were stratified
by the presence or absence of sarcopenia and stratified
by categories of MNA (<17, 17–23.5, ࣙ24). Categorical
variables were analyzed using χ 2 testing, continuous vari-
ables using a 1-way analysis of variance, and nonparametric
variables (ie, PASE score and lymphocyte number) using
Wilcoxon-Mann-Whitney. Univariate regression analysis
was used to test a correlation among muscle mass and
strength, and MNA score with other variables such as
age, BMI, WC, MMSE, GDS, CIRS-Comorbidity, CIRS-
Severity, drug number, BADLs, IADLs, PASE, 4-m walking
speed, TMT, and social support score. Statistically signifi-
cant variables were included into a multivariate regression
model as potential confounders. All statistical analyses
were performed with SPSS software (version 15.0; SPSS,
Chicago, IL). A P value <.05 was considered statistically
significant.
Results
A total of 473 elderly subjects with a mean age of 80.9
± 6.6 years was studied. Table 1 lists the anthropometric
measurements, CGA, and biochemical assessment stratified
in sarcopenia and nonsarcopenia according to EWGSOP
definition and diagnostic algorithm. The prevalence rate
of sarcopenia was 13.1%. CIRS comorbidity and severity
score, drug number, and CRP were higher, whereas 4-minute
gait speed, Tinetti score, and b-CHE were lower in elderly
subjects with sarcopenia. In addition, MNA score together
with muscle mass and grip strength were lower in elderly
subjects with sarcopenia with respect to elderly subjects
without sarcopenia (Table 1).
Table 2 lists the same parameters stratified for different
levels of MNA score (ࣙ24, 17–23.5, and <17). Prevalence
of sarcopenia progressively increased as MNA decreased
(from 6.1% to 31.4%; P < .001). Elderly subjects were
progressively older, and MMSE and Tinetti score and
881
4-m gait speed decreased as MNA score decreased. GDS
score, BADLs and IADLs lost, and low social support
increase as MNA score decreased. Similarly, both frailty
by Fried’s score and Rockwood’s score increased as MNA
score decreased. Among biomarkers, lymphocytes and b-
CHE decreased, whereas CRP increased as MNA score
decreased.
Figure 1 shows the relationship among skeletal muscle
mass and strength stratified by MNA score in nonin-
stitutionalized elderly people. Muscle mass progressively
decreased (from 17.6 ± 3.0 to 10.0 ± 2.9 kg/m2 ), as well as
muscle strength (from 41.2 ± 10.2 to 27.7 ± 8.3 kg), as MNA
score decreased.
At univariate analysis, age, CIRS-severity score, number
of drugs, BADLs lost, and CRP were negatively correlated
with skeletal muscle mass, whereas Tinetti score, 4-m gait
speed, PASE score, serum albumin level, hemoglobin, b-
CHE, and, more importantly, MNA score were positively
correlated with skeletal muscle mass. Multivariate analysis
confirms this correlation except for serum albumin level and
CRP (Table 3). The positive linear relation between skeletal
muscle mass and MNA score is shown in Figure 2A (y =
0.65x + 2.2; r = 0.72; P < .001).
Univariate and multivariate linear regression analysis on
muscle strength is shown in Table 3. Age, CIRS comorbidity
and severity, number of drugs, BADLs lost, and CRP were
negatively correlated with muscle strength, whereas Tinetti
score, 4-m gait speed, PASE score, hemoglobin, b-CHE, and
MNA score were positively correlated with muscle strength.
Except for CRP, all linear relationships were confirmed at
multivariate analysis. The positive linear relation between
muscle strength and MNA score is shown in Figure 2B
(y = 1.1x + 14.2; r = 0.46; P < .001).
Discussion
Our data show that MNA score, 1 of the most specific
tools for the assessment of malnutrition, is lower in non-
institutionalized older adults with sarcopenia than in those
without sarcopenia. Interestingly, MNA score is strongly
related to muscle mass and strength both at univariate and
multivariate analyses. This evidence suggests that malnutri-
tion state is frequently associated with a reduction of muscle
mass and strength; therefore, it should be investigated,
especially in the presence of sarcopenia in noninstitution-
alized older adults.
Sarcopenia
Age-related muscle mass decline is defined “sarcopenia,”
and it is primarily due to the progressive atrophy and
loss of type II muscle fibers and motor neurons.29 This
phenomenon tends to be reduced at a rate of 1%–2% per
year after the age of 50 years.1-3 Sarcopenia is characterized
by fibrosis and infiltration of adipose tissue determining a
882 Nutrition in Clinical Practice 33(6)

Table 1. Baseline Characteristics of the 473 Elderly Patients Enrolled in the Study Stratified for the Presence and Absence of
Sarcopenia.

Sarcopeniaa

All No (86.9%) Yes (13.1%)

Characteristics (N = 473) (n = 411) (n = 62) Pb

Anthropometric data
Age ± SD, y 80.9 ± 6.6 74.3 ± 8.1 80.5 ± 7.6 .145
Female sex, % 60.9 64.5 37.1 .057
BMI ± SD, kg/m2 27.4 ± 5.6 27.2 ± 3.1 24.2 ± 4.0 .064
Waist circumference ± SD, cm 99.7 ± 13.5 99.0 ± 10.8 96.4 ± 11.4 .007
Geriatric evaluation ± SD
Mini-Mental State Examination score 21.5 ± 6.3 23.2 ± 6.4 21.6 ± 4.4 .059
Geriatric Depression Scale score 7.6 ± 4.4 6.7 ± 4.6 8.6 ± 5.4 .187
CIRS-Comorbidity score 3.9 ± 2.2 3.40 ± 1.79 5.12 ± 2.05 .024
CIRS-Severity score 1.9 ± 0.5 1.60 ± 0.22 2.34 ± 0.55 .030
No. of drugs 6.1 ± 3.2 2.0 ± 1.0 4.0 ± 1.0 .020
BADLs lost, n 2.0 ± 1.8 1.4 ± 1.6 2.6 ± 1.8 .212
IADLs lost, n 4.2 ± 2.8 2.8 ± 1.4 4.0 ± 3.2 .321
Tinetti score 17.3 ± 7.8 22.4 ± 4.0 18.2 ± 2.2 .020
MNA score 20.9 ± 4.3 22.0 ± 4.0 15.4 ± 4.2 .024
4-m gait speed, m/s 0.46 ± 0.4 0.91 ± 0.20 0.60 ± 0.30 <.05
PASE score 37.6 ± 51.1 102.0 ± 40.8 38.4 ± 80.2 <.001
Social support score 8.3 ± 4.9 7.0 ± 2.0 9.5 ± 2.0 <.05
Fried’s frailty score 3.5 ± 1.5 2.2 ± 1.2 3.6 ± 1.2 <.05
Rockwood’s frailty score 21.2 ± 8.7 14.6 ± 9.4 28.2 ± 6.2 <.05
Muscle measurements
Grip strength, kg 26.2 ± 7.6 30.4 ± 8.6 21.8 ± 8.4 <.05
Skeletal muscle mass, kg/m2 8.0 ± 1.6 9.0 ± 1.6 6.6 ± 1.4 <.05
Biochemical measurements ± SD
Serum albumin level, g/dL 4.1 ± 0.7 4.0 ± 0.4 3.4 ± 0.8 .560
Total iron binding capacity, μg/dL 325.6 ± 54.6 296.4 ± 68.2 222.8 ± 74.2 .266
Hemoglobin, g/dL 12.9 ± 1.8 13.2 ± 1.4 12.4 ± 2.0 .059
Lymphocytes, n × 103 4.1 ± 8.2 2.2 ± 1.0 1.7 ± 1.7 .112
C-reactive protein, mg/dL 0.8 ± 0.4 0.40 ± 0.28 0.90 ± 0.44 <.05
Butyryl-cholinesterase, U/L 7480 ± 2320 8600 ± 1450 3400 ± 2850 <.05

BADLs, basic activities of daily living; BMI, body mass index; CIRS, Cumulative Index Rating Scale; IADLs, instrumental activities of daily
living; MNA, Mini Nutritional Assessment; PASE, Physical Activity Scale for the Elderly.
a Sarcopenia was defined in subjects with a low muscle strength presenting low muscle mass (skeletal muscle index <8.87 and <6.42 kg/m2 in men

and women, respectively). BMI-adjusted values were used as a cutoff point to classify low muscle strength (BMI ࣘ24, 24.1–28, and <28 was 29,
ࣘ30, and ࣘ32 kg for men; BMI ࣘ23, 23.1–26, 26.1–29, and <29 was 17, ࣘ17.3, ࣘ18, and ࣘ21 kg for women, respectively).
b P = statistical difference between the presence and the absence of sarcopenia.

reduction of functional capacity, an increase of disability
and mortality, and therefore an increase of healthcare
costs.3-5,30 Many mechanisms have been hypothesized to
explain the origin of age-related decline in muscle strength
and mass, including inflammation, physical inactivity, and
malnutrition.31,32
Malnutrition and Sarcopenia
Therefore, elderly adults may lose muscle mass and
strength.33 Accordingly, older adults usually eat less protein
than younger adults.31 In fact, community-dwelling older
adults (ࣈ10%) and those living in care homes (ࣈ30%) do not
consume the estimated average requirement for daily protein
intake (0.7 g/kg body weight/day),34,35 which represents the
minimum intake level to maintain muscle integrity for older
adults.36,37

The relationship between malnutrition and sarcopenia is
not well established, especially in elderly outpatients.29 A
disproportion between protein intake and protein needs may
determine a loss of skeletal muscle mass because of im-
balance between muscle protein synthesis and degradation.
Mini Nutritional Assessment as Tool
for Malnutrition Detection
Because malnutrition in the elderly is multifactorial, several
measures of malnutrition in geriatric people have been
Liguori et al 883

Table 2. Baseline Characteristics of the 473 Older Adult Participants Enrolled in the Study stratified for Mini Nutritional
Assessment.

Mini Nutritional Assessment score

ࣙ24 (51.0%) 17-23.5 (34.2%) <17 (14.8%) P

Characteristics (n = 241) (n = 162) (n = 70) for Trend

Anthropometric data
Age ± SD, y 79.7 ± 6.5a 81.3 ± 6.7 81.8 ± 6.3 .001
Female sex, % 27.4 46.3 26.3 .432
Body mass index, kg/m2 28.1 ± 5.0a 27.6 ± 6.2 26.1 ± 5.0 <.05
Waist circumference, cm 100.3 ± 9.4a 98.5 ± 11.2 98.2 ± 10.1 <.05
Geriatric evaluation ± SD
Mini-Mental State Examination score 24.3 ± 4.9a 21.0 ± 6.3 18.5 ± 6.8 <.001
Geriatric Depression Scale score 4.8 ± 3.6a 8.3 ± 4.0 10.1 ± 4.1 <.001
CIRS-Comorbidity score 3.1 ± 2.1 4.1 ± 2.2 4.4 ± 2.0 .456
CIRS-Severity score 1.7 ± 0.4 1.9 ± 0.5 2.0 ± 0.4 .455
No. of drugs 5.2 ± 3.3 6.5 ± 3.0 6.6 ± 3.2 .184
BADLs lost, n 1.0 ± 1.5 2.1 ± 1.8 3.2 ± 1.8 <.05
IADLs lost, n 2.5 ± 2.6 4.5 ± 2.6 5.9 ± 2.2 <.05
Tinetti score 21.9 ± 6.0 16.2 ± 7.5 12.8 ± 7.1 <.05
MNA score 25.8 ± 1.5a 20.6 ± 1.8 14.5 ± 2.2 <.001
4-m gait speed, m/s 0.66 ± 0.47 0.46 ± 0.5 0.33 ± 0.5 <.05
PASE score 65.4 ± 60.8a 28.9 ± 43.8 16.1 ± 28.7 <.05
Social support score 5.4 ± 3.9 8.9 ± 4.6 11.4 ± 4.2 <.05
Fried’s frailty score 2.6 ± 1.5 3.6 ± 1.4 4.5 ± 1.3 <.001
Rockwood’s frailty score 15.7 ± 8.0 22.2 ± 7.4 27.2 ± 7.1 <.001
Sarcopenia, n (%) 15 (6.2) 25 (15.4) 22 (31.4) <.001
Muscle measurements ± SD
Grip strength, kg 41.2 ± 10.2 38.1 ± 9.3 27.7 ± 10.3 <.001
Skeletal muscle mass, kg/m2 17.6 ± 1.5 16.8 ± 2.1 10.0 ± 2.9 <.001
Biochemical measurements ± SD
Serum albumin level, g/dL 4.2 ± 0.5 4.2 ± 0.9 3.9 ± 0.4 .052
Total iron binding capacity, μg/dL 350.4 ± 50.4 341.0 ± 60.2 265.0 ± 80.2 .163
Hemoglobin, g/dL 13.2 ± 1.8 13.0 ± 1.7 12.0 ± 1.9 .420
Lymphocytes, n × 103 5.0 ± 9.8 4.9 ± 9.3 1.2 ± 0.6 <.05
C-reactive protein, mg/dL 0.50 ± 0.42a 0.64 ± 0.50 0.95 ± 0.30 .048
Butyryl-cholinesterase, U/L 7950 ± 2150 7650 ± 2300 7000 ± 2100 .042

BADLs, basic activities of daily living; CIRS, Cumulative Index Rating Scale; IADLs, instrumental activities of daily living; MNA, Mini
Nutritional Assessment; PASE, Physical Activity Scale for the Elderly.
a P < .01 vs MNA score <17.

developed. Biochemical and clinical indices are investi-
gated by Nutritional Risk Index38 and Geriatric Nutri-
tional Risk Index39 ; anthropometry, mobility, cognitive
state, self-perceived health, and nutrition by Short Form
MNA,40 by Malnutrition Universal Screening Tool,41 and
by unintentional weight loss and poor intake (MST)42 ; and
medical history and clinical and subjective evaluation by
Nutrition Risk Screening 2002 (NRS-2002).43 In contrast
with nutrition screening tools, nutrition assessment methods
are more comprehensive and are frequently used by geria-
tricians. In particular, the Subjective Global Assessment44
and MNA45 are largely used to perform a nutrition di-
agnosis and begin nutrition management in older adult
patients.46
Sarcopenia and Mini Nutritional Assessment
Score
In our noninstitutionalized older adults, MNA score pro-
gressively decreases as muscle mass and strength are
reducing in both univariate (r = 0.72 and r = 0.46, respec-
tively) and multivariate (r = 0.62 and r = 0.40, respectively)
analyses. Interestingly, MNA score is significantly lower in
subjects with sarcopenia vs without sarcopenia (15.4 ± 4.2
vs 22.0 ± 4.0; P < .024). Some reports have investigated the
relationship between MNA and sarcopenia. However, this
relationship in noninstitutionalized older adults is poorly
investigated. In contrast, among acute geriatric patients,
malnutrition evaluated with MNA was associated to 30%
884 Nutrition in Clinical Practice 33(6)

Figure 1. Muscle mass and strength stratified by Mini

Nutritional Assessment score (ࣙ24, 17–23.5, <17) in
noninstitutionalized elderly people; P for trend = .001 for
both muscle mass (*P < .05 for <17 vs 17–23.5 and ࣙ24) and
strength (*P < .05 for <17 vs 17–23.5 and ࣙ24).

of the patients with sarcopenia.47 In addition, among male

residents in a nursing home, MNA scores were significantly
lower in residents having low compared with having normal
skeletal muscle mass (17.1 ± 3.4 vs 19.6 ± 2.5; P = .005).48
More importantly, in a prospective study, subjects with
malnutrition risk and sarcopenia have a risk for mortality 4
times higher than subjects without sarcopenia with normal
nutrition.49 Therefore, Vandewoude et al13 recently indi-
cated the MSS to depict the clinical scenario characterized
by both malnutrition and sarcopenia. Finally, it should
be emphasized that malnutrition plays a key role in the
pathogenesis of both frailty and sarcopenia.8 Malnutrition Figure 2. Regression linear relationship among muscle mass
and sarcopenia are frequently present in older adults with (A) and muscle strength (B) and Mini Nutritional Assessment
diabetes and chronic obstructive pulmonary disease that are score for noninstitutionalized elderly people.
characterized by a high degree of frailty.50,51 Interestingly,
the “frailty score” identified by Fried’s and Rockwood’s
methods was 2.6 ± 1.5 and 15.7 ± 8.0 at the highest status should be always evaluated to address therapeutical
MNA and 4.5 ± 1.3 and 27.2 ± 7.1 at the lowest MNA, interventions and lifestyle modifications (ie, increase of
respectively. protein intake and physical activity).3 Thus, it is mandatory
for clinician researchers to develop a reliable and valid
Evaluation of Nutrition Status in Patients With diagnostic algorithm to identify elderly subjects with both
Sarcopenia: Clinical Implications malnutrition and sarcopenia.

As discussed earlier, malnutrition and sarcopenia are con-

ditions frequently found in older adults that lead to nega-
tive outcomes including increased morbidity and mortality
and increased healthcare costs and rehospitalizations.8,13,49
Usually, these conditions have been separately screened,
but rarely are both conditions simultaneously assessed.8,13
Moreover, many patients concurrently show malnutrition
and sarcopenia; therefore, patient’s nutrition and functional
Conclusions
MNA score, 1 of the most accepted instruments for the
assessment of malnutrition, is lower in noninstitutionalized
older adults with sarcopenia. Thus, the presence of mal-
nutrition is strongly related to muscle mass and strength.
These findings support that low MNA score is frequently
associated to elderly outpatients with sarcopenia.
Liguori et al 885

Table 3. Univariate and Multivariate Linear Regression Analyses on Skeletal Muscle Mass and Muscle Strength.

Muscle Mass Muscle Strength

Univariate Multivariate Univariate Multivariate

Variables r P r P r P r P

Age −0.14 <.05 −0.15 <.05 −0.18 <.05 −0.16 <.05

Body mass index 0.06 .204 0.04 .220
Waist circumference 0.12 .340 0.06 .320
Mini-Mental State Examination 0.08 .120 0.08 .102
Geriatric Depression Scale −0.14 .347 −0.10 .300
CIRS-Comorbidity −0.10 .340 −0.10 <.05 −0.08 <.05
CIRS-Severity −0.07 .032 −0.06 <.05 −0.10 <.01 −0.09 <.01
No. of drugs −0.08 .030 −0.07 <.01 −0.04 .042 −0.04 <.05
BADLs −0.14 <.05 −0.12 <.05 −0.14 <.05 −0.12 <.05
IADLs −0.08 .342 −0.04 .442
Mini Nutritional Assessment 0.72 <.01 0.62 <.05 0.46 <.01 0.40 <.05
Tinetti score 0.65 <.01 0.051 .03 0.44 <.01 0.39 <.05
4-m gait speed 0.30 <.05 0.29 <.05 0.32 .05 0.28 <.05
PASE 0.57 <.05 0.30 .04 0.41 .001 0.29 <.05
Social support −0.04 .282 −0.07 .541
Serum albumin level 0.18 .022 0.14 .084 0.12 .064
Total iron binding capacity 0.14 .112 0.14 .122
Hemoglobin 0.24 <.04 0.20 <.05 0.30 <.05 0.26 .020
Lymphocytes 0.14 .114 0.15 .75
C-reactive protein −0.24 <.05 −0.08 .421 −0.20 <.05 −0.18 .325
Butyryl-cholinesterase 0.35 <.01 0.30 <.01 0.44 <.01 0.40 .036

BADLs, basic activities of daily living; CIRS, Cumulative Index Rating Scale; IADLs, instrumental activities of daily living; PASE, Physical
Activity Scale for the Elderly.

Acknowledgments 5. Iannuzzi-Sucich M, Prestwood KM, Kenny AM. Prevalence of sar-

Funding source: Fondazione Roma NCDS-2013-00000331 –
Sarcopenia and Insulin Resistance in the Elderly; Age-
Associated Inflammation as a Shared Pathogenic Mechanism
and Potential Therapeutic Target (DDM), Rome, ITALY.
Statement of Authorship
I. Liguori, F. Cacciatore, D. Bonaduce and P. Abete contributed
to conception and design; D. Della-Morte, G. Gargiulo and
G. Testa contributed to analysis and interpretation of data;
F. Curcio, G. Russo, M. Cellurale L. Aran and G. Bulli
contributed to acquisition of data.
References
1. Baumgartner RN, Koehler KM, Gallagher D, et al. Epidemiology
of sarcopenia among the elderly in New Mexico. Am J Epidemiol.
1998;147:755-763.
2. Marcell TJ. Sarcopenia: causes, consequences, and preventions. J
Gerontol A Biol Sci Med Sci. 2003;58:M911-M916.
3. Rolland Y, Czerwinski S, Abellan Van Kan G, et al. Sarcopenia: its as-
sessment, etiology, pathogenesis, consequences and future perspectives.
J Nutr Health Aging. 2008;12:433-450.
4. Rantanen T, Guralnik JM, Sakari-Rantala R, et al. Disability, physical
activity, and muscle strength in older women: the Women’s Health and
Aging Study. Arch Phys Med Rehabil. 1999;80:130-135.
copenia and predictors of skeletal muscle mass in healthy, older men
and women. J Gerontol A Biol Sci Med Sci. 2002;57:M772-M777.
6. Metter EJ, Talbot LA, Schrager M, Conwit R. Skeletal muscle strength
as a predictor of all-cause mortality in healthy men. J Gerontol A Biol
Sci Med Sci. 2002;57:B359-B365.
7. Landi F, Liperoti R, Fusco D, Mastropaolo S, Quattrociocchi D, Proia
A. Sarcopenia and mortality among older nursing home residents. J Am
Med Dir Assoc. 2012;13:121-126.
8. Cruz-Jentoft AJ, Kiesswetter E, Drey M, Sieber CC. Nutrition, frailty,
and sarcopenia. Aging Clin Exp Res. 2017;29:43-48.
9. Walrand S, Guillet C, Salles J, Cano N, Boirie Y. Physiopathological
mechanism of sarcopenia. Clin Geriatr Med. 2011;27:365-385.
10. Cooper C, Fielding R, Visser M, et al. Tools in the assessment of
sarcopenia. Calcif Tissue Int. 2013;93:201-210.
11. Mithal A, Bonjour JP, Boonen S, et al; IOF CSA Nutrition Working
Group. Impact of nutrition on muscle mass, strength, and performance
in older adults. Osteoporos Int. 2013;24:1555-1566.
12. Agarwal E, Miller M, Yaxley A, Isenring E. Malnutrition in the elderly:
a narrative review. Maturitas. 2013;76:296-302.
13. Vandewoude MF, Alish CJ, Sauer AC, Hegazi RA. Malnutrition-
sarcopenia syndrome: is this the future of nutrition screening and
assessment for older adults? J Aging Res. 2012;2012:651570.
14. van Bokhorst-de van der Schueren MA, Guaitoli PR, Jansma EP, et al.
Nutrition screening tools: does one size fit all? A systematic review of
screening tools for the hospital setting. Clin Nutr. 2014;33:39-58.
15. Guigoz Y. The Mini Nutritional Assessment (MNA) review of the
literature: what does it tell us? J Nutr Health Aging. 2006;10:466-485.
886
16. Testa G, Cacciatore F, Galizia G, et al. Waist circumference but not
body mass index predicts long-term mortality in elderly subjects with
chronic heart failure. J Am Geriatr Soc. 2010;58:1433-1440.
17. Cacciatore F, Testa G, Langellotto A, et al. Role of ventricular rate
response on dementia in cognitively impaired elderly subjects with atrial
fibrillation: a 10-year study. Dement Geriatr Cogn Disord. 2012;34:143-
148.
18. Testa G, Cacciatore F, Galizia G, et al. Depressive symptoms predict
mortality in elderly subjects with chronic heart failure. Eur J Clin Invest.
2011;41:1310-1317.
19. Testa G, Cacciatore F, Galizia G, et al. Charlson Comorbidity Index
does not predict long-term mortality in elderly subjects with chronic
heart failure. Age Ageing. 2009;38:734-740.
20. Washburn RA, Smith KW, Jette AM, Janney CA. The Physical Activity
Scale for the Elderly (PASE): development and evaluation. J Clin
Epidemiol. 1993;46:153e62.
21. Peel NM, Kuys SS, Klein K. Gait speed as a measure in geriatric
assessment in clinical settings: a systematic review. J Gerontol A Biol
Sci Med Sci. 2013;68:39-46.
22. Abete P, Cacciatore F, Ferrara N, et al. Body mass index and preinfarc-
tion angina in elderly patients with acute myocardial infarction. Am J
Clin Nutr. 2003;78:796-801.
23. Mazzella F, Cacciatore F, Galizia G, et al. Social support and long-
term mortality in the elderly: role of comorbidity. Arch Gerontol
Geriatr. 2010;51:323-328.
24. Abete P, Basile C, Bulli G, et al. The Italian version of the “frailty
index” based on deficits in health: a validation study. Aging Clin Exp
Res. 2017;29(5):913-926.
25. Fried LP, Tangen CM, Walston J, et al; Cardiovascular Health Study
Collaborative Research Group. Frailty in older adults: evidence for a
phenotype. J Gerontol A Biol Sci Med Sci. 2001;56:M146-M156.
26. Vellas B, Guigoz Y, Garry PJ, et al. The Mini Nutritional Assessment
(MNA) and its use in grading the nutritional state of elderly patients.
Nutrition. 1999;15:116-122.
27. Janssen I, Heymsfield SB, Baumgartner RN, Ross R. Estimation
of skeletal muscle mass by bioelectrical impedance analysis. J Appl
Physiol. 2000;89:465-471.
28. Volpato S, Bianchi L, Cherubini A, et al. Prevalence and clinical cor-
relates of sarcopenia in community-dwelling older people: application
of the EWGSOP definition and diagnostic algorithm. J Gerontol A Biol
Sci Med Sci. 2014;69:438-446.
29. Cruz-Jentoft AJ, Landi F, Topinkova E, et al. Understanding sarcope-
nia as a geriatric syndrome. Curr Opin Clin Nutr Metab Care. 2010;13:
1-7.
30. Goodpaster BH, Chomentowski P, Ward BK, et al. Effects of physical
activity on strength and skeletal muscle fat infiltration in older adults:
a randomized controlled trial. J Appl Physiol. 2008;105:1498-1503.
31. Morley JE, Baumgartner RN. Cytokine-related aging process. J Geron-
tol A Biol Sci Med Sci. 2004;59:924-929.
32. Soeters PB, Schols AM. Advances in understanding and assessing
malnutrition. Curr Opin Clin Nutr Metab Care. 2009;12:487-494.
33. Meng X, Zhu K, Devine A, Kerr DA, Binns CW, Prince RL. A 5-year
cohort study of the effects of high protein intake on lean mass and
BMC in elderly postmenopausal women. J Bone Miner Res. 2009;24:
1827-1834.
34. Volpi E, Campbell WW, Dwyer JT, et al. Is the optimal level of
protein intake for older adults greater than the recommended dietary
allowance. J Gerontol A Biol Sci Med Sci. 2013;68:677-681.
Nutrition in Clinical Practice 33(6)
35. Tieland M, Borgonjen-Van Den Berg K, van Loon LJ, de Groot LC.
Dietary protein intake in community-dwelling, frail, and institutional-
ized elderly people: scope for improvement. Eur J Nutr. 2012;51:173-
179.
36. Bauer J, Biolo G, Cederholm T, et al. Evidence-based recommenda-
tions for optimal dietary protein intake in older people: a position paper
from the PROT-AGE study group. J Am Med Dir Assoc. 2013;14:542-
559.
37. Deutz NE, Bauer JM, Barazzoni R, et al. Protein intake and exercise
for optimal muscle function with aging: recommendations from the
ESPEN Expert Group. Clin Nutr 2014;33:929-936.
38. Wolinsky F, Coe RM, McIntosh WA, et al. Progress in the development
of a nutritional risk index. J Nutr. 1990;120:1549-1553.
39. Cereda E, Pedrolli C. The geriatric nutritional risk index. Curr Opin
Clin Nutr Metab Care. 2009;12:1-7.
40. Rubenstein L, Harker J, Salva A, Guigoz Y, Vellas B. Screening
for undernutritionin geriatric practice: developing the short-form
mini nutritional assessment (MNA-SF). J Gerontol. 2001;56A:M366-
M372.
41. Malnutrition Advisory Group (MAG). The ‘MUST’ Explanatory
Booklet. A Guide to the ‘Malnutrition Universal Screening Tool’
(‘MUST’) for Adults: A Standing Committee of the British Association
for Parenteral and Enteral Nutrition (BAPEN). Worcestershire (UK):
Redditch; 2003.
42. Ferguson M, Capra S, Bauer J, Banks M. Development of a valid and
reliable malnutrition screening tool for adult acute hospital patients.
Nutrition. 1999;15:458-464.
43. Kondrup J, Rasmussen H, Hamberg O, et al. An ad hoc ESPEN Work-
ing Group. Nutritional Risk Screening (NRS 2002): a new method
based on analysis of controlled clinical trials. Clin. Nutr. 2003;22:321-
336.
44. Detsky A, McLaughlin J, Baker J. What is subjective global assess-
ment of nutritional status. JPEN J Parenter Enteral Nutr. 1987;11:
8-13.
45. Guigoz Y, Vellas B, Garry P. Assessing the nutritional status of
the elderly: the mini nutritional assessment as part of the geriatric
evaluation. Nutr Rev. 1996;54:S59-S65.
46. Young AM, Kidston S, Banks MD, et al. Malnutrition screening tools:
comparison against two validated nutrition assessment methods in
older medical inpatients. Nutrition. 2013;29:101-106.
47. Jacobsen EL, Brovold T, Bergland A, et al. Prevalence of factors
associated with malnutrition among acute geriatric patients in Norway:
a cross-sectional study. BMJ Open. 2016;6:e011512.
48. Tufan A, Bahat G, Ozkaya H, et al. Low skeletal muscle mass index is
associated with function and nutritional status in residents in a Turkish
nursing home. Aging Male. 2016;19:182-186.
49. Hu X, Zhang L, Wang H, et al. Malnutrition-sarcopenia syndrome
predicts mortality in hospitalized older patients. Sci Rep. 2017;7:
3171.
50. Galizia G, Cacciatore F, Testa G, et al. Role of clinical frailty on
long-term mortality of elderly subjects with and without chronic
obstructive pulmonary disease. Aging Clin Exp Res. 2011;23:
118-125.
51. Abete P, Della-Morte D, Gargiulo G, Basile C, Langellotto A, Galizia
G, Testa G, Canonico V, Bonaduce D, Cacciatore F. Cognitive
impairment and cardiovascular diseases in the elderly. A heart-brain
continuum hypothesis. Ageing Res Rev. 2014;18:41-52.
Clinical Research

Nutrition in Clinical Practice

Volume 33 Number 6
Height Prediction From Ulna Length of Critically Ill Patients December 2018 887–892

C 2017 American Society for

Parenteral and Enteral Nutrition

DOI: 10.1177/0884533617716432
wileyonlinelibrary.com
Micheli S. Tarnowski, RD1 ; Estela I. Rabito, RD, PhD2 ; Daieni Fernandes, RD, MSc3 ;
Mariane Rosa, RD4 ; Manoela L. Oliveira, RD5 ; Vânia N. Hirakata, MSc6 ;
and Aline Marcadenti, RD, PhD7

Abstract
Background: Ulna length (UL) has been used in mathematical formulas to predict the body height of healthy and sick individuals.
However, the evaluation of its use with patients admitted to intensive care units (ICU) is scarce. The objective of this study was
to develop a mathematical equation to estimate critically ill patients’ height using the UL measure and to evaluate its agreement
with measured standing height. Methods: This cross-sectional study was performed at the ICU of a tertiary hospital in Brazil. A
total of 100 patients aged ࣙ18 years who had their body height measured before ICU admission were enrolled. The equation was
developed through multiple linear regression, and its agreement was assessed through paired Student’s t test and Bland-Altman
plot. Results: The following formula was obtained: height in cm = 153.492 – (7.97 × sex [sex: male = 1, female = 2]) + (0.974 ×
UL [in cm]). The difference between means of measured height (MH) and height estimated from UL was not significant (166.26
± 8.75 cm and 166.30 ± 5.29 cm, respectively, P = .96), and a significant correlation (r = 0.624, P < .001) was detected. In the
Bland-Altman analysis, UL was in agreement with MH; however, there was a significant bias (P < .001) suggesting that it may be
disproportional and dependent on the average’s height value. Conclusion: The mathematical equation for height estimation using
UL developed in this study matched the MH of critically ill patients. However, we suggest more studies for its validation. (Nutr
Clin Pract. 2018;33:887–892)

Keywords
nutritional assessment; ulna; body height; intensive care units; critical illness

Nutrition assessment is crucial to develop therapeutic plans
and evaluate their effectiveness, particularly for critically ill
patients.1 Anthropometric evaluation is generally impaired
for critical patients because of the high complexity of their
clinical state, with limitations and difficulties related to im-
mobilization, the equipments required, and the significant
edema often present in this population.2
The Sociedad Española de Nutrición Parenteral y En-
teral (Spanish Society for Parenteral and Enteral Nutrition)
recommends that body mass and height be assessed when
patients are admitted to intensive care units (ICUs).3 Such
assessment can be used for nutrition and medical manage-
ment procedures, such as drug dose prescription, parame-
ters for mechanical ventilation, energy needs estimate, and
ideal body weight calculation.3-5
The absence of exact anthropometric information is a
reality in ICUs, where there is no gold standard for nutrition
assessment in critically ill patients. In this context, the
definition of energy and protein prescription goals must
be based on plausible estimates of body mass and height,
to avoid malnutrition, overfeeding, and incorrect nutrition
diagnosis.4 Therefore, measurements of different body com-
partments, such as the length of long bones (considered
good predictors of body height), are feasible alternatives.6
Ulna length (UL) has been used because it is a reliable height
predictor, easily obtained in bedridden patients, and less
affected by the aging process when the standing body height
measure is difficult to obtain.7
Studies suggest the use of mathematical equations for de-
termining body height through UL.6,8-14 The Malnutrition
From the 1 Health Multidisciplinary Residence Program in Intensive
Therapy, Federal University of Health Sciences of Porto Alegre,
Porto Alegre, Brazil; 2 Department of Nutrition and Postgraduate
Program in Food and Nutrition, Federal University of Parana,
Curitiba, Brazil; 3 Division of Nutrition, Brotherhood of the Santa
Casa of Porto Alegre, Porto Alegre, Brazil; 4 Unimed Grande
Florianopolis Hospital, São José, Brazil; 5 Department of Nutrition,
Federal University of Health Sciences of Porto Alegre, Porto Alegre,
Brazil; 6 Clinics Hospital of Porto Alegre, Porto Alegre, Brazil; and
7 Department of Nutrition, Federal University of Health Sciences of
Porto Alegre, and Postgraduate Program in Health Sciences:
Cardiology, Institute of Cardiology of the Rio Grande do Sul, Porto
Alegre, Brazil.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on July 20, 2017.
Corresponding Author:
Aline Marcadenti, RD, PhD, Department of Nutrition, Federal
University of Health Sciences of Porto Alegre, Rua Sarmento Leite,
245 Porto Alegre, 90050-170, Rio Grande do Sul, Brazil.
Email: alinemo@ufcspa.edu.br
888 Nutrition in Clinical Practice 33(6)

Universal Screening Tool (MUST) also suggests the use of Height was measured by the hospital nursing and nutri-
UL to estimate standing height for body mass index (BMI) tion staff, with stadiometers assembled with weighing scales
calculation to determine the nutrition risk classification.15 available in inpatient hospital units. The professionals were
However, such methods and equations were developed and trained to follow the same protocol: place the patient in a
validated with distinct population groups, based on sex, standing position, barefooted, wearing the least amount of
age, and race, but not critically ill patients. The use of clothing possible, with arms positioned alongside the body
UL to predict the height of critically ill patients would be and hands turned toward the thighs. MH was recorded in
useful, being a quick and easily applicable method. Besides, centimeters.
alternative measurements to estimate body height, such as The other anthropometric indexes were measured in the
knee height (KH), may have some limitations in the ICU, first 48 hours after ICU admission by 2 trained nutritionists:
particularly with patients who are overweight, have bilateral
amputations, or have femoral venous access.6
UL (cm): Measured with a bone anthropometer caliper
The use of UL to predict standing height was shown in
(WCS: Cardiomed, Curitiba, Brazil; range up to 280
a Brazilian population admitted to an emergency room.16
cm; 1-mm resolution) positioned between the elbow’s
However, in this study, the method suggested by MUST was
tip (olecranon process) and the midpoint of the wrist
used to identify the values of estimated height according
prominent bone (styloid process), according to MUST
to UL. Therefore, the main objective of our study was to
protocol.15 UL was measured preferentially in the left
develop a mathematical formula to estimate the standing
arm of the patient, folded in front of the patient’s chest,
body height of critically ill patients with their UL and to
with fingers pointing to the opposite shoulder.
evaluate its agreement with the measured height (MH). As
KH (cm): Measured with a bone anthropometer caliper
a secondary objective, we evaluated the agreement between
(WCS) with individuals in the supine position and the
the estimated height according to a mathematical formula
right leg forming a 90° angle with knee and ankle.
that uses the KH measurement and the MH of the subjects.
Estimated body height according to KH (KHH) was
calculated according to mathematical formulas proposed
Methods by Chumlea et al20 : for men, 64.19 – (0.04 × age in years)
+ (2.02 × KH in cm); for women, 84.88 – (0.24 × age in
This cross-sectional study was carried out between Septem-
years) + (1.83 × KH in cm).
ber 2015 and August 2016. Patients admitted to a general
ICU were enrolled (Santa Clara Hospital, Brotherhood
of the Santa Casa of Porto Alegre, Porto Alegre, Brazil). Statistical Analysis
The study included individuals of both sexes, aged ࣙ18
The analyses were performed with SPSS 22.0 (IBM,
years, whose standing body height was previously measured
Chicago, IL). Continuous variables were described with
at inpatient hospital units and recorded in their medical
mean and standard deviation (variables with normal distri-
records before being admitted to the ICU. They agreed to be
bution) or median and interquintile interval (nonparametric
part of the research and signed the informed consent form
variables); absolute numbers and frequencies were used to
(or their legal representatives). Individuals with incomplete
describe qualitative variables. Multiple linear regression,
medical records, those without MH previously registered,
in stepwise method, was used to create the body height–
and those whose KH could not be obtained at the ICU were
predictive mathematical formula based on UL. Pearson’s
excluded.
correlation test was used for correlations of the different
The data were recorded in a standardized questionnaire.
heights (MH, UL height [ULH], and KHH). Agreements
Age (years), sex, ICU length of stay (days), and reason
among the heights were evaluated through Student’s t test
for admission in the ICU (cardiovascular, gastrointestinal,
for paired samples, multiple linear regression for assessing
respiratory, postsurgery, infection/sepsis, shock, and other)
the bias significance, and Bland-Altman plot. P < .05 values
were the information collected from the electronic medical
were considered significant.
records to characterize the sample.
The study was approved by the Research Ethics Com-
The nutrition risk and/or nutrition status classification
mittee of the Brotherhood of the Santa Casa of Porto Ale-
was obtained through the Nutrition Risk in the Critically
gre (protocol 40073414.9.0000.5335) and by the Research
Ill (NUTRIC) score,17 Subjective Global Assessment,18 and
Ethics Committee of the Federal University of Health
BMI (kg/m2 ), calculated with the patient’s body mass and
Sciences of Porto Alegre.
MH. The patient’s body mass (kg) was obtained directly
through the electronic medical record (if the patient was
weighed up to 48 hours before ICU admission) or estimated
Results
through a mathematical formula19 (in the first 48 hours after In total, 100 patients who stayed a median of 8 days
of ICU admission). (range 3–15) in the ICU were evaluated. Regarding the
Tarnowski et al 889

Table 1. Characteristics of the Sample.a

Variables Total Sample (N = 100)

Sex
Male 57 (57)
Female 43 (43)
Age, y 62.47 ± 16.60
Body mass index, kg/m2 24.49 ± 4.93
NUTRIC score
High nutrition risk 22 (22)
Low nutrition risk 88 (88)

NUTRIC, Nutrition Risk in the Critically Ill.

a Values are presented as n (%) or mean ± SD.

Table 2. Linear Regression Model Used to Identify the

Predictive Equation for Height.a

Nonstandardized Figure 1. Scatter plot showing the correlation between

Coefficient measured height (MH) and height estimated from ulna length
(ULH; equation proposed by authors). r = 0.624 (P < .001).
Factors B SE t P Value

Constant 153.492 9.467 16.214 <.001

Sex −7.970 1.632 −4.884 <.001
Ulna length, cm 0.974 0.322 3.030 .003

a Dependent variable: standing measured height (cm). R2 = 0.364.

leading causes to ICU admission, 35% of the patients had

postsurgery complications; 21% were in shock; and 12%
had gastrointestinal or respiratory diseases. According to
the Subjective Global Assessment, 28% of the patients
were moderately undernourished, and 20% were severely
undernourished; 7% had a BMI <18.5 kg/m2 . Demographic
and anthropometric data related to the sample are described
in Table 1.
Two models were used for the construction of the predic-
tive mathematical formula: in the first, the linear regression
model included age, sex, and UL; in the second, sex and UL
only. The analysis showed that among all variables used, sex Figure 2. Scatter plot showing the correlation between
measured height (MH) and height estimated from knee height
and UL were the best predictors of MH, the residues of
(KHH; equations proposed by Chumlea et al20 ). r = 0.592
which showed normal distribution and were independent of (P < .001).
one another (Durbin-Watson test = 1.742). Therefore, the
following mathematical formula for height prediction was
obtained: height in cm = 153.492 − (7.97 × sex [sex: male between the averages of MH and ULH (difference = −0.04
= 1, female = 2]) + (0.974 × UL [in cm]). Statistical data cm, P = .96), suggesting an agreement between the methods.
related to the equation of ULH are presented in Table 2. The Bland-Altman plot (Figure 3) shows the limits of
Figures 1 and 2 show scatter plots and the correlations agreement (mean differences ± 2 standard deviations) as
between MH and ULH (Figure 1) and between MH and well as the bias. The result of this was significant in the linear
KHH (Figure 2). Both were statistically significant (P < regression analysis (P < .001), suggesting that the limits of
.001); however, MH and ULH presented a superior corre- agreement depend on the average height values and that
lation (r = 0.624) versus MH and KHH (r = 0.592). the bias may be disproportional: when mean height values
Regarding means of different measures of height, MH are small, the difference between methods is small; when
was 166.26 ± 8.75 cm; ULH, 166.30 ± 5.29 cm; and mean height values increase, so does the difference between
KHH, 157.56 ± 8.73 cm. There was no significant difference methods.
890
Figure 3. Bland-Altman plot showing the agreement between
measured height and height estimated from ulna length. The
dotted line indicates bias, and the continuous lines indicate the
limits of agreement. IL, inferior limit; SL, superior limit.
Figure 4. Concordance and confidence intervals for the
differences (y) and the means (x) between measured height
and height estimated from ulna length. Linear regression
equation: y = −99.6 + 0.6 × x; R2 = 0.312.
Because there was a significant association between
the differences and the averages (bias: −0.04 cm, P <
.001), we decided that the relationship between averages
and differences (estimates of maximum differences) should
be expressed according to a line of linear regression.
Figure 4 shows the concordance and confidence intervals
for the difference (cm) between MH and ULH (y) and
the average (cm) between MH and ULH (x). The linear
Nutrition in Clinical Practice 33(6)
Figure 5. Bland-Altman plot showing the agreement between
measured height and height estimated from knee height.
Dotted line indicates bias, and continuous lines indicate limits
of agreement. IL, inferior limit; SL, superior limit.
regression equation, y = −99.6 + 0.6 × x, suggests that the
difference between MH and ULH is dependent on MH. For
example, the difference between the proposed mathematical
formula (ULH) and the standing real height (MH) of an
individual at 155 cm would be −6.6 cm; for an individual
with 167 cm of body height, it would be 0.6 cm; and for an
adult measuring 189 cm, the difference would be 13.8 cm.
A significant difference between MH and KHH averages
was observed (difference: 8.69 cm, P < .001). However, this
bias was not significant in the linear regression analysis (P
= .98), suggesting that it is proportional, as observed in
Figure 5. The limits-of-agreement interval on the Bland-
Altman plot was higher when compared with ULH.
Discussion
This study evaluated the use of UL as a method to estimate
the height of critically ill patients. Because performing the
height measurement through the gold standard (upright
posture) is often very difficult in this setting, our main
goal was to develop a mathematical formula for height
estimation with UL.
We decided to exclude both the recumbent height
measurement—because it is unfeasible for many critically
ill patients—and the patient’s reported body height, due to
the trend for height reduction related to the aging process
and the fact that many individuals ignore their actual height
(ie, many elderly remember only their height measured at
youth).21
Due to alternative methods for estimating height, some
equations based on UL have been explored among dif-
ferent populations, such as the British and Portuguese,14
Tarnowski et al
Turks,13 Sudanese,22 Indians and Iraqis,12 and Sri Lankan,11
as well as with American9,23 and Australian5 pediatric
populations. However, these mathematical formulas were
created for healthy populations and were based on different
ethnicities.10 Regression analysis is considered a reliable tool
for height estimations, and one of the advantages of this
method is to create an equation with a single body measure.
However, due to its less accurate predictive capability in the
face of great variability among the body proportions of dif-
ferent populations, it is suggested that each predictive equa-
tion be specifically created for the population in question.22
Using UL, we identified an agreement between the height
estimated by our mathematical formula and the patient’s
MH, despite some reservations. We observed that the esti-
mation resulting from the mathematical formula tends to be
influenced by extremes; that is, for tall individuals, it tends
to overestimate stature, while for those of very low stature, it
tends to underestimate it. In sum, its utility is limited at the
extremes of height. The equation based on KH for height
estimation presented a systematic bias; that is, KHH was
not influenced by extremes, and the bias is uniform among
patients, being less influenced by very tall or very short
individuals. Despite the large limits-of-agreement interval
observed in our study regarding ULH (13.34 to −13.46
cm), it is lower than that observed in a study conducted in
a Brazilian emergency room (14.39 to −13.69 cm), which
used the MUST protocol to evaluate the values of estimated
height with UL.16
In a study performed in the ICU of a European univer-
sity hospital,6 researchers assessed whether different math-
ematical formulas of height prediction validated among
healthy populations would be accurate for critically ill
patients, and their correlation between the height estimated
through UL and MH was lower than the correlation ob-
served in our study (r = 0.51). It reinforces the impor-
tance of elaboration and validation of specific predictive
equations for each population. Ahmed22 assessed the height
prediction of a Sudanese adult population through the long
bones of upper limbs; with an equation based on UL, a
positive correlation was observed in relation to MH (r =
0.73 for men and r = 0.72 for women), showing that the use
of this body compartment may be a feasible alternative for
estimating height, for both men and women.
In a study with critically ill patients performed in a
Swiss university hospital,4 MH and the Chumlea method
for estimating height were compared. Similar to our re-
sults, a satisfactory correlation was observed between body
heights estimated by KH and MH; however, it presented an
overestimation of height values, with a varied and randomly
distributed dispersion.
Al-Wasfi and Puranik12 assessed the height prediction
of 196 healthy Indians and Iraqis, using measurements of
2 body parts: fibula and ulna. Based on that, 3 equations
were created: first, using fibula length; second, UL; third,
891
both. The best regression model was the one based on
both variables (fibula length + UL), showing a strong and
significant correlation (r = 0.87, P < .001). The predictive
formula based on only UL showed moderate correlations
for Indians and Iraqis (r = 0.62 and r = 0.59, respectively),
supporting our results.
Our study had some limitations. Among them, we men-
tion the sample size, because the enrollment of a larger
number of individuals could imply a bias decrease. How-
ever, our results are consistent with other studies assess-
ing healthy populations. In our study, we evaluated only
patients whose MH was listed on the medical record, and
such procedures are not always performed in hospitals—
underreporting of patients’ weight and height data is not un-
common. Although professionals were previously trained,
we cannot warrant that the protocol for height measure was
homogenized among all who were responsible for measuring
patients’ height in the hospital admission units.
Conclusion
The mathematical equation based on UL to predict body
height that was developed in this study seems to provide
good standing height estimation for critically ill patients.
We suggest, however, that our results be replicated in other
populations of hospitalized individuals, to reinforce our
findings. Regarding next steps of our research, we intend to
validate our equation in a larger population of critically ill
patients.
Acknowledgments
We thank the Brotherhood of the Santa Casa de Porto Alegre,
its nursing and central ICU staff, and the Nutrition and
Dietetics Service, as well as the Federal University of Health
Sciences of Porto Alegre. We also thank Marcelo Oliveira da
Silva (Lecttura Traduções) for English language support.
Statement of Authorship
A. Marcadenti, D. Fernandes, and E. I. Rabito equally con-
tributed to the conception and design of the research; M. Rosa
contributed to the design of the research; M. Rosa, M. L.
Oliveira, and M. S. Tarnowski contributed to the acquisition
of the data; A. Marcadenti, E. I. Rabito, and V. N. Hirakata
contributed to the analysis and interpretation of the data; and
M. S. Tarnowski and A. Marcadenti drafted the manuscript.
All authors critically revised the manuscript, agree to be fully
accountable for ensuring the integrity and accuracy of the
work, and read and approved the final manuscript.
References
1. Hejazi N, Mazloom Z, Zand F, Rezaianzadeh A, Amini A. Nutritional
assessment in critically ill patients. Iran J Med Sci. 2016;41(3):171-179.
2. Ghorabi S, Ardehali H, Amiri Z, Shariatpanahi VZ. Association of the
adductor pollicis muscle thickness with clinical outcomes in intensive
care unit patients. Nutr Clin Pract. 2016;31(4):523-526.
892
3. Ruiz-Santana S, Sánchez AJA, Abilés J. Guidelines for specialized
nutritional and metabolic support in the critically-ill patient: update.
Consensus SEMICYUC-SENPE: nutritional assessment. Nutr Hosp.
2011;26(2):12-50.
4. Berger MM, Cayeux M, Schaller M, Piazza G, Chiolero RL. Stature
estimation using the knee height determination in critically ill patients.
e-SPEN. 2008;3:e84-88.
5. Gauld LM, Kappers JBN, Carlin JB, Robertson. Height prediction
from ulna length. Dev Med Child Neurol. 2004;46:475-480.
6. L’her E, Martin Babau J, Lellouch F. Accuracy of height estimation
and tidal volume setting using anthropometric formulas in an ICU
Caucasian population. Ann Intensive Care. 2016;6(1):55.
7. Dennis DM, Hunt EE, Budgeon CA. Measuring height in recumbent
critical care patients. Am J Crit Care. 2015;24(1):41-47.
8. Lorini C, Collini F, Castagnoli M, et al. Using alternative or direct
anthropometric measurements to assess risk for malnutrition in nursing
homes. Nutrition. 2014;30(10):1171-1176.
9. Forman MR, Zhu Y, Hernandez LM, et al. Arm span and ulnar length
are reliable and accurate estimates of recumbent length and height in a
multiethnic population of infants and children under 6 years of age. J
Nutr. 2014;144(9):1480-1487.
10. Madden AM, Tsikoura T, Stott DJ. The estimation of body height from
ulna length in healthy adults from different ethnic groups. J Hum Nutr
Diet. 2012;25(2):121-128.
11. Ilayperuma I, Nananyakkara G, Palahepitiya N. A model for the
estimation of personal stature from the length of forearm. Int J
Morphol. 2010;28(4):1081-1086.
12. Al-Wasfi AAH, Puranik MG. Estimation of height from length of ulna
and length of fibula in Indian and Iraqi population. Int J Inf Res Rev.
2015;2:904-908.
13. Duyar I, Peli C. Estimating body height from ulna length: need of
a population-specific formula. Eurasian J Anthropol. 2010;1(1):11-
17.
Nutrition in Clinical Practice 33(6)
14. Barbosa VM, Stratton RJ, Lafuente E, Elia M. Ulna length to predict
height in English and Portuguese patient populations. Eur J Clin Nutr.
2012;66(2):209-215.
15. Elia M. The MUST Report. Nutritional Screening of Adults: A Multi-
disciplinary Responsibility. Development and Use of the Malnutrition
Universal Screening Tool (MUST) for Adults. Redditch, England:
BAPEN; 2003.
16. Silva FM, Figueira L. Estimated height from knee height or ulna length
and self-reported height are no substitute for actual height in inpatients.
Nutrition. 2017;33:52-56.
17. Rosa M, Heyland DK, Fernandes D, Rabito EI, Oliveira ML, Marca-
denti A. Translation and adaptation of the NUTRIC score to identify
critically ill patients who benefit the most from nutrition therapy. Clin
Nutr ESPEN. 2016;14:31-36.
18. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective
global assessment of nutritional status? JPEN J Parenter Enteral Nutr.
1987;1:8-13.
19. Chumlea WC, Guo S, Roche AF, Steinbaugh ML. Prediction of body
weight for the nonambulatory elderly from anthropometry. J Am Diet
Assoc. 1988;88(5):564-568.
20. Chumlea WC, Roche AF, Steinbaugh ML. Estimating stature from
knee height for persons 60 to 90 years of age. J Am Geriatr Soc.
1985;33(2):116-120.
21. Sorkin JD, Muller DC, Andres R. Longitudinal change in height of
men and women: implications for interpretation of the body mass
index: the Baltimore longitudinal study of aging. Am J Epidemiol.
1999;150(9):969-977.
22. Ahmed AA. Estimation of stature from the upper limb measurements
of Sudanese adults. Forensic Sci Int. 2013;228(1-3):178.e1-e7.
23. Haapala H, Peterson MD, Daunter A, Hurvitz EA. Agreement be-
tween actual height and estimated height using segmental limb lengths
for individuals with cerebral palsy. Am J Phys Med Rehabil. 2015;
94(7):539-546.
Clinical Observations

Nutrition in Clinical Practice

Volume 33 Number 6
Venovenous Extracorporeal Membrane Oxygenation December 2018 893–896

C 2018 American Society for

in an Adult Patient With Prader-Willi Syndrome: A Nutrition Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10069
Case Report wileyonlinelibrary.com

Stacy Pelekhaty, RD, LDN, CNSC1 ; and Jay Menaker, MD2

Abstract
Prader-Willi Syndrome (PWS) is a genetic condition that results in a constellation of symptoms and typically results in hyperphagia
and obesity in adulthood. Critically ill adults with PWS present a unique challenge to the nutrition professional, particularly when
they require support modalities such as extracorporeal membrane oxygenation (ECMO). The purpose of this case study is to review
the nutrition care of a critically ill adult patient with PWS who required venovenous ECMO. The patient was successfully managed
with a hypocaloric, high-protein approach, which did not result in the diagnosis of malnutrition during his hospitalization. The
patient was ultimately transitioned off extracorporeal life support and discharged to a rehabilitation facility. (Nutr Clin Pract.
2018;33:893–896)

Keywords
critical illness; enteral nutrition; extracorporeal membrane oxygenation; nutrition assessment; nutritional support; Prader-Willi
Syndrome

Prader-Willi syndrome (PWS) is a genetic condition that Case Study

results from the lack of expression of paternal genes from
chromosome 15q11.2-q13.1 Characteristic features of PWS
include hyperphagia, absent gag reflex, high pain tolerance,
developmental delays, and obesity in adults.1 However,
adult patients with PWS may not experience the same
metabolic alterations as age-, sex-, and weight-matched con-
trol subjects; therefore, generalized guidelines for obesity
should be applied with caution.2-7 Optimizing the nutrition
care of the critically ill adult patient with PWS therefore
presents a unique challenge.
Venovenous extracorporeal membrane oxygenation (VV
ECMO) is a life support method that provides oxygena-
tion and ventilation to critically ill patients in profound
respiratory failure.8 The use of ECMO to support adult
patients has increased significantly in the past 10 years,
with 2807 cases in 251 centers in the United States reported
in 2017 according to the Extracorporeal Life Support
Organization.9 A 2015 study by Kon and colleagues10
demonstrated that obese patients who require VV ECMO
had similar outcomes as nonobese patients. In addition, the
PRESERVE score has increased body mass index (BMI)
as protective against mortality in patients who require VV
ECMO.11 The purpose of this case report is to review the
nutrition management of an adult male with PWS who
requires VV ECMO.
A 24-year-old male with known PWS presented to an
outside hospital with several days of upper respiratory
symptoms. The patient was hypoxic and initially tried on
noninvasive ventilation; however, his condition continued
to deteriorate and required endotracheal intubation. The
patient’s clinical course continued to worsen, and he ex-
perienced hypotension that required vasoactive support.
Due to this shock state and overwhelming critical illness,
the patient experienced development of acute anuric renal
failure that required renal replacement therapy (RRT).
Despite maximal therapy provided by the outside hospital,
the patient’s respiratory status continued to decline and
From the 1 Department of Clinical Nutrition, University of Maryland
Medical Center, Baltimore, Maryland, USA; and 2 Department of
Surgery, University of Maryland School of Medicine, R Adams
Cowley Shock Trauma Center, Baltimore, Maryland, USA.
Financial disclosure: None.
Conflicts of interest: None.
This article originally appeared online on March 12, 2018.
Corresponding Author:
Stacy Pelekhaty, RD, LDN, CNSC, Department of Clinical
Nutrition, University of Maryland Medical Center, 22 S. Greene
Street, Baltimore, MD 21409, USA.
Email: spelekhaty@umm.edu
894
Table 1. Estimated Energy and Protein Requirements.
Nutrient Permissive Underfeeding5
Energy (kcal) 22–25 kcal/kg IBW Penn State Equation (Tmax 36.7, Mve 0) 20–30 kcal/kg
1408–1600 kcal/day 2637 × 1.1–1.2 = 2901–3164 kcal/day
Protein ࣙ2.5 g/kg IBW
ࣙ160 g/day
IBW, ideal body weight; Mve, minute ventilation (Liters per minute); Tmax, 24-hour maximum temperature in degrees Celsius.
on hospital day 3 he was transferred to our institution
for VV ECMO evaluation. On arrival to our facility, the
patient had a peak airway pressure on the ventilator of
44 cm H2 O with marginal oxygenation. The patient was
cannulated for VV ECMO based on his borderline ratio
of partial pressure of arterial oxygen (PaO2 ) and fraction
of inspired oxygen delivered by the ventilator (FIO2 ) and
high ventilatory needs. A 29 French drainage cannula was
placed in the right common femoral vein and 23 French
arterial cannula was placed in the right internal jugular vein
for return. After cannulation, the patient was transferred
to our dedicated multidisciplinary intensive care unit for
patients who require VV ECMO. Continues renal replace-
ment therapy (CRRT) was continued through the ECMO
circuit. Baseline anthropometrics according to the parents
were weight of 190 kg (420 lb) and height of 66” (167.6
cm) with a calculated ideal body weight (IBW) of 64 kg
(141 lb) according to the Hamwi equation. On arrival,
the patient met criteria for class III obesity based on a
BMI of 68 kg/m2 . Nutrition-focused physical examination
revealed diffuse fluid accumulation and significant obesity
limiting the assessment of underlying lean body mass, and
an abdomen/pelvis computed tomography (CT) scan was
not available for review.
Energy and protein requirements are summarized in
Table 1. Permissive underfeeding needs were calculated
according to the 2016 American Society for Parenteral and
Enteral Nutrition (ASPEN) and Society of Critical Care
Medicine (SCCM) guidelines.12 Energy demand during VV
ECMO has yet to be established, therefore institutional
practice is to add an additional 10%–20% to the Penn State
equation with minute ventilation of 0 L/min. International
practice surveys have established 20–30 kcal/kg as the most
commonly used method of estimating energy needs on VV
ECMO13,14 ; therefore, this is calculated as a cross-reference.
Due to the novel nature of this patient, basal metabolic rate
was also calculated. The Mifflin St Jeor equation was used
because this was found to correlate acceptably with mea-
sured energy expenditure (MEE) in community-dwelling
patients with PWS.2 Indirect calorimetry is not possible
during ECMO because gas exchange occurs via the ECMO
circuitry and is minimal in the native lung. Methods to
address this change in the site of ventilation during ECMO
using values from preoxygenator and postoxygenator blood
Nutrition in Clinical Practice 33(6)
Institutional Practice Reference
Mifflin St Jeor equation
3800–5700 kcal/day 2833 kcal/day
3–4 g protein/kg IBW 0.8 g/kg actual weight
192–256 g/day 152 g/day
gasses or attaching a calorimeter to the oxygenator have
been published but lack wide validation.15-17 In addition,
these methods require equipment not currently available at
this institution and/or blood draws that are not part of
routine patient care. Protein needs were estimated in the very
high range at 3–4 g of protein/kg IBW related to elevated
demand during CRRT and VV ECMO support.
Continuous enteral nutrition (EN) was provided with
a 1.2 kcal/mL formula containing fish oil and 75 g/L of
protein via a nasojejunal small-bore feeding tube for a goal
of 1440 mL daily using a rate-based feeding approach.
This formula was selected primarily for its high-protein
and low-fiber content given the patient’s high protein de-
mands and critical illness.12 A protein modular providing 15
g/packet was provided as a bolus of 2 packets 4 times daily.
The combined nutrition regimen provided 1728 kcal/day
(27 kcal/kg IBW) and 228 g of protein/day (3.6 g of
protein/kg IBW). During the early stage of his admission,
the patient required sedation with propofol, which uses a
10% soybean–oil based fat emulsion as the carrier fluid and
provides 1.1 kcal/mL. The patient’s sedation requirements
provided an additional 600–900 kcal/day. To avoid over-
feeding during this time, the EN regimen was adjusted to
840 mL/day with 8 packets of protein modular (2 packets
administered 4 times daily) to provide 1008 kcal (16 Kcal/kg
IBW) and 183 g of protein (2.9 g of protein/kg IBW).
On hospital day 21 the patient’s clinical status had
improved to the point of no longer requiring VV ECMO
support. After the liberation of ECMO support, the patient
required mechanical ventilation for an additional 23 days.
Indirect calorimetry was completed on hospital day 27
using the Viasys Vmax ENCORE system. MEE was 2870
kcal/day and respiratory quotient was 0.74. This was 100%
of estimated basal energy expenditure using Mifflin St Jeor.
EN was adjusted to 1560 mL/day with 2 packets of protein
modular administered 4 times daily, increasing the nutrition
provision to 1872 kcal (29 kcal/kg IBW) and 237 g protein
(3.7 g protein/kg IBW), providing 65% of MEE according
to permissive underfeeding guidelines.12 Nitrogen balance
was not available due to the patient’s continued renal failure.
The amount of goal nutrition support received during the
patient’s hospitalization is summarized in Table 2.
During the remainder of the patient’s 53-day hospital
stay, his renal function recovered and no dialysis was
Pelekhaty and Menaker
Table 2. Nutrition Support Goals and Delivery During Hospitalization.
EN Daily Protein
Hospital Infusion Goal Modular
Days (mL) Goal Goal Energy Provision
2–11 840 8 packets 1008 kcal (16 kcal/kg
IBW)
12–26 1440 8 packets 1728 kcal (27 kcal/kg
IBW)
27–34 1560 8 packets 1872 kcal (29 kcal/kg
IBW)
EN, enteral nutrition; IBW, ideal body weight.
required upon discharge. In addition, the patient success-
fully weaned off mechanical ventilation on hospital day 44,
and he passed a speech and language pathology swallow
evaluation for a pureed diet with nectar thick liquids. At
the time of discharge to a subacute facility, the patient
was consuming adequate oral nutrition to meet his estimate
energy and protein needs, and no longer required EN
support. The patient’s weight on discharge from the hospital
was recorded as 185 kg (407 lb), which is 97% of baseline
weight.
Discussion
PWS is a genetic syndrome with a worldwide prevalence
estimated to be 1 in 10,000–30,000.1 Providers working in
intensive care settings may not be familiar with the features
and unique challenges presented by this patient population.
PWS presents as poor feeding and hypotonia in the infant,
which transitions to hyperphagia around 8 years of age and
progresses to obesity by adulthood.1 Additional manifes-
tations of the syndrome include developmental delays, be-
havioral disturbances, dysmorphic features, hypogonadism,
and short stature.1
Patients with PWS may not have the same risk for
comorbidities as patients with obesity who do not have
PWS. Current research suggests that patients with PWS
have improved insulin sensitivity and reduced frequency of
nonalcoholic fatty liver disease when compared with age-,
sex-, and weight-matched control subjects.2-4 Treatment of
hypogonadism with human growth hormone in childhood
appears to further improve the metabolic profile and body
composition in adults with PWS.5 Energy expenditure in
healthy adults with PWS is less7 than that of matched
controls, and disease-specific predictive equations are not
widely validated.2
A thorough literature review did not yield guidelines for
the management of critically ill patients with obesity related
to PWS, nor do specific guidelines exist for the nutrition
care of patients who require ECMO. The 2016 ASPEN
and SCCM guidelines recommend permissive underfeed-
895
% Goal
Goal Protein Provision Received
183 g of protein (2.9 g of protein/kg 86
IBW)
228 g of protein (3.6 g of protein/kg 86
IBW)
237 g of protein (3.7 g of protein/kg 88.5
IBW)
ing of critically ill adult patients with obesity (BMI >
30 mg/kg2 ).12 Permissive underfeeding refers to providing
less energy than required to meet total energy expenditure,
while providing sufficient protein to meet requirements.
This feeding strategy may reduce the risk for metabolic
complications from overfeeding, which include hypercarbia
and hyperglycemia.12 Despite improved insulin sensitivity
and reduced incidence of diabetes mellitus compared with
matched controls in patients with PWS, efforts should be
made to minimize hyperglycemia related to critical illness.
In addition, according to a case series of critical illness and
sudden death in PWS, PWS results in abnormal response
to hypoxia18 ; therefore, reducing respiratory stress from
overfeeding is an important nutrition management consid-
eration. Overfeeding can complicate ventilatory manage-
ment by causing CO2 retention12 ; it follows that overfeeding
should be avoided while patients require VV ECMO support
for profound respiratory failure. The 2016 ASPEN and
SCCM guidelines do not provide recommendations on the
nutrition care of patients receiving VV ECMO. Based on in-
ternational surveys, current practice among ECMO centers
is to provide 20–30 kcal/kg and 1.2–2 g protein/kg.13,14 The
practice at this institution incorporates cross-referencing
this kilocalorie per kilogram (kcal/kg) range with the Penn
State equation with modifications because of the alterations
in minute ventilation on VV-ECMO to reduce the risk for
overfeeding.
In the absence of disease-specific nutrition guidelines, the
practice at this institution is to rely on general guidelines
for the nutrition management of critically ill adult pa-
tients with additional monitoring, such as nitrogen balance
and indirect calorimetry. This enhanced monitoring aug-
ments the standard practice of routinely monitoring weight,
nutrition-focused physical examination, and the amount of
goal nutrition support received. Routine nutrition-focused
physical examination conducted during the patient’s stay
did not suggest the development of fat or muscle wasting
with permissive underfeeding; however, exams were limited
by both nonpitting edema and obesity. Imaging tools such
as CT scans and bedside ultrasound (US) may provide
896
additional methods to assess the nutrition status of critically
ill patients.19 CT or US in this patient may have revealed
fat or muscle loss that could not be identified through
physical examination because of class II obesity and the
presence of nonpitting edema on arrival. However, this
patient did not require an abdomen/pelvis CT on admission,
and assessing the L3 slice for body composition may not be
practical for routine use.19 Bedside US offers a convenient
and noninvasive method of incorporating imaging into
nutrition assessment; however, there is not currently con-
sensus on the proper technique, especially in obese and
edematous patients,19 and it cannot be used to diagnose
nutrition status at this time.
Conclusion
Hypocaloric, high-protein EN support was adequate to sup-
port this critically ill adult male with PWS while he required
ECMO and CRRT. Employing this feeding strategy for
the nutrition management of this patient did not result in
the development of malnutrition during his hospital stay
based on serial monitoring of weight and nutrition-focused
physical examination. More studies are needed to investi-
gate the optimal nutrition management and monitoring of
critically ill patients with PWS, especially those who require
life support technologies such as CRRT and ECMO.
Statement of Authorship
S Pelekhaty and J Menaker equally contributed to the con-
ception and design of the research; S Pelekhaty drafted the
manuscript. All authors critically revised the manuscript, agree
to be fully accountable for ensuring the integrity and accuracy
of the work, and read and approved the final manuscript.
References
1. Cassidy SB, Schwartz S, Miller JL, Driscoll DJ. Prader-Willi syndrome.
Genet Med. 2012;14(1):10-26. https://doi.org/10.1038/gim.0b013e
31822bead0.
2. Lazzer S, Grugni S, Tringali G, Srtoria A. Prediction of basal
metabolic rate in patients with Prader-Willi syndrome. Eur J Clin Nutr.
2016;70:494-498.
3. Lacroix D, Moutel S, Coupaye M, et al. Metabolic and adipose
tissue signatures in adults with Prader Willi syndrome: a model
of extreme adiposity. J Clin Endocrinol Metab. 2015;100(3):850-859.
https://doi.org/10.1210/jc.2014-3127.
4. Bedogni G, Grugni G, Nobili V, Agosti F, Saezza A, Sartorio
A. Is non-alcoholic fatty liver disease less frequent among women
with Prader-Willi syndrome? Obes Facts. 2014;7:71-76. https://doi.org/
10.1159/000358570.
5. Fintini D, Inzaghi E, Colajacomo M, et al. Non-alcoholic fatty
liver disease (NAFLD) in children and adolescents with Prader-Willi
syndrome (PWS). Pediatr Obes. 2016;11(3):235-238. https://doi.org/
10.1111/ijpo.12052.
Nutrition in Clinical Practice 33(6)
6. Coupaye M, Lorenzini F, Lloret-Linares C, et al. Growth hormone
therapy for children and adolescents with Prader-Willi syndrome is
associated with improved body composition and metabolic status in
adulthood. J Clin Endocrinol Metab. 2013;98(2):E328-E335. https://
doi.org/10.1210/jc.2012-2881.
7. Butler MG, Theodoro MF, Bittel DC, Donnelly JE. Energy ex-
penditure and physical activity in Prader-Willi syndrome: compari-
son with obese subjects. Am J Med Genet A. 2007;143A(5):449-459.
https://doi.org/10.1002/ajmg.a.31507.
8. Makdisi G, Wang I. Extra corporeal membrane oxygenation (ECMO)
review of a lifesaving technology. J Thorac Dis. 2015;7(7):E166-E176.
https://doi.org/10.3978/j.issn.2072-1439.2015.07.17.
9. Extracorporeal Life Support Organization. ECLS Registry Report.
January 2018. https://www.elso.org/Registry/Statistics.aspx. Accessed
September 25, 2017.
10. Kon ZN, Dahi S, Evans CF, et al. Class III obesity is not a contraindi-
cation to venovenous extracorporeal membrane oxygenation support.
Ann Thorac Surg. 2015;100(5):1855-1860.
11. Schmidt M, Zogheib E, Rozé H, et al. The PRESERVE mortality
risk score and analysis of long-term outcomes after extracorporeal
membrane oxygenation for severe acute respiratory distress syndrome.
Intensive Care Med. 2013;39(10):1704-1713.
12. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for
the provision and assessment of nutrition support therapy in
the adult critically ill patient: Society of Critical Care Medicine
(SCCM) and American Society of Parenteral and Enteral Nutri-
tion (A.S.P.E.N). JPEN J Parenter Enteral Nutr. 2016;40(2):159-211.
https://doi.org/10.1177/0148607115621863.
13. Bear DE, Haslam J, Camporota L, Shankar-Hari M, Barrett NA. An
international survey of nutrition practices in adult patients receiving
veno-venous ECMO. Intensive Care Med. Exp. 2015;3(suppl 1):A295.
https://doi.org/10.1186/2197-425X-3-S1-A295.
14. Nandwani V, Tauber A, McCarthy P, Herr D. 333. Nutrition practice
patterns in adult ECMO patients: results of an international survey.
Crit Care Med. 2014;42(12 suppl 1):A1440. https://doi.org/10.1097/
01.ccm.0000457830.75362.ed.
15. Wollersheim T, Frank S, Müller MC, et al. Measuring energy ex-
penditure in extracorporeal lung support patients (MEEP)—protocol,
feasibility and pilot trial. Clin Nutr. 2018;37(1):301-307. https://doi.org/
16. De Waele A, van Zwam K, Mattens S, et al. Measuring resting
energy expenditure during extracorporeal membrane oxygenation:
preliminary clinical experience with a proposed theoretical model.
Acta Anaesthesiol Scand. 2015;59(10):1296-1302. https://doi.org/
10.1111/aas.12564.
17. Li X, Yu X, Cheypesh A, Li J. Non-invasive measurements of en-
ergy expenditure and respiratory quotient by respiratory mass spec-
trometry in children on extracorporeal membrane oxygenation—a
pilot study. Artif. Organs. 2015;39:815-819. https://doi.org/10.1111/
aor.12465
18. Stevenson DA, Anaya TM, Clayton-Smith J, et al. Unexpected death
and critical illness in Prader-Willi syndrome: report of ten indi-
viduals. Am J Med Genet A. 2004;124A(2):158-164. https://doi.org/
10.1002/ajmg.a.20370.
19. Earthman CP. Body composition tool for assessment of adult mal-
nutrition at the bedside: a tutorial on research considerations and
clinical application. JPEN J Parenter Enteral Nutr. 2015;39(7):787-822.
https://doi.org/10.1177/0148607115595227.
Clinical Observations
Stoned—A Syndrome of D-Lactic Acidosis and Urolithiasis
Casey M. Berman, MD1 ; and Russell J. Merritt, MD, PhD2
Abstract
Short bowel syndrome (SBS) can lead to many complications related to the condition and its therapy. We describe 2 children with
SBS who we believe are the second and third patients documented to have experienced both D-lactic acidosis and urolithiasis. We
review aspects of these SBS complications and recent findings on the microbiome of patients with SBS that may predispose to these
complications. (Nutr Clin Pract. 2018;33:897–901)
Keywords
D-lactate; D-lactic acidosis; dysbiosis; hypocitraturia; hyperoxaluria; lactobacillus; pediatrics; probiotics; short bowel syndrome;
small bowel bacterial overgrowth; urolithiasis
Introduction
Short bowel syndrome (SBS) usually results from surgical
resection of a large segment of the small intestine as a
result of congenital or acquired diseases.1 Common causes
of SBS in children include necrotizing enterocolitis, midgut
volvulus, intestinal atresias, and abdominal wall defects.2
The incidence of SBS is estimated to be 24.5 per 100,000
live births.2 SBS may lead to serious common seque-
lae, including dehydration, electrolyte imbalance, malnutri-
tion, chronic diarrhea, small intestinal bacterial overgrowth
(SIBO), central line associated bloodstream infections, in-
testinal failure–associated liver disease, and gallstones.1,3
Although these conditions are well described, other serious,
but less common, complications may also arise from SBS. In
this report, we describe 2 SBS patients who developed both
D-lactic acidosis and urolithiasis. Consent was obtained for
publication of these case reports per local investigational
review board guidelines.
Case 1
The first patient is a boy who first presented with D-lactic
acidosis at the age of 16 months with a weight of 9.8 kg. He
was born prematurely at 33 weeks’ gestation and had SBS
as a result of in-utero midgut volvulus with bowel necrosis.
Following surgical intervention, he had 5 cm of duodenum
remaining with a duodeno-colonic anastomosis. Prior to
presentation with D-lactic acidosis, he was maintaining
an oral intake of 950 kcal of 26 kcal/oz elemental infant
formula/d in addition to 300 ml of parenteral nutrition (PN)
composed of 20% dextrose and 4% amino acids. The high
total intake reflected his very limited enteral absorption, but
good appetite. He had missed 2 nights of PN just prior to
Nutrition in Clinical Practice
Volume 33 Number 6
December 2018 897–901

C 2018 American Society for
Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10063
wileyonlinelibrary.com
admission. He was not taking a proton-pump inhibitor or
other medications aside from vitamins and iron.
He presented to the hospital with worsening diarrhea,
low-grade fever, and altered mental status. One day prior
to admission he became very unbalanced, falling on his side
while trying to crawl. He was also sleeping more than usual
and disinterested in playing. He was febrile to 100.8°F at
home prior to admission. On admission, he was afebrile
and all vital signs were within normal limits. On exam,
he was sleeping but arousable. No focal neurologic deficits
were found. His laboratory assessment was remarkable
for metabolic acidosis with a total carbon dioxide (CO2)
of 8 mMol/L (reference range 22–30 mMol/L) and an
anion gap of 22 (reference range 5–15) without elevation of
blood urea nitrogen (BUN) or creatinine. C-reactive protein
was elevated at 2.6 mg/d (reference range 0.0–0.9 mg/dL).
Lactate was within normal limits at 15.0 mg/dL (reference
range 6–16 mg/dL). D-lactate was sent due to concern
for D-lactic acidosis. He was started on broad-spectrum
intravenous (IV) antibiotics for the history of fever with
a central venous catheter. The D-lactate sample drawn on
From the 1 New York Presbyterian–Columbia University, New York,
New York, USA; and 2 Children’s Hospital Los Angeles, Keck School
of Medicine, University of Southern California, Los Angeles,
California, USA.
Financial disclosure: None declared.
Conflicts of interests: None declared.
This article originally appeared online on March 23, 2018.
Corresponding Author:
Russell J. Merritt, MD, PhD, Children’s Hospital Los Angeles, 4650
Sunset Boulevard, MS #78, Los Angeles, CA 90027, USA.
Email: rmerritt@chla.usc.edu
898
admission resulted several days later and was elevated at
10.43 mMol/L (reference range 0–0.25 mMol/L), consistent
with a diagnosis of D-lactic acidosis. Blood cultures grew
Staphylococcus epidermidis, and he continued a 14-day
course of IV vancomycin.
Following admission, he was started on IV fluids con-
taining acetate, and his feeds were held to decrease substrate
for intestinal D-lactate production. He was given neomycin
for presumed SIBO. A contrast upper gastrointestinal series
with follow through did not show bowel dilation or obstruc-
tion. By day 2 of hospitalization, his mental status returned
to baseline. He restarted oral feeds on day 4 of hospitaliza-
tion and was discharged home on day 5 to complete a 2-
week course of neomycin, on his prior home feeding, and
PN. Metronidazole and trimethoprim/sulfamethoxazole,
alternating at 2-week intervals, were restarted following
discontinuation of neomycin.
At 4 years of age, he presented to clinic with new onset
dysuria, lethargy, and loose stools and was admitted to the
hospital. He was afebrile and vital signs were within normal
limits. He was alert and oriented on physical exam, with
no significant findings. Urinalysis showed oxalate crystals
and hematuria without signs of infection. Chemistry was
notable for a total CO2 of 9 mMol/L, with an anion gap
of 19. The D-lactate drawn on admission was later reported
as 2.96 mMol/L. Renal ultrasound showed a 1.5-cm cal-
culus in the bladder with mild right-sided hydronephrosis.
A vitamin D concentration obtained 1 month later was
32 ng/mL (reference range 20–50 ng/mL). He was started on
oral neomycin. On hospital day 8 lithotripsy was performed.
Stone analysis was not completed. He was discharged home
and a repeat renal ultrasound 2 months later showed no
stones. He has not developed gallstones.
At the age of 5 years, while on his prophylactic antibiotic
regimen, he experienced another episode of apparent D-
lactic acidosis, but the diagnosis was not documented with
laboratory testing. Upon discharge, his cyclic prophylactic
antibiotic regimen was changed to include neomycin, and
he has not experienced a recurrence.
Case 2
The second patient is an adolescent girl admitted to the
hospital at 15 years of age with a history of SBS sec-
ondary to major bowel resection at age 13. She experienced
bowel necrosis related to a congenital mesenteric defect
and a complex internal hernia. She was left with 56 cm of
small bowel and a jejuno-cecal anastomosis. She underwent
elective cholecystectomy for symptomatic cholelithiasis 7
months later. She was partially dependent on PN for 1 year
following bowel resection. At the time of D-lactic acidosis
presentation, she was on a reduced oxalate oral diet, a
proton pump inhibitor, and metronidazole and trimetho-
Nutrition in Clinical Practice 33(6)
prim/sulfamethoxazole, alternating at 2-week intervals, for
SIBO prophylaxis.
She presented with acute-onset of slurred speech, bal-
ance disturbance, and vertigo. On arrival, she was afebrile
and her vital signs were within normal limits. Physical exam
was significant for left facial paralysis, dysarthria, and right-
sided dysmetria. She was acidotic with a total CO2 of
14 mMol/L and anion gap of 18. A serum specimen for
D-lactate level was sent. The neurology service was con-
sulted in the emergency room for evaluation for a possible
cerebral vascular event. Computerized tomographic scan
of the head showed no abnormalities. Magnetic resonance
imaging showed subtle areas of T2 hyperintensity and
restricted diffusion involving the right cerebellar hemisphere
and the right hippocampus. Given her presentation and
the suspicion of D-lactic acidosis, she was admitted to
the hospital, and IV bicarbonate, oral neomycin, and
a carbohydrate-free clear liquid diet were started. The
d-lactate value returned as 1.58 mMol/L. The patient was
discharged 4 days after admission to complete a 2-week
course of neomycin and was advised to stay on a low
simple-sugar diet. After finishing the course of neomycin,
she began a prophylactic regimen of alternating neomycin
and metronidazole.
One year later, at age 16, she presented with sudden-
onset right flank pain and on computerized tomographic
scan was found to have a 5 mm calculus at the right
ureterovesicular junction with mild hydronephrosis as well
as multiple 1–3 mm nonobstructive renal calculi bilaterally.
A temporary stent was placed to relieve symptoms. Several
weeks later, she underwent ureteroscopy with ureteral stent
removal and right retrograde pyelogram, and no persistent
stones were identified. Calcifications from the stent yielded
80% calcium oxalate monohydrate and 20% calcium oxalate
dihydrate. Five months later she had a similar episode
of a left-obstructing urinary stone, which passed without
intervention. A 24-hour urine stone profile was remarkable
for elevated oxalate excretion of 47 mg/d (normal 20–
40 mg/d), supersaturation of calcium oxalate, and remark-
ably low urinary excretion of citrate of 55 mg/d (normal
>550 mg/d)—all predisposing factors to stone formation.
The urine pH was 5.6. A vitamin D concentration ob-
tained about 3 months later was 45 ng/mL. She was in-
structed to drink at least 2.5 liters of fluids/d and started
on 1200 mg/d of calcium carbonate and on potassium-
citrate supplementation. In the weeks after starting this
regimen, she reported passing several small stones, but
since then has not had recurrent stones. On repeat 24-
hour urine stone analysis 4 months after initiation of
potassium citrate and calcium carbonate supplementation,
urinary citrate increased to 468 mg/d, but oxalate excretion
remained elevated at 117 mg/d. She has subsequently done
well on a low oxalate diet with calcium and potassium
citrate supplementation and a low simple-sugar diet with
Berman and Merritt 899

prophylactic alternating neomycin and metronidazole at 1- Table 1. Clinical Features Seen With D-Lactic Acidosis.
week intervals.
Clinical settings
Short bowel syndrome with colon in continuity
Discussion Malabsorption syndromes
Bariatric intestinal bypass surgery
To our knowledge, only 1 prior patient with SBS has been Neurologic manifestations
reported with both D-lactic acidosis and urolithiasis.4 We Altered behavior
believe these findings may be interrelated because of the Somnolence
increased intestinal bacteria lactate production and acetate Drunk appearing
Aggressive
catabolism in SBS dysbiosis and the effect of acidosis
Confused/disoriented/delirious
on renal citrate excretion as described in the following Neurologic examination findings
paragraphs. Gait disturbance/ataxia/clumsiness
Plasma D-lactate of >3 mMol/L is often used to make Weakness
the diagnosis of D-lactic acidosis.5 However, there is not Dysarthria
a good correlation between D-lactate concentration and Ptosis
symptom severity.6,7 SBS puts patients at increased risk for Nystagmus
Blurred vision
harboring large numbers of D-lactate, producing intestinal
Supportive laboratory findings
bacteria related to their reduced carbohydrate absorption, Metabolic acidosis
absence of an ileocecal valve, and SIBO with increased Elevated anion gap
fermentation of lumenal carbohydrates.5,8,9 Organic acid Negative urine ketones
production may exceed the buffering of pancreatic bicar- Normal L-lactate
bonate secretion.8,10 Lactate-producing bacteria, including
Lactobacillus fermenti, Lactobacillus acidophilus, and Lac-
tobacillus mucosae, thrive in a more acidic environment and Table 2. Management of D-Lactic Acidosis in Short Bowel
quickly become the predominant bacteria of the intestinal Syndrome.
flora of patients with SBS.5,8-12 The bacteria present in SBS
can produce large amounts of D-lactate.9 This dysbiosis Acute therapy
is also characterized by deficiencies of Bacteroides and the Intravenous base (acetate or bicarbonate)
Fasting or enteral carbohydrate restriction
Firmicutes, including Clostridia.11,13 Enteral antibiotics (neomycin may be a preferred drug)
The presentation of D-lactic acidosis resembles alcohol Chronic prevention of recurrences
intoxication (see Table 1 for list of common symptoms). Simple carbohydrate restriction as feasible
Encephalopathy appears to be universal on presentation.14 Cyclic antibiotic therapy; neomycin may be especially
Slurred speech is the most common symptom (52% of pa- helpful
tients) observed followed by ataxia (32%), gait disturbance Consider probiotics (not lactobacillus or D-lactate
(29%), weakness (16%), aggressive behavior (6%), feeling producing organisms)
Recalcitrant cases
drunk (3%), and blurry vision (3%).14 The etiology of the Consider:
central nervous system symptoms appears to be multifac- Surgical therapy of underlying anatomy facilitating small
torial. D-lactate is neurotoxic,15 and brain tissue lacks D- intestinal bacterial overgrowth
2-hydroxyacid dehydrogenase.15 Acidosis inhibits pyruvate Fecal transplantation
decarboxylation, which in turn reduces acetyl Coenzyme
A and adenosine triphosphate production and decreases
neurotransmitter production.15 The cerebellum, with less
pyruvate dehydrogenase, is more vulnerable to this effect.15 Although dietary modification alone may be effective,
Other potentially neurotoxic substances are also produced antibiotic treatment is commonly implemented. Neomycin
by enteric bacteria.16 Comorbid nutrition deficiencies in was used successfully in the first documented case of
SBS such as thiamin deficiency may also contribute to the D-lactic acidosis.21 Another case report of a 12-year-old
symptoms.14,17,18 boy with SBS and D-lactic acidosis supports the utility of
Treatment of D-lactic acidosis is described in Table 2 a combination of neomycin with transient withholding of
and begins with IV correction of academia.5,8 Withholding feedings15 as was done in our patients.
of enteral carbohydrates is helpful in acute management Probiotic administration for the treatment of D-lactic
to reduce D-lactate production.8,19 A 9-year-old girl was acidosis is controversial, as these patients already have
managed with dietary polymeric carbohydrate and an exclu- presumed SIBO.22 Some probiotics can colonize the in-
sion of simple sugars with resolution of recurrent D-lactic testine with flora incapable of D-lactate production. A 3-
acidosis.20 year-old boy with SBS and chronic symptoms of D-lactic
900
acidosis responded to probiotics that changed his fecal
flora from predominantly gram-positive to predominantly
gram-negative with a nearly complete elimination of uri-
nary D-lactate excretion.23 Fecal transplantation has also
been described as successful therapy for recurrent D-lactic
acidosis.24
Long-term therapy to prevent recurrences includes sim-
ple carbohydrate restriction, which may prove challeng-
ing in a young child. Prophylactic antibiotics to pre-
vent bacterial overgrowth appear to be helpful. Of inter-
est, both of our patients presented with D-lactic acido-
sis while they were taking alternating metronidazole and
trimethoprim/sulfamethoxazole prophylaxis. Neither had a
recurrence after neomycin was included in their rotating
antibiotic regimens.
Both of our patients with D-lactic acidosis experienced
urolithiasis. Normally, oxalate is bound by calcium in the
intestine, leading to the fecal excretion of calcium oxalate.4
Fat malabsorption in SBS patients leads to excess lumenal
fatty acids and formation of calcium-fatty acid soaps.25
Unbound oxalate is absorbed, and intestinal oxalate perme-
ability appears to be increased in SBS due to toxic mucosal
effects of malabsorbed free fatty acids and deconjugated bile
salts.4,25 The resultant hyperoxaluria can lead to urinary cal-
cium oxalate stones. In a general U.S. pediatric population,
renal stones were present in about 14.8 per 100,000 children
in 2003–2008.26 Although data specific to pediatric SBS pa-
tients do not appear to exist, an adult cohort study reported
that at a mean of 5 years of follow-up, nephrolithiasis was
present in 24% of SBS patients with a colon.27
In children with calcium oxalate stones, hyperoxaluria,
hypercalciuria, and hypocitraturia are common4 and were
documented in our female patient. A low oxalate diet, potas-
sium citrate supplementation, and calcium supplementation
are beneficial. Calcium therapy for hyperoxaluria deceases
oxalate absorption linearly with increasing dietary calcium
dose of up to 1200 mg/d.28
Hyperoxaluria in SBS may be related to intestinal dys-
biosis. Intestinal oxalate is degraded by bacterial enzymes,
not host enzymes.29 Oxalobacter formigenes, with enzymes
for oxalate degradation, is absent in patients with cal-
cium oxalate nephrolithiasis.29 O. formigenes is sensitive
to various antibiotics and may be reduced by persistent
antibiotic use, as given to many SBS patients. Urinary
citrate prevents the formation of calcium oxalate crystals
and stones. Hypocitraturia occurs in 19%–68% of patients
with urolithiasis.30,31
Diet is a factor in hypocitraturia, including low consump-
tion of magnesium and potassium.32 In 63 children with
urolithiasis, daily intake of magnesium and potassium and
urinary concentrations of magnesium and potassium were
lower in patients with hypocitraturia when compared with
those who had normal urinary citrate.32 Most individuals
with hypomagnesuria exhibit hypocitraturia.30,32 Fruits and
Nutrition in Clinical Practice 33(6)
Table 3. Management of Short Bowel Syndrome–Associated
Oxalate Urolithiasis.
Interventions that may be helpful after acute urologic
management of stone(s):
Analysis of stone composition
Quantification of urinary calcium, magnesium, oxalate, and
citrate
Assure good hydration
Diet rich in fruits and vegetables, as feasible
Low oxalate diet
Oral calcium supplementation
Oral potassium citrate supplementation to normalize
urinary citrate
Oral magnesium supplementation, as tolerated
Incompletely evaluated intervention
Probiotic therapy with Oxalobacter formigenes or other
oxalate degrading probiotic organism(s)
vegetables provide dietary potassium, and low fruit intake
was found in the hypocitraturic group. Addition of fruits
and vegetables to adult nephrolithiasis patients with hypoci-
traturia significantly increased urinary citrate.33 In addition,
potassium chloride supplementation led to significant in-
creases in citrate excretion,34 and magnesium salts increase
urinary pH, which enhances the excretion of citrate.
Intestinal dysbiosis found in SBS may contribute to
hypocitraturia. Some bacteria metabolize citrate.35 A study
comparing the microbiome of SBS patients vs healthy con-
trols found that the citrate cycle was amplified in the bacteria
of the SBS patients.13 Some lactic acid bacteria metabolize
citrate.36-40 It is likely that upregulation of citrate catabolism
depletes intestinal citrate and may lead to reduced urinary
citrate.
Hypocitraturia has also been attributed to acidosis in re-
nal tubular acidosis. Intracellular acidosis leads to decreased
urinary citrate excretion.32 This may occur with D-lactate
acidosis and offers a possible mechanism by which D-lactic
aciduria may predispose to urolithiasis. Interestingly, when
our first patient presented with urolithiasis, he had D-lactic
acidosis. Management of calcium oxalate stones in patients
with SBS is described in Table 3.
Conclusions
We identified 2 pediatric SBS patients who experienced both
symptomatic D-lactic acidosis and urolithiasis. D-lactic
acidosis should be considered in SBS patients who present
with central nervous system symptoms and hematuria
and flank pain suggest possible urolithiasis. Based on the
literature related to both of these SBS complications, we
speculate that the dysbiosis in a subset of SBS patients, likely
characterized in part by increased lactobacilli, may put
them at risk for both of these complications via increased
D-lactate production and increased citrate catabolism.
Berman and Merritt
Statement of Authorship
RJM contributed to the conception and design of the research.
CMB and RJM contributed to the acquisition and analysis of
the data. RJM and CMB contributed to the interpretation of
the data and CMB drafted of the manuscript. Both authors
critically revised the manuscript, agree to be fully accountable
for ensuring the integrity and accuracy of the wok and read and
approved the final manuscript.
References
1. Diamond IR, de Silva N, Pencharz PB, Kim JH, Wales PW. Neonatal
short bowel syndrome outcomes after the establishment of the first
Canadian multidisciplinary intestinal rehabilitation program: prelim-
inary experience. J Pediatr Surg. 2007;42(5):806-811.
2. Wales PW, de Silva N, Kim J, Lecce L, To T, Moore A. Neonatal
short bowel syndrome: population-based estimates of incidence and
mortality rates. J Pediatr Surg. 2004;39(5):690-695.
3. Vanderhoof JA, Langnas AN. Short-bowel syndrome in children and
adults. Gastroenterology. 1997;113(5):1767-1778.
4. Rahman N, Hitchcock R. Case report of paediatric oxalate urolithiasis
and a review of enteric hyperoxaluria. J Pediatr Urol. 2010;6(2):112-
116.
5. Fabian E, Kramer L, Siebert F, et al. D-lactic acidosis—case report and
review of the literature. Z Gastroenterol. 2017;55(1):75-82.
6. De Vrese M, Koppenhoefer B, Barth CA. D-lactic acid metabolism
after an oral load of DL-lactate. Clin Nutr. 1990;9:23-28.
7. Thurn JR, Pierpont GL, Ludvigsen CW, Eckfeldt JH. D-lactate
encephalopathy. Am J Med. 1985;79(6):717-721.
8. Bongaerts G, Bakkeren J, Severijnen R, et al. Lactobacilli and acidosis
in children with short small bowel. J Pediatr Gastroenterol Nutr.
2000;30(3):288-293.
9. Petersen C. D-lactic acidosis. Nutr Clin Pract. 2005;20(6):634-645.
10. Bongaerts GP, Tolboom JJ, Naber AH, et al. Role of bacteria in the
pathogenesis of short bowel syndrome-associated d-lactic acidemia.
Microb Pathog. 1997;22(5):285-293.
11. Mayer C, Gratadoux JJ, Bridonneau C, et al. Faecal D/L lactate ratio
is a metabolic signature of microbiota imbalance in patients with short
bowel syndrome. PLoS One. 2013;8(1):e54335.
12. Davidovics ZH, Carter BA, Luna RA, Hollister EB, Shulman RJ,
Versalovic J. The fecal microbiome in pediatric patients with short
bowel syndrome. JPEN J Parenter Enteral Nutr. 2016;40(8):1106-
1113.
13. Piper HG, Fan D, Coughlin LA, et al. Severe gut microbiota dysbiosis
is associated with poor growth in patients with short bowel syndrome.
JPEN J Parenter Enteral Nutr. 2017;41:1202-1212.
14. Kowlgi NG, Chhabra L. D-lactic acidosis: an underrecognized com-
plication of short bowel syndrome. Gastroenterol Res Pract. 2015:
476215.
15. Zhang DL, Jiang ZW, Jiang J, Cao B, Li JS. D-lactic acidosis sec-
ondary to short bowel syndrome. Postgrad Med Jl. 2003;79(928):110-
112.
16. Halperin ML, Kamel KS. D-lactic acidosis: turning sugar into acids in
the gastrointestinal tract. Kidney Intl. 1996;49(1):1-8.
17. Hudson M, Pocknee R, Mowat NAG. D-Lactic acidosis in short
bowel syndrome—an examination of possible mechanisms. Q J Med.
1990;74(274):157-163.
18. Puwanant M, Mo-Suwan L, Patrapinyokul S. Recurrent d-lactic aci-
dosis in a child with short bowel syndrome. Asia Pac J Clin Nutr.
2015;14(2):195-198.
19. Perlmutter DH, Boyle JT, Campos JM, Egler JM, Watkins JB. D-
Lactic acidosis in children: an unusual metabolic complication of small
bowel resection. J Pediatr. 1983;102(2):234-238.
901
20. Mayne AJ, Handy DJ, Preece MA, George RH, Booth IW. Dietary
management of D-lactic acidosis in short bowel syndrome. Arch Dis
Child. 1990;65(2):229-231.
21. Oh MS, Phelps KR, Traube M, Barbosa-Saldivar JL, Boxhill C, Carroll
HJ. D-Lactic acidosis in a man with the short-bowel syndrome. N Engl
J Med. 1979;301(5):249-252.
22. Munakata S, Arakawa C, Kohira R, Fujita Y, Fuchigami T,
Mugishima H. A case of D-lactic acid encephalopathy associated with
use of probiotics. Brain Dev. 2010;32(8):691-694.
23. Schoorel EP, Giesberts MA, Blom W, van Gelderen HH. D-lactic
acidosis in a boy with short bowel syndrome. Arch Dis Child.
1980;55(10):810-812.
24. Davidovics ZH, Vance K, Etienne N, Hyams JS. Fecal transplantation
successfully treats recurrent d-lactic acidosis in a child with short bowel
syndrome. JPEN J Parenter Enteral Nutr. 2017;41(5):896-897.
25. Bhasin B, Urekli HM, Atta MG. Primary and secondary hyperox-
aluria: understanding the enigma. World J Nephrol. 2015;4(2):235-244.
26. Dwyer ME, Krambeck AE, Bergstralh EJ, Milliner DS, Lieske JC, Rule
AD. Temporal trends in incidence of kidney stones among children: a
25-year population based study. J Urology. 2012;188:247-252.
27. Nightingale MD, Lennard-Jones JE, Gertner DJ, Wood SR, Bartram
CI. Colonic preservation reduces need for parenteral therapy, increases
incidence of renal stones, but does not change high prevalence of gall
stones in patients with a short bowel. Gut. 1992;33:1493-1497.
28. Holmes RP, Knight J, Assimos DG. Lowering urinary oxalate
excretion to decrease calcium oxalate stone disease. Urolithiasis.
2016;44(1):27-32.
29. Vaidyanathan S, von Unruh GE, Watson ID, Laube N, Willens S,
Soni BL. Hyperoxaluria, hypocitraturia, hypomagnesiuria, and lack
of intestinal colonization by Oxalobacter formigenes in a cervical
spinal cord injury patient with suprapubic cystostomy, short bowel, and
nephrolithiasis. Sci World J. 2006;6:2403-2410.
30. Kato Y, Yamaguchi S, Yachiku S, et al. Changes in urinary parameters
after oral administration of potassium-sodium citrate and magnesium
oxide to prevent urolithiasis. Urology. 2004;63(1):7-11.
31. Bevill M, Kattula A, Cooper CS, Storm DW. The modern metabolic
stone evaluation in children. Urology. 2017;101:15-20.
32. Kovacevic L, Wolfe-Christensen C, Edwards L, Sadaps M, Laksh-
manan Y. From hypercalciuria to hypocitraturia—a shifting trend in
pediatric urolithiasis? J Urol. 2012;188(4 suppl):1623-1627.
33. Meschi T, Maggiore U, Fiaccadori E, et al. The effect of fruits and
vegetables on urinary stone risk factors. Kidney Int. 2004;66(6):2402-
2410.
34. Domrongkitchaiporn S1, Stitchantrakul W, Kochakarn W. Causes of
hypocitraturia in recurrent calcium stone formers: focusing on urinary
potassium excretion. Am J Kidney Dis. 2006;48(4):546-554.
35. Edin-Liljegren A, Hedelin HH, Grenabo L, Pettersson S. Impact of
Escherichia coli on urine citrate and urease-induced crystallization.
Scanning Microsc. 1995;9(3):901-905.
36. Ostlie HM, Helland M, Narvhus JA. Growth and metabolism of
selected strains of probiotic bacteria in milk. Int J Food Microbiol.
2003;87(1-2):17-27.
37. Palles T, Beresford T, Condon S, Cogan TM. Citrate metabolism in
Lactobacillus casei and Lactobacillus plantarum. J Appl Microbiol.
1998;85(1):147-154.
38. Hugenholtz J. Citrate metabolism in lactic acid bacteria. FEMS
Microbiol Rev.1993;12(1-3):165-178.
39. Diaz-Muniz I, Banavara DS, Budinich MF, Rankin SA, Dudley EG,
Steele JL. Lactobacillus casei metabolic potential to utilize citrate as
an energy source in ripening cheese: a bioinformatics approach. J Appl
Microbiol. 2006;101(4):872-882.
40. Mortera, P, Pudlik A, Magni C, Alarcon S, Lolkema JS. Ca2+-citrate
uptake and metabolism in lactobacillus casei ATCC 334. Appl Enviorn
Microbiol. 2013;79(15):4603-4612.
Clinical Observations

Nutrition in Clinical Practice

Volume 33 Number 6
Avoidance of Overt Precipitation and Patient Harm December 2018 902–905

C 2017 American Society for

Following Errant Y-Site Administration of Calcium Chloride Parenteral and Enteral Nutrition
DOI: 10.1177/0884533617723865
and Parenteral Nutrition Compounded With wileyonlinelibrary.com
Sodium Glycerophosphate

Collin Anderson, PharmD, PhD, BCPS, BCPPS; Chanelle Stidham, PharmD, BCPS;
Sabrina Boehme, PharmD, BCPS, BCPPS; and Jared Cash, PharmD, BCPS

Abstract
Calcium phosphate precipitates present 1 of many challenges associated with parenteral nutrition (PN) compounding. Extensive
research has led to the establishment of solubility curves to guide practitioners in the prescription and preparation of stable PN.
Concurrent dosing of intravenous products via y-site administration with PN can alter the chemical balance of the solution
and modify solubility. Medications containing calcium or phosphate should not be administered in the same line as PN, due
to the high potential for precipitation. Herein a case is reported from a pediatric cardiac intensive care unit where a physician
ordered the administration of calcium chloride. The bedside nurse added the calcium chloride intermittent infusion as a y-site
administration with the patient’s PN. The patient’s PN had been compounded with sodium glycerophosphate, temporarily available
in the United States during a sodium phosphate shortage. The patient did not experience any observable adverse effects from the
y-site administration with PN. Following this event, the scenario was replicated to investigate any precipitation risk associated with
the y-site administration. Additionally, a separate PN solution containing sodium phosphate rather than glycerophosphate was
compounded and used in a laboratory setting to demonstrate the potential for harm had the patient’s PN been compounded with an
inorganic phosphate source. This replication of the error demonstrates the additional safety gained in relation to precipitation risk
when PN solutions are compounded with sodium glycerophosphate in place of sodium phosphate. (Nutr Clin Pract. 2018;33:902–
905)

Keywords
drug compounding; food-drug interactions; nutritional support; parenteral nutrition

Calcium phosphate precipitation in compounded parenteral
nutrition (PN) products has been well documented and
researched. Deaths attributed to calcium phosphate pre-
cipitates administered within the context of PN led to a
U.S. Food and Drug Administration safety alert and are
referenced in early recommendations by a national advi-
sory group regarding safe practices for PN formulations.1,2
Extensive research established solubility curves for varying
PN compositions. Due to their everyday utility, these graphs
were compiled in tertiary references and handbooks for easy
access to practitioners involved in prescribing, preparing,
and administering PN. Pharmacists initially referred to
printed graphs for each PN prescription to determine if
the formulation would likely precipitate. More recently,
compounding software has incorporated those graphs into
platforms that automatically suggest when a proposed PN
product may lead to precipitation. Immediate warnings
are provided to the compounding personnel, allowing for
adjustments in the formulation prior to production. Many
factors contribute to PN solubility. Primary among these
are calcium and phosphate concentrations, form of calcium
utilized, amino acid product and concentration, pH, and
temperature.3 Other precipitates, such as zinc phosphate,
can occur in PN. In neonatal and pediatric patient pop-
ulations, practitioners can be limited in their ability to
provide adequate amounts of calcium and phosphorus due
to volume constraints in the setting of increased mineral
needs. Generally, significant time and resources are devoted
to ensuring that PN preparations are safe and do not pose a
From Intermountain Healthcare Primary Children’s Hospital, Salt
Lake City, Utah, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on September 27, 2017.
Corresponding Author:
Collin Anderson, PharmD, PhD, BCPS, BCPPS, Pharmacy,
Intermountain Healthcare Primary Children’s Hospital, 100 N Mario
Capecchi Dr, Salt Lake City, UT 84113, USA.
Email: collin.anderson@imail.org
Anderson et al
Figure 1. Sodium glycerophosphate.
significant risk for forming calcium phosphate precipitates.
Y-site administration of any other solutions containing
calcium or phosphate with PN should be strictly avoided,
due to the extreme concentrations encountered during such
administration. It is important to remember that in addition
to standard mineral intravenous (IV) replacement products,
other medications may contain phosphate in their formula-
tion (eg, dexamethasone sodium phosphate).
Sodium glycerophosphate (Figure 1) is an organically
bound phosphate that has a significantly improved calcium
phosphate solubility profile in comparison with inorganic
phosphate preparations. Phosphate is bound to glycerol
within the organic phosphate salt. Upon IV infusion, the
phosphate is readily cleaved from the glycerol component
through an enzymatic process. The pharmacokinetic pro-
files of sodium phosphate and sodium glycerophosphate are
nearly identical, and the products are both valid options as
a phosphate source.4 In terms of solubility in the presence
of calcium, sodium glycerophosphate has been shown to
be superior. PN solutions containing glycerophosphate are
less likely to precipitate and allow for greater flexibility
in calcium and phosphate dosing. European approval of
the drug first occurred nearly 2 decades ago, and the
product is readily available in those markets. Due to pre-
vious inorganic phosphate shortages within the U.S. drug
distribution system, the Food and Drug Administration al-
lowed for temporary importation and utilization of sodium
glycerophosphate. Studies of sodium glycerophosphate and
calcium were performed following methodology originally
used to determine PN calcium phosphate solubility curves.
Research has demonstrated a remarkable solubility profile
for sodium glycerophosphate in solutions with calcium-
containing salts, essentially eliminating concern for precipi-
tation in the range of clinically relevant PN preparations.5,6
Situation
A 3.1-kg male was admitted to the cardiac intensive care
unit of our free-standing pediatric institution following
open heart surgery. The patient received PN according to
an institutional cardiac nutrition protocol. Blood calcium
levels were persistently low (ionized calcium <1.2 mmol/L),
requiring correction with IV calcium chloride. Approxi-
mately 12 hours after PN was started, an order was placed
903
for 90 mg of IV calcium chloride, which was obtained
from an automated dispensing cabinet and administered
via y-site administration with PN downstream from the
filter. After administration, during nurse hand-off at shift
change, the oncoming nurse immediately recognized the
error that occurred in administering calcium with PN and
contacted the unit clinical pharmacist. Bedside practitioners
did not observe any adverse effects surrounding the y-site
administration of the calcium chloride and PN. It was noted
that the patient’s PN contained sodium glycerophosphate as
its sole phosphate source.
Method
The situation was re-created to study the difference in y-
site administration of calcium chloride with the patient’s PN
formulation vs a similar PN containing inorganic phosphate
in place of sodium glycerophosphate. The original PN con-
tained 49.8 mEq/L of calcium gluconate and 21.5 mmol/L
of sodium glycerophosphate, which is within published
solubility limits for these agents. This PN was compounded
again to re-create the y-site administration that occurred
in the intensive care unit. An identical PN containing
inorganic phosphate in place of glycerophosphate could
not be compounded at the same concentrations, because it
would exceed the solubility limit. Therefore, a similar in-
organic phosphate-containing PN was compounded within
the solubility curve limit by reducing calcium gluconate
to 39.6 mEq/L and sodium phosphate to 17.1 mmol/L
(Table 1). With an Alaris infusion pump, each PN was run
at the ordered rate of 4.8 mL/hr through a CareFusion (San
Diego, CA) 0.2-μm filter infusion set. Downstream from the
filter, a bifuse administration set was used to run 90 mg of
calcium chloride, 10%, over 30 minutes via a separate pump.
The contents of the catheter output for both PN solutions
were collected over the time of administration and then im-
mediately passed through a 0.8-μm nitrocellulose filter disk.
The filter from the y-site infusion of the sodium phosphate–
containing PN showed particulate matter presumed to be
calcium phosphate, which was visible to the naked eye.
The filter from the y-site infusion of the glycerophosphate-
containing PN contained 6 crystals, which were seen upon
microscopic evaluation (Figure 2).
Discussion
A sentinel event was averted due to the favorable compatibil-
ity profile of sodium glycerophosphate. Calcium phosphate
pulmonary emboli were implicated in the deaths of 2 indi-
viduals in the 1990s. The same publication that reported the
deaths replicated the implicated PN formulation. Infusion
of the PN in healthy pigs caused death within 4 hours.7
Simulation of the scenario that occurred at our hospital
likewise demonstrates the grave danger associated with
exceeding calcium phosphate solubility limits. Precipitation
904 Nutrition in Clinical Practice 33(6)

Table 1. Original PN Composition With Glycerophosphate Versus Inorganic Phosphate–Containing PN.

Original PN-Containing Glycerophosphate Experimental PN-Containing Sodium Phosphate

Component Weight-Based Dose Concentration Weight-Based Dose Concentration

Dextrose 7.4 g/kg 20% 9.3 g/kg 20%

Trophamine 1.8 g/kg 4.9% 1.8 g/kg 3.9%
Calcium gluconate 1.85 mEq/kg 50 mEq/L 1.85 mEq/kg 40 mEq/L
Sodium glycerophosphate 0.8 mmol/kg 22 mmol/L NA NA
Sodium phosphate NA NA 0.8 mmol/kg 17 mmol/L
Magnesium sulfate 0.4 mEq/kg 10.8 mEq/L 0.4 mEq/kg 8.6 mEq/L
Sodium acetate 0.65 mEq/kg 17 mEq/L 0.85 mEq/kg 18 mEq/L
Sodium chloride 0.65 mEq/kg 17 mEq/L 0.85 mEq/kg 18 mEq/L
Carnitine 10 mg/kg 0.27 g/L 10 mg/kg 0.21 g/L
Total volume 37 mL/kg 115 mL/d 47 mL/kg 145 mL/d

NA, not applicable; PN, parenteral nutrition.

Figure 2. A area of the filter (3 × 3 mm2 ) pictured under 30× magnification. The simulated y-site administration of the sodium
glycerophosphate–containing parenteral nutrition with calcium chloride (left) and the analogous inorganic phosphate–containing
parenteral nutrition with calcium chloride (right).

of the y-site admixture was clearly observable without
magnification.
Research has demonstrated that within PN formulations,
sodium glycerophosphate and calcium chloride are solu-
ble at concentrations up to 50 mmol/L and 50 mEq/L,
respectively.5,6 Of note, the concentration of calcium and
phosphate in the replicated y-site admixture was calculated
to be 418 mEq/L of calcium and 15.7 mmol/L of phosphate,
far exceeding concentrations in published solubility studies.
The filtered output from the sodium glycerophosphate–
containing PN and y-site calcium chloride did result in
6 crystals that were discovered through microscopic ex-
amination of the filtered solution. As controlled physical
compatibility studies involving sodium glycerophosphate
and calcium chloride in compounded PN have not been
performed at the replicated y-site concentration reported
herein, it is not advisable to exceed previously published
limits.
Like other institutions, our hospital has had long-
standing policies, procedures, and education regarding
the incompatibility surrounding y-site administration of
calcium and/or phosphate solutions with PN, as recom-
mended by the American Society for Parenteral and Enteral
Nutrition.8,9 Electronic and print resources are available to
nursing personnel, and pharmacists are regularly consulted
regarding compatibility. In the case reported, human err
bypassed those safeguards, resulting in potential danger.
Fortunately, the patient’s PN was compounded with sodium
Anderson et al
glycerophosphate, and there was no apparent harm to the
patient. The favorable solubility profile of sodium glyc-
erophosphate vs inorganic phosphate leads to a better safety
profile in regard to the formation of harmful precipitates
in PN. While PN remains a dangerous medication even
when compounded with sodium glycerophosphate, the risk
of calcium phosphate precipitation is dramatically lower
than in standard PN solutions compounded with inorganic
phosphate.
Conclusion
When compared with sodium phosphate, sodium glyc-
erophosphate minimizes risk of calcium phosphate pre-
cipitation within PN. Making sodium glycerophosphate
available within the United States would decrease the pre-
cipitation risks associated with PN. It is recommended
that sodium glycerophosphate be used in place of sodium
phosphate in PN.
Statement of Authorship
All authors contributed to the conception and design of the
research; C. Anderson and C. Stidham contributed to the
acquisition, analysis, and interpretation of the data; and C.
Anderson drafted the manuscript. All authors critically revised
the manuscript, agree to be fully accountable for ensuring the
905
integrity and accuracy of the work, and read and approved the
final manuscript.
References
1. McKinnon BT. FDA safety alert: hazards of precipitation associated
with parenteral nutrition. Nutr Clin Pract. 1996;11:59-65.
2. Mirtallo J, Driscoll D, Helms R, Kumpf V, McKinnon B. Safe practices
for parenteral nutrition formulations. JPEN J Parenter Enteral Nutr.
1998;22:49-66.
3. Newton D, Driscoll D. Calcium and phosphate compatibility: revisited
again. Am J Health Syst Pharm. 2008;65:73-80.
4. Topp H, Hochfeld O, Bark S, et al. Glycerophosphate is interchangeable
with inorganic phosphate in terms of safety and serum pharmacokinet-
ics. Pharmacology. 2011;88:193-200.
5. MacKay M, Anderson C. Physical compatibility of sodium glycerophos-
phate and calcium gluconate in pediatric parenteral nutrition solutions.
JPEN J Parenter Enteral Nutr. 2015;39:725-728.
6. Anderson C, MacKay M. Physical compatibility of calcium chloride
and sodium glycerophosphate in pediatric parenteral nutrition solutions.
JPEN J Parenter Enteral Nutr. 2016;40:1166-1169.
7. Hill SE, Heldman LS, Goo EDH, Whippo PE, Perkinson JC. Fatal
microvascular pulmonary emboli from precipitation of a total nutrient
admixture solution. JPEN J Parenter Enteral Nutr. 1996;20:81-87.
8. Ayers P, Adams S, Boullata J, et al. ASPEN parenteral nutrition
safety consensus recommendations. JPEN J Parenter Enteral Nutr.
2014;38:296-233.
9. Boullata JI, Gilbert K, Sacks G, et al. ASPEN clinical guidelines:
parenteral nutrition ordering, order review, compounding, labeling, and
dispensing. JPEN J Parenter Enteral Nutr. 2014;38:334-377.
Standards of Practice

Nutrition in Clinical Practice

Volume 33 Number 6
Standards for Nutrition Support: Adult Hospitalized Patients December 2018 906–920

C 2018 American Society for

Parenteral and Enteral Nutrition

DOI: 10.1002/ncp.10204
wileyonlinelibrary.com
Andrew Ukleja, MD, AGAF1 ; Karen Gilbert, RN, MSN, CNSC, ACNP2 ;
Kris M. Mogensen, MS, RD-AP, LDN, CNSC3 ; Renee Walker, MS, RD, LD, CNSC,
FAND4 ; Ceressa T. Ward, PharmD, BCPS, BCNSP, BCCCP5 ; Joe Ybarra, PharmD,
BCNSP6 ; Beverly Holcombe, PharmD, BCNSP, FASHP, FASPEN7 ; Task Force
on Standards for Nutrition Support: Adult Hospitalized Patients, the American Society
for Parenteral and Enteral Nutrition

Abstract
The American Society for Parenteral and Enteral Nutrition defines standards as benchmarks representing a range of performance
of competent care that should be provided to assure safe and efficacious nutrition care in most circumstances. Standards
are documents that define the structure needed to provide competent care. These Standards for Nutrition Support for Adult
Hospitalized Patients are an update of the 2010 Standards. These practice-based standards are intended for use by healthcare
professionals charged with the care of adult hospitalized patients receiving nutrition support therapy in any hospital with or
without a formal nutrition support service or team. These Standards address professional responsibilities as they relate to patient
assessment, diagnosis, education, care plan development, implementation, clinical monitoring, evaluation, and professional issues
around nutrition support. (Nutr Clin Pract. 2018;33:906–920)

Keywords
enteral nutrition; hospitalization; nutrition assessment; nutrition support; parenteral nutrition; standard of care

Introduction usually address professional responsibilities as they relate to

The American Society for Parenteral and Enteral Nu-
trition (ASPEN) is dedicated to improving patient care
by advancing the science and practice of clinical nutri-
tion and metabolism. Founded in 1976, ASPEN is an
interdisciplinary organization whose members are involved
in the provision of clinical nutrition therapies, including
parenteral and enteral nutrition. With more than 6,500
members from around the world, ASPEN is a community of
dietitians, nurses, pharmacists, physicians, scientists, stu-
dents, and other health professionals from every facet of
nutrition support clinical practice, research, and educa-
tion. ASPEN envisions an environment in which every
patient receives safe, efficacious, and high-quality nutrition
care. ASPEN’s mission is to improve patient care by
advancing the science and practice of clinical nutrition
and metabolism. These Standards for Nutrition Support
patient assessment, diagnosis, education, care plan devel-
opment, implementation, clinical monitoring, evaluation,
and professional issues. ASPEN publishes discipline-based
(eg, dietitian, nurse, pharmacist, or physician) and practice-
based (eg, adult hospitalized patients, pediatric hospitalized
patients, home and alternate site care) standards. Standards
are presented in the most generic terms possible. The details
of specific tests, therapies, and protocols are left to the
From the 1 Beth Israel Deaconess Medical Center, Division of
Gastroenterology, Boston, Massachusetts, USA; 2 Thomas Jefferson
University Hospital, Philadelphia, Pennsylvania, USA; 3 Department
of Nutrition, Brigham and Women’s Hospital, Boston,
Massachusetts, USA; 4 Michael E. DeBakey Veteran Affairs Medical
Center, Houston, Texas, USA; 5 Emory Healthcare, Atlanta, Georgia,
USA; 6 Medical City McKinney, McKinney, Texas, USA; and
7 American Society for Parenteral and Enteral Nutrition, Silver

for Adult Hospitalized Patients are an update of the 2010
standards.1 They are intended for use by any hospital with
or without a formal nutrition support service (or team).
ASPEN defines standards as benchmarks representing
a range of performance of competent care that should be
provided to assure safe and efficacious nutrition care in
most circumstances.2 Standards are documents that define
the structure needed to provide competent care. Standards
Spring, Maryland, USA.
Financial disclosure: None declared.
Conflicts of interest: None declared.
This article originally appeared online on October 15, 2018.
Corresponding Author:
Beverly Holcombe, PharmD, BCNSP, FASHP, FASPEN, ASPEN,
8401 Colesville Rd, Suite 510, Silver Spring, MD 20910.
Email: beverlyh@nutritioncare.org
Ukleja et al
discretion of individual healthcare facilities. Each health-
care facility shall strive to provide the best nutrition support
care that is possible given the resources of the organization.
The standards aim to ensure sound and efficient nutrition
care for those in need of nutrition support therapy.
Important Note
These standards do not constitute medical or other pro-
fessional advice and should not be taken as such. To the
extent that the information published herein may be used
to assist in the care of patients, this is the result of the sole
professional judgment of the attending healthcare profes-
sional whose judgment is the primary component of quality
medical care. The information presented in these standards
is not a substitute for the exercise of such judgment by the
healthcare professional. Circumstances in clinical settings
and patient indications may require actions different from
those recommended in this document and in those cases, the
judgment of the treating professional should prevail.
Audience for Standards
These practice-based standards are intended for use by
healthcare professionals charged with the care of adult
hospitalized patients receiving nutrition support therapy.
Level of Care
As limited by the Important Note above, these Standards
of Practice present a range of performance of competent
care that should be provided by healthcare professionals
caring for adult hospitalized patients receiving nutrition
support therapy. Terminologies included in each standard
are specified as:
(a) “Shall”: Indicates standards to be followed strictly.
(b) “Should”: Indicates that among several possibilities
one is particularly suitable, without mentioning or
excluding others, or that a certain course of action
is preferred but not necessarily required.
(c) “May”: Indicates a course of action that is permis-
sible within the limits of recommended practice.
These standards have been developed by the ASPEN
Task Force on Standards for Nutrition Support: Adult
Hospitalized Patients, reviewed by the ASPEN Clinical
Practice Committee, and approved by the ASPEN Board
of Directors on July 25, 2018. These Standards of Prac-
tice should be used in conjunction with the previously
published ASPEN Clinical Guidelines, Standards, Position
Papers, and other Board Approved documents, which can
be accessed at the ASPEN Documents Library, http://www.
nutritioncare.org/Clinical_Practice_Library/.
907
Chapter I: Organization
Standard 1. Nutrition Support Service (or
Team)
A nutrition support service (or team) should assess and
in collaboration with patients’ primary teams, manage the
nutrition support therapy of patients who require or may
require nutrition support therapy. These patients are of-
ten, but not always, determined to be nutritionally-at-risk
at admission or upon subsequent evaluation.3 Organized
nutrition support services (or teams) are associated with
improved patient outcomes, decreased length of hospital-
ization, and improved cost effectiveness.4-19 If a hospital
does not have a designated nutrition support service (or
team), the care used to provide nutrition support therapy
should be interprofessional. The scope and design of the
nutrition support service (or team) and their respective
activities vary according to the unique attributes of each
hospital. Among various organizations, management of
nutrition support may comprise a spectrum of activities
including no formal structure, an administrative nutrition
committee only, a consultative nutrition support service (or
team), or a nutrition support service (or team) that assumes
responsibility for the nutrition care of patients who receive
nutrition support therapy.
1.1 When an organized nutrition support service (or
team) exists, it shall be directed by a clinician who
has appropriate education, specialized training, pa-
tient care experience, or experience in managing
nutrition support services (teams).
1.2 An organized nutrition support service (or team)
should include a physician, nurse, dietitian, and
pharmacist, each following the standards of prac-
tice for their discipline, as available.20-23
1.3 If a nutrition support service (or team) is not
established, nutrition support therapy should be
managed with an interprofessional approach that
includes the patient’s physician, nurse, dietitian,
and pharmacist.
Standard 2. Policies and Procedures
Written policies and procedures for providing nutrition
support therapy shall be current.
2.1 The policies and procedures shall be developed with
the input of and review by all members of the
nutrition support service (or team) and/or nutrition
support committee.
2.2 The policies and procedures shall be reviewed
periodically and revised as appropriate to define
optimal patient care and therapeutic outcomes. (See
3.2.)
908
Standard 3. Performance Improvement
The nutrition support service (or team) and/or nutrition
support committee shall regularly review and report on ser-
vice performance, quality indicators, patient outcome data,
and adverse events related to nutrition support therapies.24
These reports shall be shared with all internal stakeholders
and reported to external agencies as required.
3.1 The nutrition support service (or team) and/or nu-
trition support committee shall recommend policy,
procedure, or protocol changes that improve and/or
enhance the safety and efficacy of nutrition support
therapy.
3.2 The review of service performance should assess
the appropriateness and effectiveness of nutrition
support therapy.
Chapter II: Nutrition Care
Nutrition care and the administration of nutrition support
therapy shall proceed according to a series of steps with
feedback loops. These steps include nutrition screening,
formal nutrition assessment, creation of a nutrition care
plan, implementation of the plan, patient monitoring, eval-
uation of the plan, evaluation of the care setting, and
reformulation of the plan or termination of therapy. (See
Figure 1: ASPEN Adult Nutrition Care Pathway.)
Standard 4. Nutrition Screening
Nutrition screening is defined as “a process to identify
an individual who is malnourished or who is at risk for
malnutrition to determine if a detailed nutrition assessment
is indicated.”1 Patients who are nutritionally-at-risk shall be
identified by a validated screening process and by periodic
rescreening per institutional policy or standard.3,25-33 This
process should be created, approved, and regularly reviewed
by a group with organizational authority, preferably a
designated nutrition committee.
4.1 Results of the nutrition screening shall be docu-
mented and communicated and appropriate inter-
vention shall be initiated within the time frame
specified by the hospital or as clinically indicated.
4.2 A procedure for rescreening of patients not imme-
diately identified as nutritionally-at-risk should be
implemented and regularly reviewed.
Standard 5. Nutrition Assessment
All patients identified as nutritionally-at-risk based
on the nutrition screening shall undergo a nutrition
assessment.3,27-33 This nutrition assessment shall be
documented and made available to all patient care providers.
The intent of the nutrition assessment is to document
Nutrition in Clinical Practice 33(6)
baseline nutrition parameters, identify nutrition risk
factors and specific nutrition deficits, determine individual
nutrition needs, and identify medical, psychosocial, and
socioeconomic factors that may influence the prescription
and administration of nutrition support therapy.34,35
5.1 The nutrition assessment shall be performed within
the time frame specified by the hospital and by a
dietitian or a clinician with documented specialized
expertise in nutrition.
5.2 The nutrition assessment shall include evaluation
of the patient’s current nutrition status and nutri-
tion requirements.
5.2.1 A malnutrition diagnosis, if present, and
degree of malnutrition shall be clearly doc-
umented to facilitate appropriate diagnosis
coding.
5.2.2 Degree of obesity (ie, class I, class II, or class
III), if applicable, shall also be documented.
5.3 The patient’s nutrition requirements shall be sum-
marized based on the findings of the nutrition
assessment and should include energy, macronutri-
ent (protein, and as appropriate, carbohydrate and
fat), as well as fluid, electrolyte, and micronutrient
requirements, as appropriate.
5.4 Nutrition assessment shall include a review and
documentation of factors relevant to delivery of
nutrition support therapy. Relevant factors may
include, but are not limited to, the following: ability
to eat safely and adequately, patient’s goals, assess-
ment of aspiration risk, functional status of the
gastrointestinal tract, cognitive function/abilities,
enteral and vascular access, and results of tests and
invasive procedures.
Chapter III: The Nutrition Care Plan
Standard 6. Goals
The process of nutrition care is multifactorial and shall
include multiple levels of intervention including screening
for nutrition risk factors. The nutrition care plan shall
be created from a comprehensive review and analysis of
information gathered from many aspects of the patient’s
care. The nutrition care plan should include “statements
of nutrition goals and monitoring/evaluation parameters,
the most appropriate route of administration of nutrition
therapy, method of nutrition access, anticipated duration of
therapy, and training and counseling goals and methods.”2
A formal nutrition assessment provides the basis for
the nutrition care plan. The nutrition care plan guides
comprehensive nutrition therapy by defining its ratio-
nale, describing appropriate intervention and monitoring,
and delineating recommended reassessment and reevalu-
ation parameters. This process facilitates changes in care
Ukleja et al 909

Figure 1. The American Society for Parenteral and Enteral Nutrition adult nutrition care pathway.

appropriate to the clinical setting while considering the
continuum of care. Revision of the nutrition care plan based
on changes in clinical status and achievement of goals of
therapy should occur before discontinuation of nutrition
support therapy.
Standard 7. Interprofessional Approach
The nutrition care plan should be developed using an inter-
professional team approach involving the patient, caregiver
(if applicable), the nutrition support service (or team), the
910 Nutrition in Clinical Practice 33(6)

Figure 1. Continued.
Ukleja et al
patient’s physician(s), dietitian(s), nurse(s), pharmacist(s),
and other appropriate healthcare professionals.
Standard 8. Patient and Caregiver
Communication
The nutrition care plan should include patient and/or care-
giver(s) education about nutrition support therapy, goals,
and expectations and should incorporate the wishes of the
patients and/or caregiver(s). Appropriate routes of admin-
istration shall be defined, identification of intake goals shall
be included, and estimated duration of therapy as well as
criteria for discontinuation of therapy should be addressed.
Standard 9. Selection of Route
The route selected to provide nutrition support therapy shall
be appropriate to the patient’s clinical status or condition
and shall periodically be assessed for continued appropri-
ateness as well as for its adequacy in meeting goals of the
nutrition care plan.36 (See Figure 2: Route of administration
algorithm.)
Standard 10. Selection of Formulation
The enteral nutrition (EN) or parenteral nutrition (PN)
formulation shall be appropriate for the patient’s disease
process and compatible with the route of access.37,38
10.1 The EN or PN formulation shall be adjusted as ap-
propriate based on the patient’s clinical response.
10.2 The EN or PN formulation shall be adjusted
accordingly when significant amounts of nutrients
are provided (eg, parenteral infusions, medica-
tions) through means other than the EN for-
mula or PN admixture or lost/eliminated through
mechanical procedures or anatomical defects
(eg, renal replacement therapy, enterocutaneous
fistula) .
Chapter IV: Implementation
Standard 11. Ordering Process
Implementation of the nutrition care plan shall follow
nutrition assessment and development of a formal nutrition
care plan.
11.1 Authority to prescribe nutrition support therapy
shall be determined by hospital policy and appli-
cable professional licensure laws.
11.1.1 Hospital policy should clearly articu-
late the appropriate credentials, training,
and/or certifications and competencies re-
quired for clinicians who prescribe nutri-
tion support therapy.
911
11.1.2 As delineated by clinical privileges and
applicable professional licensure laws, a
nutrition support clinician may enter/write
orders for feeding formulations, laboratory
tests, and adjunctive therapy (eg, intra-
venous [IV] fluids, insulin, IV/oral elec-
trolytes) and adjust regimens based on
response to therapy, changing clinical con-
dition(s), altered laboratory values, and
nutrition assessment parameters.
11.1.3 Hospital policy should include a com-
petency for nutrition support therapy
prescribing.39
11.2 Orders for nutrition support therapy shall be
documented in the patient’s medical record before
administration.
11.2.1 A standardized order format and review
process for nutrition support therapies
orders shall be used to minimize the
risk of adverse events and error. This
process shall include standardized elec-
tronic orders (eg, computerized provider
order entry [CPOE] system) for prescrib-
ing PN and EN. Handwritten order to
prescribe PN and EN should be avoided
due to potential for error. Verbal and
telephone orders and text messaging of
orders should be avoided.40,41 See ASPEN
Parenteral Nutrition Safety Consensus
Recommendations39 and ASPEN Safe
Practices for Enteral Nutrition Therapy40
for details of the prescribing process for
PN and EN.
11.3 Nutrition care plans shall be implemented to
promote safe, accurate, and effective nutrition sup-
port therapy based on the patient’s needs and clin-
ical condition and will provide resource-efficient
and fiscally responsible care.
Standard 12. Nutrition Support Access
Access for nutrition support therapy shall be achieved and
maintained in a manner that minimizes risk to the patient
and optimizes therapeutic outcome(s).36,40-43
12.1 Standard techniques and policies should be estab-
lished and followed for access device insertion and
routine care. (See section 16.5.)
12.1.1 The selection of a venous access site (cen-
tral vs peripheral vein) should depend on
expected duration of therapy, nutrition
requirements, and patient’s vascular con-
dition and preferences.40,43,44 When PN is
912
Figure 2. Route of administration algorithm. GI, gastrointestinal; PN, parenteral nutrition.
administered via a central access device,
the femoral vein site should be avoided, es-
pecially in the obese, to minimize infection
risks associated with nontunneled central
venous catheters (CVCs).43,45-48 Periph-
erally inserted central catheters (PICC)
should not be used as a strategy to re-
duce central line-associated bloodstream
infection (CLABSI).46 A CVC with the
fewest number of lumens or ports required
for that patient should be used for PN
administration.46
12.1.2 The selection of an enteral access device
(nasoenteric vs enterostomy [ie, gastros-
tomy, jejunostomy]) should depend on the
patient’s disease state, needs and goals,
ethical situation, gastrointestinal anatomy
and function, expected duration of EN
therapy, and the ability to safely access
the gastrointestinal tract via radiologic,
surgical, endoscopic techniques, or other
guided technology.41,42,49-51 (See Table 1:
Selection of Enteral Access Based on Gas-
tric Tolerance and Anticipated Feeding
Duration.)
12.1.3 Appropriate access devices shall be placed
by a physician, nurse, or trained health-
care professional who is competent to
Nutrition in Clinical Practice 33(6)
Table 1. Selection of Enteral Access Based on Gastric
Tolerance and Anticipated Feeding Duration41 .
Duration
Normal Long Term
Gastric Short Term (Longer Than
Motility (4–6 Weeks) 6 Weeks)
Yes Nasogastric Gastrostomy
No Nasoduodenal Jejunostomy
Nasojejunal
place the specific access device. Guid-
ance technology for placement of cen-
tral venous access and enteral access de-
vices should be used only by clinicians
who have completed prerequisite training
and credentialing required by the respec-
tive institution.50,52,53 Professionals with
knowledge in preventing, recognizing, and
managing complications associated with
the placement and maintenance of the
access devices should monitor the use of
the access devices.41,43,44,46-48
12.1.4 Proper placement of central venous ac-
cess devices shall be confirmed using ap-
propriate technology and documented in
Ukleja et al
the medical record before initial use.43,44
For enteral access devices, the auscaltatory
method shall not be relied upon to dif-
ferentiate between pulmonary, gastric, and
small bowel placement of a nasoenteric
tube.50 When using enteral access system
or guidance technology to place enteral
access devices, if any difficulty occurs dur-
ing insertion, confirmation of the final
tube position should be done per insti-
tution protocol.53 Radiographic confirma-
tion is the gold standard for determining
the exact tube position after insertion and
should be used.41,49-52
12.1.5 Central venous access should be used for
the delivery of PN admixtures with an
osmolarity greater than 900 mOsm/L.54
The catheter tip should be positioned in
the lower segment of the superior vena
cava adjacent to the cavo-atrial junction.43
Peripheral PN may be administered, if in-
dicated, through a peripheral access device
provided the osmolarity of the admixture
is less than or equal to 900 mOsm/L. Lipid
injectable emulsion (ILE) may be concur-
rently infused.43,44,54
12.1.6 Monitoring procedures for nutrition sup-
port therapy administration shall include
visual inspection of the patient’s enteral
or parenteral access devices and insertion
site.
12.2 Complications related to an access device and
outcome(s) of the interventions to manage the
complication(s) shall be clearly documented in the
medical record.
Standard 13. PN Admixture Preparation
PN shall be prepared accurately as prescribed and stored
safely according to United States Pharmacopeia (USP)
General Chapter<797>: Pharmaceutical Compounding-
Sterile Products.55
13.1 PN formulations shall be prepared using current
policies and procedures regarding manufacturing,
compatibility, and stability. These procedures shall
be supervised by a licensed pharmacist with ap-
propriate credentials and experience.40,44
13.1.1 A hospital-specific standardized process
for PN preparation shall be used. This
may include the use of standardized PN
formulations when appropriate.40,56
13.1.2 A pharmacist shall review the contents
of a PN order for appropriateness and
913
compare it with previous orders when
applicable.40,56
13.2 In hospitals that use automated compounding
devices (ACDs) for preparation of PN formu-
lations, policies and procedures shall be devel-
oped to address responsibilities for operation and
maintenance, staff training, and monitoring ACD
performance (ie, quality assurance).40,57
13.2.1 Adequate training of personnel shall in-
clude use of computer software to as-
sist in daily use and trouble shooting of
ACDs.40
13.2.2 PN substrate dosing limit alerts shall be ac-
tivated in the computer software and used
in the assessment of the PN formulation
prior to compounding.40
13.2.3 Documents generated by the ACD or
other electronic devices shall be compared
with the ordered PN formulation.40
13.2.4 The pharmacist and/or pharmacy techni-
cian shall monitor the equipment during
the preparation process to assure proper
operation.40
13.2.5 End-product and validation testing of PN
admixtures should be completed.
13.3 In hospitals that outsource preparation of PN
admixtures, policies and procedures shall be de-
veloped for appropriate ordering, storage, prepa-
ration, labeling, and dispensing of PN admixtures.
Hospitals should ensure that the outsource agency
prepares PN formulations in accordance with
USP General Chapter <797>: Pharmaceutical
Compounding-Sterile Products.55
13.4 In hospitals that use standardized, commercially
available PN products, policies and procedures
shall be developed for appropriate ordering, stor-
age, preparation, labeling, and dispensing of PN
admixtures.40,56
13.5 PN admixtures shall be sterile and free from phys-
ical contaminants (foreign materials and phys-
ical matter) and minimize patient exposure to
aluminum.40,59
13.6 A pharmacist should refer to ASPEN, American
Society of Health-system Pharmacists (ASHP),
FDA Drug Shortages, or other appropriate re-
source(s) on managing shortages and outages
of PN components and develop hospital-specific
strategies to provide optimal PN therapy during
shortages.
13.7 Nonnutrient medication (eg, insulin) should be
added to PN only when supported by physiochem-
ical compatibility and stability data.54
13.8 A pharmacist shall conduct a visual inspection of
the final PN admixture prior to dispensing.39
914
13.10 Additions to PN admixture shall not be made
outside of the pharmacy sterile compounding
environment.
Standard 14. EN Formula Preparation
EN formulas and products shall be prepared accurately
and safely as prescribed and stored according to the
manufacturers’ directions and published safety consensus
recommendations.41
14.1 EN formulas shall be prepared by trained per-
sonnel under professional supervision in a clean
environment. Aseptic technique shall be used in
the preparation of EN formulas.41
14.1.1 Preparation equipment shall be sanitized
regularly.
14.1.2 Open-system containers shall be filled with
EN formula using aseptic technique.
14.2 Any addition of modular products or water to
the formula shall be ordered by the prescribing
clinician or designee.
14.2.1 Additions to EN formulas shall not be
done at the bedside.
14.2.2 Additions to closed-system EN containers
shall not be made.
Standard 15. Packaging and Labeling
PN admixtures and EN formula containers shall be ap-
propriately packaged and labeled in a standardized fashion
according to hospital policy and procedure.
15.1 PN admixtures shall be packaged in adminis-
tration containers that can assure maintenance
of sterility and allow visual inspection during
preparation, storage, and infusion.
15.1.1 The PN admixtures and ILE administered
as a separate infusion shall be labeled
with the following as described in the AS-
PEN Parenteral Nutrition Safety Consen-
sus Recommendations:40
r Two patient identifiers (eg, name, med-
ical record number, date of birth)
r Patient location or address
r Administration date and time
r Beyond-use date and time
r Route of administration (central vein
vs peripheral vein)
r Prescribed volume
r Method of administration (continuous
vs cyclic)
r Complete name of all ingredients ex-
pressed in the same units of measure
as the PN order
Nutrition in Clinical Practice 33(6)
r All PN ingredients shall be ordered
as amount per day (ie, grams per day,
mEq per day)
r Name of compounding institution or
pharmacy
15.1.2 Auxiliary labels should be affixed to PN
admixture packaging to reduce risk of
error (eg, for central line only).
15.1.3 The PN admixture shall be stored in a
refrigerator (per established guidelines),
unless the admixture will be administered
immediately to the patient.40,45,55
15.2 EN formulas shall be packaged in adminis-
tration containers, which assure accuracy of
volume, cleanliness, and minimize the risk for
contamination.41
15.2.1 Open-system administration containers
should be used if the EN formula will be
modified with modular products. However,
the addition of modular products to an
open-system container may result in an
unacceptable risk of contamination in
hyperthermal environments.41
15.2.2 Hospital-prepared EN formulas shall be
stored in a refrigerator (per established
guidelines), unless the formula will be ad-
ministered immediately to the patient.
15.3 EN labels shall be standardized.
15.3.1 EN formula containers shall be labeled
accurately with the contents and 2 pa-
tient identifiers (eg, name, medical record
number, date of birth), product name and
strength, additives, volume, and appropri-
ate hang time.41
15.3.2 EN formula container labels shall also
contain delivery site/access, route (enteral),
and method of administration (eg, contin-
uous, cyclic, bolus).41
15.3.3 EN formula container labels shall contain
a statement indicating that the product is
for enteral administration only.41
15.3.4 EN formula container labels shall contain
a statement indicating that the product is
not for IV administration.41
Standard 16. Administration of Nutrition
Support Therapy
EN formulas and PN admixtures shall be administered
safely and accurately in accordance with the prescribed
order and consistent with the patient’s tolerance.40,41
16.1 Nutrition support therapy shall be administered
by or under the supervision of trained personnel.
Ukleja et al
16.2 Hospital-specific procedures shall exist regarding
techniques used to administer nutrition support
therapy. Organizations should use infusion pumps
with the ability to reduce errors.40
16.3 Acute care facilities should establish a policy that
prohibits the use of a PN admixture prepared
for administration at home or in subacute or
long-term care facilities. PN should be discontin-
ued prior to discharge or transport to another
facility.40
16.4 Each PN admixture should be inspected prior to
and during administration. If visual changes are
present, the admixture shall not be administered,
and the pharmacy shall be notified.40,60
16.5 Before nutrition support therapy is administered
to the patient, the label on the container shall
be checked against the order and the patient’s
identity shall be verified per hospital policy
to assure the prescribed formulation is deliv-
ered to the appropriate patient and administered
by the correct route at the designated/intended
time.40,41,61,62 Administration tubing should be
attached to PN containers immediately prior
to use.40
16.6 The administration rate of the prescribed nutri-
tion support therapy shall be checked each time
a new volume is ordered or initiated and peri-
odically during its administration.40,41 Use of an
independent double-check verification should be
performed by a second clinician prior to beginning
a PN infusion.40,41,62
16.7 Procedures shall be written to prevent and manage
vascular or enteral access device occlusion and IV
extravasation.41,42,44,63
16.8 EN and PN processes shall be documented in
the patient’s medical record including tolerance,
administration volumes, and hourly rates. The
amount of nutrition therapy ordered vs amount
administered should be noted and reasons for
discrepancies evaluated.41
16.9 Policies and procedures shall exist to prevent,
diagnose, manage, and monitor patient infections
caused by contamination of the PN admixture
or the equipment/devices used in its administra-
tion, as PN is an independent risk factor for
CLABSI.46-48,64
16.9.1 Infection prevention strategies shall be
used to minimize CLABSI including
a bundle of targeted, evidence-based
catheter insertion and maintenance
practices.46-48
16.9.2 Access ports shall be disinfected with an
appropriate antiseptic prior to catheter
manipulation and manipulation should
915
be minimized; vascular access devices
used for PN should not be used for blood
sampling.46,47,64 Coinfusion of fluids or
medications into the PN system should
be avoided, if possible. If no alternatives
are available, a pharmacist shall review
the compatibility and stability data for
coinfusion prior to administration.40,44
16.9.3 Unit-specific data regarding CLABSI
shall be shared with all internal stake-
holders and reported to external agencies
as required.47
16.9.4 PN admixtures shall be labeled with the
beyond-use date and time and discarded
as indicated. Once the delivery system
is accessed, the administration of a PN
admixture shall be completed within 24
hours.40,44-46,55
16.9.5 Administration sets for PN shall be
changed every 24 hours or with each new
PN container. A 1.2-micron filter shall be
used for all total nutrient admixtures and
a 0.22-micron filter for dextrose/amino
acids (2-in-1) admixtures.40,44,65
16.9.6 ILE administered separately from PN ad-
mixtures (dextrose/amino acids) shall be
infused through a 1.2-micron filter and be
completed within 12 hours of initiating
the infusion.40,44,65
16.10 A policy shall exist regarding the maximal rate
of administration for ILE. Manufacturers’ recom-
mendations should be considered in formulating
this statement.
16.11 Cycling of PN admixtures should be considered
for patients with or at risk for liver dysfunction,
on long-term PN, or those who are stable, active,
and may benefit from infusion-free time.66,67
16.12 Prevention strategies shall be used to minimize
the risk of microbial contamination of EN for-
mulations. (Refer to the Enteral Nutrition Prac-
tice Recommendations41 and Guidelines for the
Provision and Assessment of Nutrition Support
Therapy in the Adult Critically Ill Patient31 for
details.)
16.13 Procedures and protocols to minimize the risk of
regurgitation and aspiration of EN formulations
should be implemented.41
16.13.1 All patients receiving EN shall be as-
sessed for risk of aspiration and steps
employed to reduce aspiration risk and
pneumonia.31,41,68-70
16.13.2 The head of the bed should be elevated
30–45° during EN administration unless
contraindicated.31,41,69-71
916
16.14 An enteral feeding protocol should be designed
to assure that optimal nutrients are delivered
and unnecessary interruption of feeding mini-
mized. Gastric residual volumes may be used to
assess EN tolerance as part of a multifaceted
approach.31,41,68
16.15 Policies and procedures shall exist to minimize the
risk of enteral misconnections.41,72-74
16.15.1 Hospitals should conform to Interna-
tional Organization of Standardization
(ISO) standard 80369-3 that is driving
the production of products with incom-
patible connectors by designing features
that make incorrect connections impos-
sible (eg, ENFit).41,74-77 Enteral deliv-
ery devices (administration sets, feeding
tubes, and enteral syringes) with con-
nectors that can physically connect with
other nonenteral connectors shall not be
purchased.
16.15.2 Standard Luer syringes shall not be used
to administer oral or enteral medications
or EN formula.41,74
16.15.3 Tubes or catheters shall be traced
from the patient to the point of origin
before connecting any new device or
infusion.41,74,76
16.15.4 Tubes and catheters having different pur-
poses should be routed in different, stan-
dardized directions (eg, IV lines routed
toward the head; enteric lines toward the
feet) and labeled at proximal and distal
tubing ends.73
16.16 Protocols shall be established for administering
medications and modular products through an
enteral access device.41
16.16.1 Medications should not be mixed
directly with EN formulas due to
potential drug-drug and drug-nutrient
interactions.41,78-81
16.16.2 Medication orders should specify the
route of delivery (eg, PO, NG tube, G
tube, J tube) and be administered accord-
ing to current guidelines. Special con-
siderations include use of proper dosage
form, administration of the drug separate
from the EN formula and other drugs,
and the location of drug delivery in the
gastrointestinal tract.41,78-83
16.16.3 Enteral access device(s) should be
flushed appropriately before and after
each medication administration and
restarting EN administration to help
prevent occlusion.41,80,82
Nutrition in Clinical Practice 33(6)
Standard 17. Adverse Events Management
An adverse event, including sentinel events related to the
administration of nutrition support therapy and the equip-
ment/access devices, shall be documented and reported
according to hospital protocol to promote a culture of
patient safety. Protocols should be developed and followed
to decrease the risk of adverse events.
Chapter V: Monitoring and Reevaluating
the Nutrition Care Plan
Standard 18. Parameters and Frequency
A plan for monitoring the effect of nutrition support
therapy interventions should be stated in the nutrition care
plan.36,40,41
Monitoring parameters are chosen relative to the therapy
goals of the nutrition care plan. The nutrition care plan shall
be revised to optimize nutrition support therapy and achieve
predetermined goals, as indicated.
18.1 The frequency of monitoring should depend on
severity of illness, level of metabolic stress, nu-
trition status, as well as the patient’s clinical
condition.31,36,40,41
18.1.1 Daily or more frequent monitoring should
be required in patients who are critically ill,
have debilitating diseases (eg, diabetes mel-
litus) or infection, are at risk for refeeding
syndrome complications, are transitioning
between PN or EN and oral diet, or have
experienced complications associated with
nutrition support therapy.
18.1.2 Weekly or as clinically indicated moni-
toring may be needed in patients who
are clinically and metabolically stable with
documented stable laboratory parameters.
18.2 Monitoring parameters should include the follow-
ing:
r Physical assessment, including clinical signs of
fluid and nutrient excess or deficiency
r Functional status
r Vital signs
r Actual nutrient intake (oral, enteral, and
parenteral)
r Weight
r Laboratory data
r Diagnostic tests
r Review of all medications
r Changes in gastrointestinal function
r Input and output/fluid balance
18.3 Appropriate changes in nutrition support therapy
shall be made based on results of monitored
parameters. Recommended changes in nutrition
Ukleja et al
support therapy including EN formula/PN admix-
ture or administration route and resulting out-
comes shall be documented in the nutrition care
plan.84
18.4 Protocols should be established to maintain blood
glucose control in patients receiving EN or
PN.85-87
Standard 19. Reevaluation of Nutrition Care
Plan
The patient shall be monitored for progress toward short-
and long-term goals as defined in the nutrition care
plan.36,41
19.1 Appropriate parameters should be measured
serially during nutrition support therapy and
documented.36,40,41 Parameters may include
weight change, changes in laboratory data,
adequacy of intake, ability to transition to oral
diet, functional status performance, and quality
of life.
19.2 The monitoring parameters should be compared
with the goals of the nutrition care plan. If goals
are not being met, a new clinical issue or complica-
tion develops, and/or an adverse event occurs, the
nutrition care plan should be modified.
Chapter VI: Transition of Therapy
Standard 20. Adequacy of Intake
The transition of nutrition therapies shall be monitored.
Recommendations for improving oral and EN intake shall
be documented. Adequacy of energy and nutrient intake
is based on clinical judgment and shall be assessed and
documented before discontinuation of nutrition support
therapy.31,35,40 (See Figure 2: Route of administration algo-
rithm.)
Standard 21. Continuity of Care
Continuity of the nutrition support therapy shall oc-
cur through active communication with all members of
the patient care team, the patient, and caregiver(s). (See
Figure 1: Adult nutrition care pathway.)
21.1 A plan shall be developed for transition of nu-
trition support therapy to an alternate healthcare
facility or to home care and should include
identification of the primary clinician responsi-
ble for coordinating, monitoring, providing ed-
ucation, and ordering home nutrition support
therapy.31,36,40,41,87,88
21.2 Indications for home nutrition support therapy
shall be documented.
917
21.3 Appropriate education should be provided to pa-
tient and/or caregiver(s) and documented before
discharge. Communication with home infusion
and healthcare agencies and with the patient’s
home nutrition support management team should
be established prior to hospital discharge.
21.4 The nutrition support therapy prescription and
administration schedule should be documented
and communicated with home infusion and home
health agencies before discharge.36,40,41,45,88 Specif-
ically, with PN, there should be pharmacist-
to-pharmacist communication to the alternate
healthcare facilities or home agencies.
21.5 Periodic monitoring should be recommended de-
pending on patient’s condition.35,39,40,86
Standard 22. Nutrition Therapy at End-of-Life
Care
The decision on nutrition support therapy in an end-of-
life setting should be determined by patient autonomy
and the patient’s family member(s) or surrogate decision
maker. The patient or the patient’s family member(s) or
surrogate decision maker shall decide on acceptance or
refusal of medical therapy.36,41,86,88-90
22.1 The clinician has no obligation to provide nutri-
tion support therapy and hydration to a patient in
the end-of life situation.88-90
22.2 Decisions at end-of-life are often made based on
healthcare and spiritual literacy of the patient
and his/her family; they shall be involved in the
healthcare process of end-of-life.
References
1. Ulkeja A, Freeman KL, Gilbert K, et al. Standards for nutrition
support: adult hospitalized patients. Nutr Clin Pract. 2010;25(4):403-
414.
2. Robinson D, Walker R, Adams S, et al. American Society for Parenteral
and Nutrition (A.S.P.E.N.) Definition of Terms, Enteral Style, and
Conventions Used in A.S.P.E.N. Board of Directors–Approved
Documents. https://www.nutritioncare.org/uploadedFiles/Documents/
Guidelines_and_Clinical_Resources/ASPEN%20Definition%
20of%20Terms,%20Style,%20and%20Conventions%20Used%20in%
20ASPEN%20Board%20of%20Directors-Approved%
20Documents.pdf. Accessed July 1, 2018.
3. The Joint Commission PC.01.02.01, PC 01.02.03. Comprehensive Ac-
creditation Manual for Hospitals: The Official Handbook. Oak Brook,
IL: Joint Commission Resources, 2018.
4. Schneider PJ. Nutrition support teams: an evidence-based practice.
Nutr Clin Pract. 2006;21(1):62-67.
5. Winkler MF. Improving safety and reducing harm associated with
specialized nutrition support. Nutr Clin Pract. 2005;20(6):595-596.
6. Gurgueira GL, Leite HP, Taddei JA, de Carvalho WB. Outcomes in a
pediatric intensive care unit before and after the implementation of a
nutrition support team. JPEN J Parenter Enteral Nutr. 2005;29(3):176-
185.
918
7. O’Brien DD, Hodges RE, Day AT, Waxman KS, Rebello T. Rec-
ommendations of nutrition support team promote cost containment.
JPEN J Parenter Enteral Nutr. 1986;10(3):300-302.
8. Traeger SM, Williams GB, Milliren G, Young DS, Fisher M, Haug
MT. Total parenteral nutrition by a nutrition support team: improved
quality of care. JPEN J Parenter Enteral Nutr.1986;10(8):408-412.
9. Goldstein M, Braitman LE, Levine GM. The medical and financial
costs associated with termination of a nutrition support nurse. JPEN
J Parenter Enteral Nutr. 2000;24(6):323-327.
10. Naylor CJ, Griffiths RD, Fernandez RS. Does a multidisciplinary
total parenteral nutrition team improve patient outcomes? A systematic
review. JPEN J Parenter Enteral Nutr. 2004;28(4):251-258.
11. Gales BJ, Riley DG. Improved total parenteral nutrition therapy man-
agement by a nutritional support team. Hosp Pharm.1994;29(5):469-
470, 473-475.
12. Dalton MJ, Schepers G, Gee JP, Alberts CC, Eckhauser FE, Kirking
DM. Consultative total parenteral nutrition teams: effect on the
incidence of total parenteral nutrition-related complications. JPEN J
Parenter Enteral Nutr. 1984;8(2):146-152.
13. Maurer J, Weinbaum F, Turner J, et al. Reducing the inappropriate
use of parenteral nutrition in an acute care teaching hospital. JPEN J
Parenter Enteral Nutr. 1996;20(4):272-274.
14. Chris Anderson D, Heimburger DC, Morgan SL, et al. Metabolic
complications of total parenteral nutrition: effects of a nutri-
tion support service. JPEN J Parenter Enteral Nutr. 1996;20(3):
206-210.
15. A.S.P.E.N. Practice Management Task Force: DeLegge M, Wooley JA,
Guenter P, et al. The state of nutrition support teams and update on
current models for providing nutrition support therapy to patients. Nutr
Clin Pract 2010;25(1):76-84.
16. Braun K, Utech A, Velez ME, Walker R. Parenteral nutrition elec-
trolyte abnormalities and associated factors before and after nutrition
support team initiation. JPEN J Parenter Enteral Nutr. 2018;42(2):387-
392.
17. Hassell JT, Games AD, Shaffer B, Harkins LE. Nutrition support team
management of enterally fed patients in a community hospital is cost-
beneficial. J Am Diet Assoc. 1994;94(9):993-998.
18. Kennedy JF, Nightingale JM. Cost savings of an adult hospital
nutrition support team. Nutrition. 2005;21(2):1127-1133.
19. DeLegge MH, Kelly AT. The state of nutrition support teams. Nutr
Clin Pract. 2013;28(6):691-697.
20. Mascarenhas MR, August DA, DeLegge MH, et al. Standards of prac-
tice for nutrition support physicians. Nutr Clin Pract. 2012;27(2):295-
299.
21. DiMariua-Ghalili RA, Gilbert K, Lord L, et al. Standards of nutrition
care practice and professional performance for nutrition support and
generalist nurses. Nutr Clin Pract. 2016;31(4):527-547.
22. Tucker A, Ybarra J, Bingham A, et al. American Society for Parenteral
and Enteral Nutrition (A.S.P.E.N.) Standards of practice for nutrition
support pharmacists. Nutr Clin Pract. 2015;30(1):139-146.
23. Brantley SL, Russell MK, Mogensen KM, et al. American Society
for Parenteral and Enteral Nutrition and Academy of Nutrition
and Dietetics: revised 2014 standards of practice and standards of
professional performance for registered dietitian nutritionists (com-
petent, proficient, and expert) in nutrition support. Nutr Clin Pract.
2014;29(6):792-828.
24. Mathieson K, Vogt EAM. Quality improvement in clinical practice. In:
Mueller CM, ed. The ASPEN Adult Nutrition Support Core Curriculum
(3rd ed.). Silver Spring, MD: American Society for Parenteral and
Enteral Nutrition; 2017:805-818.
25. Weekes CE, Elia M, Emery PW. The development, validation and reli-
ability of a nutrition screening tool based on the recommendations of
the British Association for Parenteral and Enteral Nutrition (BAPEN).
Clin Nutr. 2004;23(5):1104-1112.
Nutrition in Clinical Practice 33(6)
26. Kondrup J, Rasmussen H, Hamberg O, et al. Nutritional risk screening
(NRS 2002): a new method based on an analysis of controlled clinical
trials. Clin Nutr. 2003;22(3):321-336.
27. Mueller C, Compher C, Druyan ME, A.S.P.E.N. Board of Directors.
A.S.P.E.N. Clinical Guidelines: nutrition screening, assessment, and
intervention in adults. JPEN J Parenter Enteral Nutr. 2011;35(1):16-
24.
28. Lacey K, Pritchett E. Nutrition care process and model: ADA adopts
road map to quality care and outcomes management. J Am Diet Assoc.
2003;103(8):1061-1072.
29. American Dietetic Association. Identifying patients at risk: ADA’s
definitions for nutrition screening and nutrition assessment. Council
on Practice (COP) Quality Management Committee. J Am Diet Assoc.
1994;94(8):838-839.
30. JeVenn AK, Galang M, Hipskind P, Bury C. Malnutrition screening
and assessment. In: Mueller CM, ed. The ASPEN Adult Nutrition
Support Core Curriculum (3rd ed.). Silver Spring, MD: American
Society for Parenteral and Enteral Nutrition; 2017:185-212.
31. McClave SA, Taylor BE, Martindale RG, et al. Guidelines for the
provision and assessment of nutrition support therapy in the adult
critically ill patient. JPEN J Parenter Enteral Nutr. 2016;40(2):159-211.
32. Lim SlL, Ang E, Foo YL, et al. Validity and reliability of nutrition
screening administered by nurses. Nutr Clin Pract. 2013;28(6):730-736.
33. Patel V, Roman M, Corkins MR. Nutrition screening and assessment in
hospitalized patients: a survey of current practice in the United States.
Nutr Clin Pract. 2014;29(4):483-490.
34. White JV, Guenter P, Jensen G, et al. Consensus Statement: Academy
of Nutrition and Dietetics and American Society for Parenteral and
Enteral Nutrition: Characteristics recommended for the identification
and documentation of adult malnutrition (under-nutrition). JPEN J
Parent Enteral Nutr. 2012;36(3):275-283.
35. Guenter P, Jensen G, Patel V, et al. Addressing disease-related malnu-
trition in hospitalized patients: A call for a national goal. Jt Comm J
Qual Patient Saf. 2015;41(10):469-473.
36. Worthington P, Balint J, Bechtold M, et al. When is PN appropriate?
JPEN J Parenter Enteral Nutr. 2017;41(3):324-377.
37. Doley J, Phillips W. Overview of enteral nutrition. In: Mueller CM, ed.
The ASPEN Adult Nutrition Support Core Curriculum (3rd ed.). Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition;
2017:213-226.
38. Mirtallo J. Overview of parenteral nutrition. In: Mueller CM, ed.
The ASPEN Adult Nutrition Support Core Curriculum (3rd ed.). Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition;
2017:285-296.
39. Guenter P, Boullata JI, Ayers P, et al. Standardized competencies
for PN prescribing: the American Society of Parental and Enteral
Nutrition model. Nutr Clin Pract. 2015;30(4):570-576.
40. Ayers P, Adams SC, Boullata J, et al. A.S.P.E.N. Parenteral nutrition
safety consensus recommendations. JPEN J Parenter Enteral Nutr.
2014;38(2):296-333.
41. Boullata JI, Carrera AL, Harvey L, et al. ASPEN safe practices for
enteral nutrition therapy. JPEN J Parenter Enteral Nutr. 2017;41(1):15-
103.
42. Fang JC, Kinikini M. Enteral access devices. In: Mueller CM, ed.
The ASPEN Adult Nutrition Support Core Curriculum, (3rd ed.) Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition.
2017;251-264.
43. Neal AM, Drogan K. Parenteral access devices. In: Mueller CM, ed.
The ASPEN Adult Nutrition Support Core Curriculum (3rd ed.). Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition;
2017:321-344.
44. Gorski LA, Hadaway L, Hagle M, McGoldrick M, Orr M, Doellman
D. Infusion therapy standards of practice. J Infus Nurs. 2016;39(1
Suppl):S1-S159.
Ukleja et al
45. Mirtallo J, Canada T, Johnson D, et al. Task Force for the Revision of
Safe Practices for Parenteral Nutrition. Safe practices for parenteral
nutrition [erratum in JPEN J Parenter Enteral Nutr. 2006;30:177].
JPEN J Parenter Enteral Nutr. 2004;28(6):S39-S70.
46. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for the preven-
tion of intravascular catheter-related infections. https://www.cdc.gov/
infectioncontrol/guidelines/pdf/bsi/bsi-guidelines-H.pdf. Accessed
March 5, 2018.
47. Marschall J, Mermel L, Fakih M, et al. Strategies to prevent central
line-associated bloodstream infections in acute care hospitals: 2014
update. Infect Control Hosp Epidemiol. 2014;35(7):753-771.
48. Chopra V, Krein S, Olmsted R, Safdar N, Saint S. Prevention of central
line-associated bloodstream infections: brief update review. In: Shekelle
PG, Wachter RM, Pronovost PJ, eds. AHRQ. Making Healthcare Safer
II: An Updated Critical Analysis of the Evidence for Patient Safety
Practices. AHRQ Publication No. 13-E001-EF. Rockville, MD: Agency
for Healthcare Research and Quality. 2013:88-109. https://www.
ahrq.gov/sites/default/files/wysiwyg/research/findings/evidence-based-
reports/services/quality/patientsftyupdate/ptsafetyII-full.pdf.
Accessed March 7, 2018.
49. Koopman MC, Kudsk KA, Szotkowski MJ, et al. A team-based pro-
tocol and electromagnetic technology eliminate feeding tube placement
complications. Ann Surg. 2011;253(2):297-302.
50. American Association of Critical Care Nurses. (2016). AACN prac-
tice alert: initial and on-going verification of feeding tube place-
ment in adults. (applies to blind insertions and placements with
an electromagnetic device). http://www.aacn.org/wd/practice/content/
feeding-tube-practice-alert.pcms?menu=practice. Accessed March 7,
2018.
51. Powers J, Lubbehusen M, Spitzer T, et al. Verification of an electro-
magnetic placement device compared with abdominal radiograph to
predict accuracy of feeding tube placement. JPEN J Parenter Enteral
Nutr. 2011:35(4):535-539.
52. Metheny NA, Meert KL. Effectiveness of an electromagnetic feeding
tube placement device in detecting inadvertent respiratory placement.
Am J Crit Care. 2014;23(3):240-248.
53. U.S. Food and Drug Administration. Feeding tube placement systems-
letter to health care providers. http://www.fda.gov/MedicalDevices/
Safety/LetterstoHealthCareProviders/ucm591838.htm. Accessed June
20, 2018.
54. Boullata JI, Gilbert K, Sacks G, et al. A.S.P.E.N. Clinical guidelines:
parenteral nutrition ordering order review, compounding, labeling,
and dispensing. JPEN J Parenter Enteral Nutr. 2014;38(3):334-
377.
55. Pharmaceutical compounding—sterile preparations (general informa-
tio chapter 797). In: The United States Pharmacopeia, 31st rev., and
the National Formulary 26th ed. Rockville, MD: The United States
Pharmacopeial Convention; 2008:334-351.
56. A.S.P.E.N. Board of Directors and Task Force on Parenteral Nutrition
Standardization: Kochevar M, Guenter P, Holcombe B, Malone A,
Mirtallo J. A.S.P.E.N. statement on parenteral nutrition standardiza-
tion. JPEN J Parenter Enteral Nutr. 2007;31(5):441-448.
57. American Society of Health-System Pharmacists. ASHP guidelines
on the safe use of automated compounding devices for the prepa-
ration of parenteral nutrition admixtures. Am J Health-Syst Pharm.
2000;57(14):1343-1348.
58. American Society of Health-System Pharmacists. ASHP guidelines on
outsourcing sterile compounding services. Am J Health-Syst Pharm.
2015;72(19):1664-1675.
59. The American Society for Parenteral and Enteral Nutrition
(A.S.P.E.N.) Aluminum Task Force; Charney PJ. A.S.P.E.N. statement
on aluminum in parenteral nutrition solutions. Nutr Clin Pract.
2004;19(4):416-417.
919
60. Rollins CJ. Total nutrient admixtures: stability issues and their impact
on nursing practice. J Intraven Nurs. 1997;20(6):299-304.
61. Pennsylvania Patient Safety Authority. The five rights: not the gold
standard for safe medication practices. PA-PSRS Patient Safety
Advisory. http://patientsafety.pa.gov/ADVISORIES/Pages/200506_09
.aspx. Accessed March 8, 2018.
62. Institute for Safe Medication Practices. Independent double checks:
undervalued and misused: selective use of this strategy can play an
important role in medication safety. ISMP Medication Safety Alert!
2013;18(12):1-4. https://www.ismp.org/newsletters/acutecare/show
article.aspx?id=51. Accessed March 8, 2018.
63. Benedetta B, Andres C. Images in clinical medicine: Extravasation
of peripherally administered parenteral nutrition. N Engl J Med.
2011;364(10):e20.
64. The Joint Commission. Preventing central line–associated bloodstream
infections: a global challenge, a global perspective. http://www.jointcom
mission.org/assets/1/18/clabsi_monograph.pdf. Published May 2012.
Accessed March 7, 2018.
65. Institute for Safe Medication Practices. IV fat emulsion needs a filter.
ISMP Medication Safety Alert! 2016;21(1):2-3. http://www.ismp.org/
Newsletters/acutecare/issue.aspx?id=1113. Accessed March 8, 2018.
66. Stout SM, Cober MP. Metabolic effects of cyclic parenteral nutrition
infusion in adults and children. Nutr Clin Pract. 2010;25(3):277-
281.
67. Suryadevara S, Celestin J, DeChicco R, et al. Type and prevalence
of adverse events during the parenteral nutrition cycling process in
patients being prepared for discharge. Nutr Clin Pract. 2012;27(2):268-
273.
68. Malone A, Seres D, Lord L. Complications of enteral nutrition. In:
Mueller CM, ed. The ASPEN Adult Nutrition Support Core Curriculum
(3rd ed.). Silver Spring, MD: American Society for Parenteral and
Enteral Nutrition; 2017:265-284.
69. American Association of Critical Care Nurses. AACN practice alert:
prevention of aspiration in adults. Crit Care Nurse. 2016;36(1):e20-e24.
https://www.aacn.org/clinical-resources/practice-alerts/prevention-of-
aspiration. Accessed March 7, 2018.
70. DiBardino D, Wunderink R. Aspiration pneumonia: a review of
modern trends. J Crit Care. 2015;30(1):40-48.
71. Tablan OC, Anderson LJ, Besser R, et al. Guidelines for preventing
health-care-associated pneumonia, 2003: recommendations of CDC
and the Healthcare Infection Control Practices Advisory Committee.
MMWR Recomm Rep. 2004;53(RR-3):1-36.
72. Guenter P, Hicks RW, Simmons D, et al. Enteral feeding misconnec-
tions: a consortium position statement. Jt Comm J Qual Patient Saf.
2008;34(5):285-292.
73. Guenter P. New enteral connectors: raising awareness. Nutr Clin Pract.
2014; 29(5):612-614.
74. The Joint Commission. Managing risk during transition to new ISO tub-
ing connector standards. Sentinel Event Alert 53. http://www.jointcom
mission.org/sea_issue_53/. Published August 20, 2014. Accessed March
8, 2018.
75. Premier Safety Institute. Tubing misconnections. http://www.premier
safetyinstitute.org/safety-topics-az/tubing-misconnections/tubing-
misconnections/. Accessed March 8, 2018.
76. Simmons D, Symes L, Graves K. Tubing misconnections: normaliza-
tion of deviance. Nutr Clin Pract. 2011;26(3):286-293.
77. Institute for Safe Medication Practices. EnFit enteral devices are
on their way . . . important safety considerations for hospitals. ISMP
Medication Safety Alert! 2015;20(7):1-7.
78. Verdell A, Rollins CJ. Drug-nutrient interactions. In: Mueller CM, ed.
The ASPEN Adult Nutrition Support Core Curriculum (3rd ed.). Silver
Spring, MD: American Society for Parenteral and Enteral Nutrition;
2017:361-378.
920
79. Boullata JI. Drug administration through an enteral feeding tube. Am
J Nurs. 2009;109(10):34-42.
80. Williams NT. Medication administration through enteral feeding
tubes. Am J Health-System Pharm. 2008;65(24):2347-2357.
81. Institute for Safe Medication Practices. Preventing errors when admin-
istering drugs via an enteral feeding tube. ISMP Medication Safety
Alert! 2010;15(9)1-4. https://www.ismp.org/newsletters/acutecare/
articles/20100506.asp. Accessed March 8, 2018.
82. McIntyre CM, Monk H. Medication absorption considerations in
patients with postpyloric enteral feeding tubes. Am J Health-System
Pharm. 2014;71(7):549-556.
83. Institute for Safe Medication Practices. Oral dosage forms that should
not be crushed 2016. http://www.ismp.org/tools/DoNotCrush.pdf. Ac-
cessed March 8, 2018.
84. McClave SA, DiBaise JK, Mullin GE, Martindale RG. ACG clinical
guideline: nutrition therapy in the adult hospitalized patient. Am J
Gastroenterol. 2016;111(3):315-334.
Nutrition in Clinical Practice 33(6)
85. McMahon MM, Nystrom E, Braunschweig C, Miles J, Compher C,
ASPEN Board of Directors. A.S.P.E.N. Clinical Guidelines: nutrition
support of adult patients with hyperglycemia. JPEN J Parenter Enteral
Nutr. 2013;38(1):23-36.
86. Mundi MS, Nystrom E, Hurley DL, McMahon MM. Management of
parenteral nutrition in hospitalized patients. JPEN J Parenter Enteral
Nutr. 2017;41(4):535-549.
87. Durfee SM, Adams SC, Arthur E, et al. A.S.P.E.N. Standards for
nutrition support: home and alternate site care. Nutr Clin Pract.
2014;29(5):542-555.
88. Ganzini L. Artificial nutrition and hydration at the end of life: ethics
and evidence. Palliat Support Care. 2006;4(2):135-143.
89. Barrocas A, Geppert C, Durfee SM, et al. A.S.P.E.N. ethics position
paper. Nutr Clin Pract. 2010;25(6):672-679.
90. Schwartz DB, Barrocas A, Wesley JR, et al. Gastrostomy tube place-
ment in patients with advanced dementia or near end of life life. Nutr
Clin Pract. 2014;29(6):829-840.
Special Report

Nutrition in Clinical Practice

Volume 33 Number 6
Pediatric Nasogastric Tube Placement and Verification: Best December 2018 921–927

C 2018 American Society for

Practice Recommendations From the NOVEL Project Parenteral and Enteral Nutrition
DOI: 10.1002/ncp.10189
wileyonlinelibrary.com

Sharon Y. Irving, PhD, CRNP, FCCM, FAAN1,2 ; Gina Rempel, MD, FRCPC3,4 ;
Beth Lyman, RN, MSN, CNSC, FASPEN5 ; Wednesday Marie A. Sevilla, MD, MPH,
CNSC6 ; LaDonna Northington, DNS, RN, BC7 ; Peggi Guenter, PhD, RN, FAAN,
FASPEN8 ; and The American Society for Parenteral and Enteral Nutrition

Abstract
The placement of a nasogastric tube (NGT) in a pediatric patient is a common practice that is generally perceived as a benign bedside
procedure. There is potential risk for NGT misplacement with each insertion. A misplaced NGT compromises patient safety,
increasing the risk for serious and even fatal complications. There is no standardized method for verification of the initial NGT
placement or reverification assessment of NGT location prior to use. Measurement of the acidity or pH of the gastric aspirate is the
most frequently used evidence-based method to verify NGT placement. The radiograph, when properly obtained and interpreted,
is considered the gold standard to verify NGT location. However, the uncertainty regarding cumulative radiation exposure related
to radiographs in pediatric patients is a concern. To minimize risk and improve patient safety, there is a need to identify best practice
and to standardize care for initial and ongoing NGT location verification. This article provides consensus recommendations for best
practice related to NGT location verification in pediatric patients. These consensus recommendations are not intended as absolute
policy statements; instead, they are intended to supplement but not replace professional training and judgment. These consensus
recommendations have been approved by the American Society for Parental and Enteral Nutrition (ASPEN) Board of Directors.
(Nutr Clin Pract. 2018;33:921–927)

Keywords
enteral nutrition; misplacement; NG feeding tube; pediatric

Introduction when initially placed or for reverification of NGT location

In 2014, a report on 255,140 hospital discharges in the
United States revealed that 25% of all patients receiving
enteral nutrition (EN) were children. Of those, 6% of the
total number of patients were under 12 months of age.1 This
report indicates that a significant number of hospitalized
children require feeding tube placement for the adminis-
tration of EN. A 2016 study reported that approximately
25% of hospitalized pediatric patients require a temporary
feeding tube.2 The most common enteral feeding tube
used in hospitalized children is a nasogastric tube (NGT).2
Insertion of an NGT is a high-volume practice commonly
performed by nurses as a blind procedure, without the use
of technology to guide or visualize the internal path of the
tube. Although the vast majority of blind NGT insertions
result in successful placement in the intended location—
the stomach—each tube can potentially be misplaced, even
when the procedure is performed by a healthcare provider
experienced in NGT placement. A misplaced NGT compro-
mises patient safety, increasing the risk for severe and even
fatal complications.3-6 Clinicians practicing in the United
States lack guidance for best practice to verify NGT location
before use in pediatric patients. There is a need to identify
best practice and standardize care for initial and ongoing
verification of NGT location to decrease the risk of a
misplaced NGT. Therefore, the objective of this article is to
From the 1 University of Pennsylvania School of Nursing,
Philadelphia, Pennsylvania, USA; 2 Department of Critical Care
Nursing, The Children’s Hospital of Philadelphia, Philadelphia,
Pennsylvania, USA; 3 Rady Faculty of Health Sciences, Max Rady
College of Medicine, University of Manitoba, Winnipeg, Canada;
4 Children’s Hospital Winnipeg, Winnipeg, Manitoba, Canada;
5 Nutrition Support Team, Children’s Mercy Hospital, Kansas City,
Missouri, USA; 6 Division of Pediatric Gastroenterology, Children’s
Hospital of Pittsburgh, University of Pittsburgh Medical Center,
Pittsburgh, Pennsylvania, USA; 7 University of Mississippi Medical
Center School of Nursing, Jackson, Mississippi, USA; and 8 Clinical
Practice, Quality, and Advocacy, American Society for Parenteral and
Enteral Nutrition (ASPEN), Silver Spring, Maryland, USA.
This article originally appeared online on September 6, 2018.
Corresponding Author:
Sharon Y. Irving, PhD, CRNP, FCCM, FAAN, Assistant Professor,
University of Pennsylvania School of Nursing, Claire M. Fagin Hall,
418 Curie Blvd., RM 427, Philadelphia, PA 19104.
Email: ysha@nursing.upenn.edu
922
develop and disseminate recommendations for best practice
related to NGT location verification in pediatric patients
based on the available literature.
The consensus recommendations presented here are not
intended as absolute policy statements. Use of these prac-
tice recommendations does not in any way guarantee any
specific benefit in outcome or survival. The professional
judgment of the attending health professional is the primary
component of quality medical care delivery. Because con-
sensus recommendations cannot account for every variation
in circumstances, practitioners must always exercise pro-
fessional judgment when applying these recommendations
to individual patients. These consensus recommendations
are intended to supplement, but not replace, professional
training and judgment.
Background
The literature is replete with case reports in both pediatric
and adult patients describing NGT misplacements.5 An
early account of submucosal placement of an NGT that
occurred in an adult patient under anesthesia was reported
by Daly in 1953 and again by Lind et al. in 1978.7,8 Compli-
cations in adult patients related to NGT placement can be
found from the 1970s through the present day.8-11 Similar
reports of misplaced NGTs in children describe insertions
into the esophagus,4,12,13 pylorus,4 pharyngeal mucosa,14,15
intracranium,16 and most commonly the respiratory tract.17
Both gastric and bladder perforation related to NGT
misplacement have been described in children.18,19 These
reports of misplaced NGTs in both adult and pediatric
patients demonstrate the risks associated with insertion,
emphasizing that NGT placement is not always a benign
procedure, as is often perceived.
Although NGT misplacements have frequently been
described in the United States, it is difficult to quantify
an actual number of misplacements, particularly in chil-
dren. There are limitations to state reporting mechanisms,
a lack of a national reporting system, and a presumed
hesitancy among institutions to provide public access to
such information. Without a denominator for the number of
NGTs placed and a definitive number of misplacements, it
is virtually impossible to accurately quantify the number of
NGT misplacements that occur. Pennsylvania is one of few
states where reporting of NGT misplacements is required. A
2017 Pennsylvania Patient Safety Authority report describes
166 enteral tube misplacements documented from 2011–
2016.20 In this report, 10.2% of the misplacements occurred
in pediatric patients, with many of these misplacements
associated with adverse events.20 This report is one of
few references that quantifies the number of enteral tube
misplacements in a defined time period, thus depicting the
scope of the problem for verification of NGT location.
Despite the lack of robust prevalence data describing NGT
Nutrition in Clinical Practice 33(6)
misplacements in pediatric patients in the United States, it
is well recognized that even 1 undetected, misplaced NGT
may have major implications for the individual patient, the
family, and the entire healthcare system.21
Although the reporting of NGT misplacements is incon-
sistent in the United States, the National Health Service
(NHS) in the United Kingdom has issued a series of Patient
Safety Alert (PSA) reports, the most recent in 2016.22
In that 2016 PSA report spanning 5 years (2011–2016),
there were 95 occurrences of NGT misplacement, with 32
deaths from the over 3 million NGT (or orogastric tube)
placements recorded.22 The most common error cited was
the inaccurate interpretation of a radiograph used to verify
NGT position. Other errors cited include the use of unap-
proved methods for NGT placement verification, nursing
error in performing pH testing, and communication failures
for which the NGT location was not checked before use.
Failure at the organizational level to implement previously
identified processes to ensure correct NGT placement was
determined to be a primary cause of the misplacements.
Thus, a mandate was issued by the NHS to declare NGT
misplacement a never event, and a PSA was directed to
organizations emphasizing the seriousness with which NGT
placement is to be regarded, holding both the provider and
the organization accountable for patient safety related to
NGT location verification.22,23
Current Practice
Correct NGT placement begins with accurate measurement
of the length of the tube to be inserted into the patient to
reach the stomach. Two methods commonly used by nurses
to determine NGT depth are the Nose→Earlobe→Xiphoid
process→Midline of the Umbilicus (NEMU) method and
the Nose→Earlobe→Xiphoid (NEX) method of measure-
ment. When the 2 methods were evaluated, the NEMU
method demonstrated superior accuracy (97% vs 59%) over
the NEX method for placement in the stomach.24 Upon
closer evaluation, use of the NEX method often resulted in
high esophageal NGT misplacement, increasing the risk of
aspiration and jeopardizing patient safety.24,25
Current practice to verify NGT location is variable
among institutions, patient care units, and providers.2,11,26
Commonly used methods for NGT location verification
include: auscultation, aspiration with visual inspection
of gastric fluids, pH testing of gastric secretions, and
radiography.27 Safety and practice alerts warn against the
use of auscultation28 and visual inspection of gastric
aspirate29 as the means of NGT location verification be-
cause neither method is confirmatory, and either can give
false affirmation of correct NGT placement. Despite these
warnings and practice alerts, recent studies found that these
methods are still widely used by nurses caring for both
pediatric and adult patients.2,30
Irving et al
Use of Radiographs to Verify NGT Placement
The current gold standard to verify NGT placement is a
properly obtained and interpreted radiograph. However,
uncertainty regarding cumulative radiation exposure related
to radiograph frequency,31 as well as concerns over the ac-
curate and consistent interpretation and reporting of NGT
location by both radiologists32,33 and nonradiologists,34
raise questions regarding the use of radiography for NGT
location verification as the gold standard in pediatric pa-
tients. Accurate NGT location by radiographic verification
depends on clearness of the image, interpretation, and
the accuracy and clarity of the radiographic report. The
report should contain information on the path of the NGT
and the exact location of the tube tip that indicates its
readiness for use.32 Concerns of radiation exposure, the
variability of technique, and the lack of standardization
for the amount of radiation used has led many institutions
to use the as low as reasonably achievable35 concept for
pediatric imaging. Given these factors, radiograph is often
not the first-line method used to verify NGT placement in
many children’s hospitals.2 However, it is the standard by
which all other methods of verification are compared for
accuracy in establishing NGT location.
Use of Gastric pH to Verify NGT Placement
Measurement of the acidity of the gastric aspirate is an
evidence-based method used to verify NGT placement.
Commercial products to measure pH from gastric aspi-
rate show variance in measurement increments of 0.5–1.0.
Studies have demonstrated that obtaining a pH ࣘ 5.5 from
gastric aspirate obtained from an NGT is a reliable indicator
that the tube is properly placed in the stomach.4,23,29,36-38
Standard practice in the United Kingdom is to obtain a gas-
tric pH measurement as the primary method to determine
NGT location; a pH value of 1–5.5 is considered indicative
of correct gastric placement for an NGT.22
An issue to consider when using gastric aspirate pH
measurement to determine NGT location is the use of
histamine-2 receptor antagonists (H2 RAs) and/or proton
pump inhibitors (PPIs). These classes of medications lower
gastric acidity and can result in a pH measurement > 5.5,
causing concern about tube misplacement.4 However, a
discrepancy of accurate gastric aspirate pH measurements
when these medications are used has not been validated in
the literature. A recently completed retrospective study of
neonates demonstrated that 97% of 6979 pH measurements
obtained in 1024 infants were ࣘ5.39 Whereas not many of
the infants received a PPI or H2 RA, 95% and 92% of those
who did had pH ࣘ 5, respectively. Additionally, a 2016
retrospective study from a large medical record database
reports a significant increase in H2 RA and PPI prescriptions
in children over a 6-year (2005–2011) period, with children
923
less than 12 months of age having the highest prescription
rate for H2 RA medications.40
A true gastric aspirate is necessary to obtain an accurate
pH measurement; therefore, a second issue to consider
concerns reverification of correct NGT location in those
patients receiving continuous enteral feeding. These patients
may require a period of cessation of formula infusion and
water or air flush of the NGT to ensure accuracy when
obtaining a gastric aspirate to measure pH, although there
are no data available to support or refute this practice.
A limitation to using gastric pH to assess NGT location
arises if the tube is misplaced into the esophagus, where
it is difficult to withdraw secretions. More importantly, if
esophageal placement occurs, the pH of any aspirate may
mimic that of gastric secretions due to aspiration or reflux
at the time of pH testing, which carries a high risk of
false representation of NGT placement. Similarly, distal
migration of the NGT to the level of the pylorus or beyond
can also result in difficulty obtaining a gastric sample to test
pH for accurate gastric placement. Although the use of pH
measurement for tube tip verification is the best evidence-
based method, there are issues—as outlined above—that
need to be considered.
Other Methods Used for NGT Verification
The use of an electromagnetic sensor-guided device for
NGT placement verification in pediatric patients is contro-
versial. A PSA issued in the United Kingdom described 2
patient deaths associated with the use of an electromag-
netic sensor-guided device.41 The PSA mandates gastric pH
measurement to accompany the use of the electromagnetic
sensor-guided device to verify NGT placement. A limitation
associated with this device is tube size. Currently, the
smallest tube that accompanies the device is 8 French, which
is often too large for many pediatric patients. Also, the
external sensor necessary for use with the device may be too
large and too heavy to be placed on the chest of smaller
infants and children. Safety concerns with the use of the
electromagnetic sensor-guided device include injury to the
intestinal intima42 and inconsistency between the actual
tube location and the image projected by the device.43,44 A
US Food and Drug Administration (FDA) alert issued in
January 2018, regarding use of an electromagnetic sensor-
guided device for NGT placement, recommends user train-
ing from the manufacturer with mandated competency in
device operation, and use of an additional method of
NGT location verification.45 Studies in the United States
have demonstrated the use of an electromagnetic sensor-
guided device to be helpful for the placement of transpyloric
feeding tubes in adult and pediatric patients46,47 ; however,
definitive data for successful NGT placement in pediatric
patients is limited.
924
Figure 1. NGT placement and verification decision tree.
NG, nasogastric.
The use of ultrasound technology as a noninvasive
substitute for radiologic imaging to verify NGT placement
shows promise. A recent study in a pediatric intensive
care unit demonstrated 100% sensitivity with the use of
ultrasound for correct placement of NGTs at the bedside
when operated by a radiologist.48 Further investigation
into the feasibility and applicability of ultrasound to verify
NGT location in children at the bedside by nonradiologists
Nutrition in Clinical Practice 33(6)
is needed. Its portability, absence of radiation, and
noninvasive properties make ultrasound a potentially
useful method for verifying NGT location. An additional
method for verification of NGT placement, capnography,
has demonstrated enteral placement with 98% accuracy
in one study49 ; however, it is not currently recommended
to be used as an independent method to verify NGT
placement.
Irving et al
Recommendations
Based on the available evidence and as outlined in Figure
1, the following are recommendations for best practice
standards to verify NGT location in pediatric patients:
r Provide education
- Education should be provided for all clinicians
placing NGTs within institutions and across
care settings.
- Education should include competency valida-
tion for placement, pH measurement, decision
making to determine need for radiographic eval-
uation, documentation of tube placement, and
patient tolerance of the procedure.
- Competency-based education should be in place
for providers interpreting radiographs to verify
NGT placement.
r Use appropriate NGT placement and securing
methods.
- Use the Nose→Earlobe→Xiphoid process→
Midline of the Umbilicus (NEMU) method for
determination of NGT insertion length.
- Document the centimeter marking on the tube,
where it exits the nose or mouth, once cor-
rect tube placement is confirmed periodically
depending on the policy of the healthcare set-
ting.
- In an NGT with a stylet in place prior to inser-
tion, if the NGT has been flushed with sterile
water to facilitate stylet lubrication and removal
after insertion, aspirate the entire fill volume of
sterile water and discard. A second aspiration
is necessary to obtain gastric secretions for pH
testing.
r Measure gastric pH.
- Use gastric pH testing as the first-line method
for NGT location verification.
- A gastric pH value of 1–5.5 without a change
in the patient’s clinical status is indicative of
gastric placement.
- When used intermittently for enteral feedings
and/or medication administration, establish a
schedule for frequency of NGT location confir-
mation.
- When used continuously for enteral feeding, de-
termine the need for frequency of confirmation
with documentation of NGT location.
r Consider a radiograph for any patient in whom there
is any concern for correct NGT placement, such as:
925
- Difficulty placing the NGT
- NGT placement in any patient at high risk
of misplacement. This includes those with
known history of facial fractures, neurologic
injury/insult/baseline abnormality, respiratory
concerns, decreased or absent gag reflex, and
those who are critically ill.
- In any patient whose condition deteriorates
shortly after NGT placement
r Improve interpretation and communication about
the radiograph.
- The radiograph requisition should clearly re-
quest “NGT placement verification” or similar
language.
- The radiograph report should contain a state-
ment of the tube path, the location of the tube
tip, and confirmation that the tube is positioned
in the desired location and is appropriate for use.
Future Considerations
Challenges persist surrounding the placement and location
verification of NGTs in pediatric patients. The following are
potential solutions to be considered:
r Adoption of these recommendations with units and
institutions implementing them to meet their specific
needs
r Partnering with key stakeholders for technology and
product development to allow for placement verifica-
tion and reverification in real time for the duration of
NGT use
r Research on best practice for ongoing reverification
of correct NGT placement, including data on fre-
quency of assessment for reverification of correct
NGT location for patients on intermittent feedings
and/or continuous feedings and/or NGT medication
administration
r Standardized, mandated, state and federal reporting
mechanisms for NGT misplacements, followed by a
root cause analysis to evaluate the misplacement and
to ascertain opportunities for improved education
and practice
Conclusions
Placement of NGTs in pediatric patients to facilitate provi-
sion of EN, medication administration, and fluid infusion
is a common practice performed by nurses without the use
of technology. Despite the frequency of NGT placement, it
carries the potential of patient harm if the tube is misplaced.
Various methods for determining location verification exist.
Radiography is the gold standard and measurement of
926
gastric pH is a validated method; however, there is no
standardization in the United States on best practice for
NGT location verification. The recommendations presented
here are a necessary first step in establishing best prac-
tice related to NGT placement verification in the pedi-
atric patient. The future considerations present numerous
opportunities for collaboration between nurse scientists,
clinicians, researchers, institutional leaders, policy makers,
and industry partners to improve patient safety related to
NGT placement and verification in children.
References
1. Healthcare Cost and Utilization Project (HCUP). Overview of
the National (Nationwide) Inpatient Sample (NIS). Agency for
Healthcare Research and Quality. 2014. Available at: www.hcup-us.
ahrq.gov/nisoverview.jsp. Accessed August 24, 2017.
2. Lyman B, Kemper C, Northington L, et al. Use of temporary
enteral access devices in hospitalized neonatal and pediatric pa-
tients in the United States. JPEN J Parenter Enteral Nutr. 2016;40:
574–580.
3. Cirgin Ellett ML, Cohen MD, Perkins SM, Smith CE, Lane KA,
Austin JK. Predicting the insertion length for gastric tube placement
in neonates. J Obstet Gynecol Neonatal Nurs. 2011;40:412–421.
4. Ellett ML, Croffie JM, Cohen MD, Perkins SM. Gastric tube place-
ment in young children. Clin Nurs Res. 2005;14:238–252.
5. Baskin WN. Acute complications associated with bedside placement of
feeding tubes. Nutr Clin Pract. 2006;21:40–55.
6. Yardley IE, Donaldson LJ. Patient safety matters: reducing the risks of
nasogastric tubes. Clin Med (Lond). 2010;10:228–230.
7. Daly W. Unusual complication of nasal intubation: report of case.
Anesthesiology. 1953;14:96.
8. Lind LJ, Wallace DH. Submucosal passage of a nasogastric tube
complicating attempted intubation during anesthesia. Anesthesiology.
1978;49:145–147.
9. Aronchick JM, Epstein DM, Gefter WB, Miller WT. Pneumotho-
rax as a complication of placement of a nasoenteric tube. JAMA.
1984;252:3287–3288.
10. Bankier AA, Wiesmayr MN, Henk C, et al. Radiographic detection
of intrabronchial malpositions of nasogastric tubes and subsequent
complications in intensive care unit patients. Intensive Care Med.
1997;23:406–410.
11. Metheny NA, Meert KL, Clouse RE. Complications related to feeding
tube placement. Curr Opin Gastroenterol. 2007;23:178–182.
12. Crisp CL. Esophageal nasogastric tube misplacement in an infant
following laser supraglottoplasty. J Pediatr Nurs. 2006;21:454–455.
13. Ellett ML, Cohen MD, Croffie JM, Lane KA, Austin JK, Perkins SM.
Comparing bedside methods of determining placement of gastric tubes
in children. J Spec Pediatr Nurs. 2014;19:68–79.
14. De Espinosa H, Garcı́a de Paredes C. Traumatic perforation of the
pharynx in a newborn baby. J Pediatr Surg. 1974;9:247–248.
15. Wynne DM, Borg HK, Geddes NK, Fredericks B. Nasogastric tube
misplacement into Eustachian tube. Int J Pediatr Otorhinolaryngol.
2003;67:185–187.
16. Arslantas A, Durmaz R, Cosan E, Tel E. Inadvertent insertion of
a nasogastric tube in a patient with head trauma. Childs Nerv Syst.
2001;17:112–114.
17. Creel AM, Winkler MK. Oral and nasal enteral tube placement errors
and complications in a pediatric intensive care unit. Pediatr Crit Care
Med. 2007;8:161–164.
18. Glüer S, Schmidt AI, Jesch NK, Ure BM. Laparoscopic repair of
neonatal gastric perforation. J Pediatr Surg. 2006;41:e57–58.
Nutrition in Clinical Practice 33(6)
19. Mattar MS, al-Alfy AA, Dahniya MH, al-Marzouk NF. Urinary
bladder perforation: an unusual complication of neonatal nasogastric
tube feeding. Pediatr Radiol. 1997;27:858–859.
20. Wallace SC. Data Snapshot: Complications Linked to Iatrogenic En-
teral Feeding Tube Misplacements. PA Patient Saf Advis. 2017;14:1-60.
21. Kemper C, Northington L, Wilder K, Visscher D. A call to action:
the development of enteral access safety teams. Nutr Clin Pract.
2014;29:264–266.
22. National Health Service. Nasogastric tube misplacement: continuing
risk of death and severe harm. Patient Safety Alert. 2016. NHS/PSA/
RE/2016/006. Available at: https://improvement.nhs.uk/news-alerts/na
sogastric-tube-misplacement-continuing-risk-of-death-severe-harm/.
Accessed November 7, 2017.
23. National Health Service. Placement devices for nasogastric tube inser-
tion do not replace initial position checks. Patient Safety Alert. 2013.
NHS/PSA/W/2013/001. Available at: https://www.england.nhs.uk/
wp-content/uploads/2013/12/psa-ng-tube.pdf.
24. Ellett ML, Cohen MD, Perkins SM, Croffie JM, Lane KA, Austin JK.
Comparing methods of determining insertion length for placing gastric
tubes in children 1 month to 17 years of age. J Spec Pediatr Nurs.
2012;17:19–32.
25. Klasner AE, Luke DA, Scalzo AJ. Pediatric orogastric and nasogastric
tubes: a new formula evaluated. Ann Emerg Med. 2002;39:268–272.
26. Simons SR, Abdallah LM. Bedside assessment of enteral tube place-
ment: aligning practice with evidence. Am J Nurs. 2012;112:40–46; quiz
48, 47.
27. Irving SY, Lyman B, Northington L, Bartlett JA, Kemper C. Nasogas-
tric tube placement and verification in children: review of the current
literature. Nutr Clin Pract. 2014;29:267–276.
28. (NACH) NACH. Blind Pediatric NG Tube Placements Continue to
Cause Harm. Overland Park, KS: Child Health Patient Safety Orga-
nization, Inc.; 2012.
29. Metheny N. AACN Practice Alert. Initial and ongoing verification of
feeding tube placement in adults (applies to blind insertions and place-
ments with an electromagnetic device). Critl Care Nurse. 2016;36(2):
e8–e13.
30. Metheny NA, Stewart BJ, Mills AC. Blind insertion of feeding tubes
in intensive care units: a national survey. Am J Crit Care. 2012;21:
352–360.
31. Kleinerman RA. Cancer risks following diagnostic and therapeu-
tic radiation exposure in children. Pediatr Radiol. 2006;36 Suppl 2:
121–125.
32. Cohen MD, Ellett M. Quality of communication: different patterns of
reporting the location of the tip of a nasogastric tube. Acad Radiol.
2012;19:651–653.
33. Rollins H, Arnold-Jellis J, Taylor A. How accurate are X-rays to check
NG tube positioning? Nurs Times. 2012;108:14–16.
34. Lee KH, Cho HJ, Kim EY, et al. Variation between residents and
attending staff interpreting radiographs to verify placement of nutri-
tion access devices in the neonatal intensive care unit. Nutr Clin Pract.
2015;30:398–401.
35. Don S. Pediatric digital radiography summit overview: state of confu-
sion. Pediatr Radiol. 2011;41:567–572.
36. Gilbertson HR, Rogers EJ, Ukoumunne OC. Determination of a
practical pH cutoff level for reliable confirmation of nasogastric tube
placement. JPEN J Parenter Enteral Nutr. 2011;35:540–544.
37. Stock A, Gilbertson H, Babl FE. Confirming nasogastric tube position
in the emergency department. Pediatr Emerg Care. 2008;24:805–809.
38. Metheny NA, Pawluszka A, Lulic M, Hinyard LJ, Meert KL. Test-
ing placement of gastric feeding tubes in infants. Am J Crit Care.
2017;26:466–473.
39. Kemper C, Haney B, Oschman A, et al. Use of pH testing to verify
enteral feeding tube placement in neonates. Submitted to Advances in
Neonatal Care, 2018.
Irving et al
40. Faden HS, Ma C. Trends in oral antibiotic, proton pump inhibitor,
and histamine 2 receptor blocker prescription patterns for children
compared with adults: implications for clostridium difficile infection in
the community. Clinical Pediatrics. 2016;55: 712–716.
41. National Health Service. Nasogastric tube misplacement: continuing
risk of death and severe harm. Patient Safety Alert. 2016. NHS/PSA/
RE/2016/006. Available at: https://improvement.nhs.uk/news-alerts/n
asogastric-tube-misplacement-continuing-risk-of-death-severe-harm/.
Accessed March 20, 2018.
42. U.S. Food and Drug Administration. MAUDE – Manufacturer and
user facility device experience. us department of health and hu-
man services. 2017. Available at: https://www.accessdata.fda.gov/scripts
/cdrh/cfdocs/cfmaude/search.cfm. Accessed August 22, 2017.
43. Bryant V, Phang J, Abrams K. Verifying placement of small-bore feed-
ing tubes: electromagnetic device images versus abdominal radiographs.
Am J Crit Care. 2015;24:525–530.
44. Metheny NA, Meert KL. Update on effectiveness of an electromagnetic
feeding tube-placement device in detecting respiratory placements. Am
J Crit Care. 2017;26:157–161.
927
45. Drug USFa, (FDA) A, Division of Industry and Consumer Ed-
ucation. Feeding Tube Placement Systems, Letter to Health Care
Providers. Silver Spring, MD: U.S. Food and Drug Administration
2018. Available at: https://www.fda.gov/MedicalDevices/Safety/Lett
erstoHealthCareProviders/ucm591838.htm.
46. Kline AM, Sorce L, Sullivan C, Weishaar J, Steinhorn DM. Use
of a noninvasive electromagnetic device to place transpyloric feeding
tubes in critically ill children. Am J Crit Care. 2011;20:453–459;
quiz 460.
47. Brown AM, Perebzak C, Handwork C, Gothard MD, Nagy K. Use of
electromagnetic device to insert postpyloric feeding tubes in a pediatric
intensive care unit. Am J Crit Care. 2017;26:248–254.
48. Atalay YO, Aydin R, Ertugrul O, Gul SB, Polat AV, Paksu MS. Does
bedside sonography effectively identify nasogastric tube placements in
pediatric critical care patients? Nutr Clin Pract. 2016.
49. Chau JP, Lo SH, Thompson DR, Fernandez R, Griffiths R. Use
of end-tidal carbon dioxide detection to determine correct place-
ment of nasogastric tube: a meta-analysis. Int J Nurs Stud. 2011;48:
513–521.