Clinical Nutrition ESPEN xxx (2018) e1ee8

Contents lists available at ScienceDirect

Clinical Nutrition ESPEN

journal homepage: http://www.clinicalnutritionespen.com

Original article

Assessing the appropriateness of parenteral nutrition use in

hospitalized patients. A comparison on parenteral nutrition bag
prescription in different wards and nutritional outcomes
Monica Laura Ponta a, *, Laura Rabbione a, Cristina Borgio a, Eliana Quirico a,
Elena Patrito a, Maria Novella Petrachi a, Elisabetta Girotto a, Marisa Sillano a,
Silvio Geninatti b, Michela Zanardi a, Andrea Pezzana a
a
Department of Dietetics and Clinical Nutrition, San Giovanni Bosco Hospital, Piazza del Donatore di Sangue, 3, 10154, Turin, Italy
b
Statistical and Epidemiological Research Unit, Local Health Service Turin ASL TO2, Corso Svizzera, 185, 10149, Turin, Italy

a r t i c l e i n f o s u m m a r y

Article history: Purpose: Our aim is to assess parenteral nutrition (PN) bag prescription in hospitalized patients and
Received 21 December 2016 evaluate clinical outcomes linked to PN therapy.
Accepted 29 January 2018 Methods: We performed an observational longitudinal retrospective study on PN prescription in a
General Public Hospital in Turin, Italy, on ninety-five patients receiving PN prescribed by the Nutrition
Keywords: Support Team (NST). We described patients' demography and assessed nutritional outcomes, as well as
Nutritional bag prescription
PN bag prescription in different wards. Medians were calculated for several clinical parameters before
Parenteral nutrition regimen
and after PN therapy. A z-test for proportions has been performed to better understand the impact of
Clinical outcomes
Nutritional status
various conditions on clinical outcomes and to compare differences between administered nutrients and
Nutrition support team required amounts.
Results: The NST resulted responsible for only 18% of bags prescribed in the geriatrics ward and for 48% in
the surgery wards. PN was not able to fulfill nutritional requirements resulting in a median lack of 3.1
calories and 0.23 g of proteins per kilogram of reference body weight per day. Despite this, PN therapy
was able to improve total blood proteins and calcium blood levels in our cohort. The NST changed the
prescription in 55.8% of the pre-existing PN regimens.
Conclusions: More strict adherence to guidelines is needed in order to maximize effectiveness of PN and
observe a positive impact on clinical parameters.
© 2018 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights
reserved.

1. Introduction subjects, is representative of the Italian population and it is aligned

Malnutrition is a condition related to significantly higher length
of hospital stay (LOS), costs and subsequent home health care needs
in hospitalized patients compared with a good nutritional status
[1]. A high number of hospitalized patients is malnourished or at
risk for malnourishment. The Project: Iatrogenic MAlnutrition in
Italy (PIMAI) showed that prevalence of malnutrition in Italian
hospitalized patients is 30.7% [2]. This data, obtained from 1583
* Corresponding author. Department of Medicine, Division of Gastroenterology,
Toronto General Hospital Research Institute e University Health Network, 200
Elizabeth Street, Toronto, ON, M5G 2C4, Canada.
E-mail address: monicalaura.ponta@gmail.com (M.L. Ponta).
with the European prevalence of malnutrition of 31.1% [3]. Amer-
ican studies offer a heterogeneous picture of the prevalence of
malnutrition in hospital settings with the prevalence ranging from
30% to 55% [4e7]. Appropriately selected nutritional support,
including parenteral nutrition (PN), can address the problem of
malnutrition, improve clinical outcomes and help to reduce the
costs of health care [8]. Enteral feeding is the most physiologic way
of nourishment and should be preferred when a well functioning
gastrointestinal tract is available. However, there are conditions in
which oral/enteral nutrition is not possible, such as major upper
gastrointestinal surgery, high output gastrointestinal fistula, diffuse
peritonitis, intestinal obstruction, ileus, intractable vomiting or
diarrhea, gastrointestinal ischemia. In all these cases it is

https://doi.org/10.1016/j.clnesp.2018.01.072
2405-4577/© 2018 European Society for Clinical Nutrition and Metabolism. Published by Elsevier Ltd. All rights reserved.

Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
e2 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8

mandatory to correct or prevent malnutrition with a parenteral With this information we calculated the following parameters:
approach [9]. Several nutritional bags are available to be used in length of stay (LOS), number of days between surgery and the first
hospital settings, for both central and peripheral access. A variety of medical advice (when applicable), body mass index (BMI), duration
different compositions allow the physician to choose the bag that of PN therapy. In relation to the blood tests, we compared patient's
better fulfills patient's requirements. Since parenteral nutrition values with reference values of the hospital laboratory (Table 2).
(PN) therapy may be quite expensive (even if the three- We identified values that were in range and those that were
compartment bags used in our hospital are more economical under or above the range. We also asked the hospital pharmaceu-
than the customized ones [10]), it is mandatory to choose the best tical service, the amount of PN bags used by the geriatrics and
possible bag type and volume in order to maximize this nutritional general surgery wards in order to obtain the number of bags pre-
therapy. The aim of this study is to assess the appropriate use of PN scribed by the attending physicians compared to those prescribed
therapy in hospitalized patients and evaluate clinical outcomes on by the NST. Lastly, we obtained the number of all the inpatients of
patients followed by the Nutrition Support Team (NST) at San each ward in the study period.
Giovanni Bosco Hospital in Turin, Italy.
2.2. Statistical analysis
2. Materials and methods
Statistical analysis was performed using SAS version 9.2.2. We
characterized our cohort evaluating the distribution of gender, age
2.1. Participants and data
ranges (<30, 30e49, 50e64, 65e80, >80), admitting ward, admit-
ting discipline (surgical or medical with sub-group of geriatrics),
We performed an observational analytical longitudinal retro-
number of comorbidities divided into ranges (0, 1, 2, 3, >3), route of
spective study on PN prescription at San Giovanni Bosco Hospital,
administration of PN, PN regimen, the prevalence of surgery and
Turin, Italy in a twelve months period (from June 1st 2014 to May
lastly the prevalence of pre-existing PN therapy (with the per-
30th 2015). We identified eleven wards of interest: geriatric, in-
centages of confirmation, or of modification of PN therapy). We also
ternal medicine A and B, psychiatry, emergency medicine,
calculated the mean ± standard deviation, or the median with
neurology, cardiology, acute coronary care unit, general surgery,
interquartile range when appropriate, of age, BMI and reference
urology and vascular surgery. We excluded nephrology and dialysis
body weight.
wards because of the specific needs in this particular population.
To evaluate patient and nutritional outcomes, we calculated the
Starting from June 1st, we selected all the patients receiving PN
median LOS and the median of duration of PN, the median amount
prescribed by the NST. We included both patients that were already
of days between surgery and the medical consult (with two sub-
on PN before the referral and those with a new PN onset prescribed
groups: medical consult performed before or after surgery). We
by the specialist. We collected medical records of ninety-five pa-
identified the prevalence of patients who needed a subsequent
tients from the Hospital Information System Galileo 1.5.3.3.2787
medical consult after the first one. We evaluated the patient's
(NoemaLife S.p.A.) and from the worksheets used during the visits
malnutrition status considering the prevalence of patients with
where the NST recorded nutritional data.
total blood protein before PN being <6.6 g/dl. Then we sorted out
We set up a database with the following information: gender,
age, ward of admission, date of admission and discharge, admitting
disciplines (medical or surgical ward), date of surgery (if appli- Table 2
cable), number of comorbidities, date of the first and second (if Reference values of blood tests.
applicable) medical advice, actual and reference body weight, Parameter Measure unit Minimum Maximum
height, calories and proteins requirements, type of PN regimen
Lymphocytes /UL 1500 e
already prescribed by attending physicians (if applicable: type of Glucose mg/dl 60 100
bag, volumes infused, additions, calories and proteins adminis- Creatinine mg/dl 0.5 0.9
tered; Table 1), date of start and stop of PN, route of PN adminis- Sodium mEq/l 136 145
tration (peripheral or central), type of PN regimen prescribed by the Potassium mEq/l 3.5 5.1
Magnesium mEq/l 1.22 2.14
NST (type of bag, volumes infused, additions, total calories and
Calcium mmol/l 2.15 2.50
proteins administered, calories and proteins per kilogram of Phosphorus mg/dl 2.5 4.5
reference weight), blood tests before starting PN and at the end of Total protein g/dl 6.6 8.7
the hospitalization (lymphocytes, glucose, creatinine, sodium, po- Albumin g/l 35 50
tassium, magnesium, calcium, phosphorus, total protein, albumin, Transferrin mg/dl 200 360
Reactive C protein mg/dl e 0.5
transferrin, C reactive protein).

Table 1
Composition of parenteral nutrition bags.

Name Vol (ml) AA (gr) Lip (gr) Hc (gr) Total kcal Na (mMol) K (mMol) Ca (mMol) Mg (mMol) P (mMol) Osm (mOsm/l)

Nutriperilipid 1875 60 75 120 1435 75 45 4.5 4.5 11.25 920

Olimel N4 1500 38 45 112 1050 31.5 24 3 3.3 12.7 760
Olimel N5 2000 66.8 80 230 1980 70 60 7 8 30 1120
Olimel N7 1500 66.4 60 210 1710 52.5 45 5.3 6 22.5 1360
Olimel N9 2000 113.9 80 220 2140 0 0 0 0 6 1170
Smofkabiven 986 50 38 125 1100 40 30 2.5 5 12 1500
Smofkabiven 1477 75 56 187 1600 60 45 3.8 7.5 19 1500
Smofkabiven 1970 100 75 250 2200 80 60 5 10 25 1500

Vol: bag volume, AA: amino acids, Lip: lipids, Hc: carbohydrates. Amounts provided per bag. Nutriperilip contains 50% soybean oil and 50% medium chain triglycerides (MCT)
as source of lipids, Olimel contains 80% olive oil and 20% soybean oil as source of lipids, Smofkabiven contains 30% soybean oil, 30% MCT, 25% olive oil and 15% fish oil as source
of lipids.

Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8 e3

the percentages of malnourished patients at baseline whose total
blood proteins had been corrected after PN therapy. We calculated
the median required amounts of calories and proteins and we
compared them with the median amounts of calories and proteins
administered. Then we calculated the difference of the two sets of
data. A z-test for proportions has been performed in order to
evaluate the statistical significance of the differences. We also
compared the median amounts of calories and proteins given with
the central venous catheter and the peripheral venous catheter. We
quantified the prevalence of bag adjustments made from the NST
and those made from the ward physician on pre-existing PN pre-
scription (Table 4).
We calculated the median values (±standard deviation) of each
of the twelve blood tests before and after PN. A z-test for pro-
portions has been performed in order to verify the statistical sig-
nificance of the differences with the purpose of verifying the real
improvement of each blood test (Table 5). Then we calculated the
prevalence of patients whose blood tests were in range or below/
above the range before and after PN therapy to quantify how many
patients improved.
To better understand the impact of certain conditions, we also
compared blood outcomes between different groups of patients:
patients receiving PN via central vs. peripheral venous catheter;
adjustment made to the bag (yes vs. no); length of stay, shorter or

Table 3
Blood test outcomes.

Variable Before PN therapy After PN therapy Statistics

Mean total blood proteins (±SD; range) 5.4 (±1.00; 3.4e8.9) [n ¼ 86] 5.9 (±0.94; 3.3e7.6) [n ¼ 75]
Total blood proteins, n (%)
Low 76 (88.3%) 53 (70.7%)
In range 9 (10.5%) 22 (29.3%) p ¼ 0.001
High 1 (1.2%) 0
Mean blood creatinine (±SD; range) 1.0 (±0.84; 0.2e6.0) [n ¼ 92] 1.1 (±0.87; 0.3e5.6) [n ¼ 85]
Blood creatinine, n (%)
Low 15 (16.3%) 7 (8.2%)
In range 48 (52.2%) 48 (56.5%) p ¼ 0.283
High 29 (31.5%) 30 (35.3%)
Mean blood albumin (±SD; range) 26.0 (±6.21; 17.0e41.6) [n ¼ 36] 26.0 (±6.27; 12e40) [n ¼ 26]
Blood albumin, n (%)
Low 32 (88.9%) 23 (88.5%)
In range 4 (11.1%) 3 (11.5%) p ¼ 0.479
High 0 0
Mean blood transferrin (±SD; range) 151.5 (±41.05; 61e240) [n ¼ 18] 151.0 (±31.18; 84e209) [n ¼ 15]
Blood transferrin, n (%)
Low 18 (100%) 15 (100%)
In range 0 0 n.d.
High 0 0
Mean blood CRP (±SD; range) 6.3 (±7.19; 0e32.8) [n ¼ 72] 4.3 (±7.15; 0e32.8) [n ¼ 72]
Blood CRP, n (%)
In range 5 (6.9%) 4 (5.6%) n.d.
High 67 (93.1%) 68 (94.4%)
Mean blood lymphocytes (±SD; range) 1130 (±698.7; 310e3530) [n ¼ 84] 1220 (±751.6; 210e4100) [n ¼ 79]
Blood lymphocytes, n (%)
Low 61 (72.6%) 49 (62.0%)
In range 23 (27.4%) 30 (38.0%) p ¼ 0.075
Mean blood glucose (±SD; range) 114.5 (±38.29; 60e271) [n ¼ 84] 111.5 (±60.26; 54e387) [n ¼ 62]
Blood glucose, n (%)
Low 0 1 (1.4%)
In range 23 (27.4%) 23 (31.9%) p ¼ 0.267
High 61 (72.6%) 48 (66.7%)
Mean blood sodium (±SD; range) 138 (±7.5; 112e159) [n ¼ 92] 137 (±6.9; 122e165) [n ¼ 88]
Blood sodium, n (%)
Low 35 (38.0%) 29 (33.0%)
In range 44 (47.9%) 50 (56.8%) p ¼ 0.114
High 13 (14.1%) 9 (10.2%)
Mean blood potassium (±SD; range) 4.1 (±0.67; 2.5e5.7) [n ¼ 91] 4.5 (±0.76; 2.2e6.7) [n ¼ 88]
Blood potassium, n (%)
Low 17 (18.5%) 6 (6.8%)
In range 68 (73.9%) 69 (78.4%) p ¼ 0.240
High 7 (7.6%) 13 (14.8%)
Mean blood calcium (±SD; range) 2.0 (±0.20; 1.5e2.5) [n ¼ 64] 2.1 (±0.47; 1.7e5.7) [n ¼ 69]
Blood calcium, n (%)
Low 49 (76.6%) 37 (53.7%)
In range 15 (23.4%) 29 (42.0%) p¼0.011
High 0 3 (4.3%)
Mean blood magnesium (±SD; range) 1.6 (±0.47; 0.7e2.9) [n ¼ 21] 1.7 (±0.31; 1.0e2.2) [n ¼ 24]
Blood magnesium, n (%)
Low 3 (14.3%) 2 (8.3%)
In range 16 (76.2%) 21 (87.5%) p ¼ 0.161
High 2 (9.5%) 1 (4.2%)
Mean blood phosphorus (±SD; range) 3.0 (±0.78; 1.8e3.8) [n ¼ 7] 3.9 (±1.22; 3.2e6.0) [n ¼ 4]
Blood phosphorus, n (%)
Low 2 (28.6%) 0
In range 5 (71.4%) 3 (75%) n.d.
High 0 1 (25%)

Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
e4 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8

Table 4
Bag consumption in geriatrics and surgery wards.

Bag type Total geriatrics Prescribed by attending Prescribed Total surgery Prescribed by attending Prescribed by
bags (n) physicians, n (%) by NST, n (%) bags (n) physicians, n (%) NST, n (%)

Nutriperilipid 91 57 (62.6) 34 (37.4) 184 184 (100) 0 (0)

Olimel N4 532 440 (82.7) 92 (17.3) 180 41 (22.8) 139 (77.2)
Olimel N5 0 0 (0) 0 (0) 76 4 (5.3) 72 (94.7)
Olimel N7 4 4 (100) 0 (0) 168 110 (65.5) 58 (34.5)
Olimel N9 24 24 (100) 0 (0) 24 24 (100) 0 (0)
Smofkabiven 986 0 0 (0) 0 (0) 8 8 (100) 0 (0)
Smofkabiven 1477 0 0 (0) 0 (0) 124 0 (0) 124
Smofkabiven 1970 48 48 (100) 0 (0) 224 142 (63.4) 82 (36.6)
Totals 863 737 (85.4) 126 (18.0) 988 513 (51.9) 475 (48.1)

Table 5 the total patients evaluated, fifty-two (54.7%) already had a PN bag
Prevalence of patients seen by Nutrition Support Team. prescribed by the attending physicians. Of these pre-existing PN
Ward Total inpatients/year n Patients evaluated therapies, the NST confirmed twenty-three (44.2%) and modified
by NST n (%) the remaining twenty-nine (55.8%). A second medical advice has
Geriatric 525 7 (1.3%) been performed in 55% of our patients. The mostly used route of
Internal medicine A 989 12 (1.2%) parenteral nutrition administration was through central vein that,
Internal medicine B 794 15 (1.9%) however, accounted for only 54.7% of subjects, while in the
Psychiatry 396 1 (0.3%)
remaining 45.3% a peripheral vein has been used. The median
Emergency medicine 433 12 (2.8%)
Neurology 473 3 (0.6%) duration of PN therapy was 13 days. Thirty-three patients out of 89
Cardiology 1356 1 (0.1%) received PN for more than fifteen days, 13 of these patients (39.4%)
Acute coronary care unit 105 1 (1.0%) had a peripheral catheter. Table 3 shows also the prevalence of
General surgery 1130 31 (2.7%)
different PN regimen prescribed by the NST at the time of the first
Urology 1196 5 (0.4%)
Vascular surgery 403 7 (1.7%)
medical advice. The mostly prescribed bags were Olimel N4 for the
peripheral route and Smofkabiven 1477 for the central route.

3.2. Patient and nutritional outcomes

greater than the mean LOS; duration of PN therapy shorter of
greater than the mean duration (Supplemental Tables 1e4). We
also compared the number of bags prescribed by the NST with the
total amount of bags prescribed by attending physicians of geriat-
rics and surgery wards, obtaining the prevalence of nutrition bags
prescribed by the specialist in each ward.
Lastly, we compared the number of patients followed by the NST
with the number of all the inpatients hospitalized in the study
period for each ward. Results were considered statistically signifi-
cant when p < 0.05.
3. Results
3.1. Characterization of study subjects
Demographic data (Supplemental Table 1) showed no statisti-
cally significant differences between genders. The median age was
75 (ranging from 18 to 90).
Patients were distributed in different age ranges: 3.2% were
under 30 years of age, 11.6% ranged from 30 to 49 while 17.9% were
between 50 and 64, the higher prevalence of subjects (49.4%) were
65e80 years old and the remaining 17.9% were over 80. In the
occasion of the medical advice, the study subjects were hospital-
ized in different wards with the following distribution: geriatrics
7.4%, internal medicine A 12.6%, internal medicine B 15.7%, psy-
chiatry 1.1%, emergency medicine 12.6%, neurology 3.2%, cardiology
1.1%, acute coronary care unit 1.1%, general surgery 32.5%, urology
5.3% and vascular surgery 7.4%. Only two patients (2.1%) had no
comorbidity, eleven (11.6%) had only one comorbidity, twenty-
eight (29.5%) had two comorbidities, sixteen (16.8%) had three
comorbidities and the remaining thirty-eight (40.0%) had three or
more comorbidities. The median reference body weight was
64.0 kg with a range from 35.0 kg to 95.0 kg. The mean body mass
index calculated on the reference body weight was 22.6 kg/m2. The
actual body weight has been recorded in 42% of subjects only. Of
We observed that the median LOS was 26 days with a wide
range from a minimum of 6 days and a maximum of 171 days.
For patients who underwent surgery (51.6% of our cohort), the
medical advice had been performed preoperatively in 16.3% of
patients, with a median of 6.5 days in advance, and after the sur-
gery, in the remaining 83.7% of subjects with a median of 8.0 days
following surgery. Malnutrition prevalence, was calculated using
total blood proteins below 6.6 g/dl as a surrogate, was 88% before
PN therapy. Only 23.8% of these patients improved their nutritional
status, while in the remaining 76.2%, malnutrition persisted.
Considering patients' nutritional needs, the average amount of
calories per kilogram per day was estimated as being 27 kcal/kg/
day (range from 21 to 44), while for the protein requirements the
median was 1.26 g/kg/day with a range from 0.83 to 1.93. The
amounts given with the PN therapy were 24 kcal/kg/day (range
9e41) and 1.03 g/kg/day for the proteins (range 0.42e1.89). Results
showed a median lack of 3.1 calories and 0.23 g of proteins per
kilogram of reference body weight per day. The differences are
statistically significant, as confirmed by p-values obtained with t-
test. As expected, the median amount of both calories and proteins
given via the central catheter is higher (27.1 kcal/kg/day and 1.15 g
of proteins/kg/day respectively) than those given through the pe-
ripheral catheter (19.1 kcal/kg/day and 0.75 g of proteins/kg/day
respectively). When taking into consideration adjustments made to
bags, we observed that 76.8% of bag prescribed by the NST had also
some sort of supplement added while only 3.8% of bags prescribed
by attending physicians had supplements. For a summary of data
presented here, see Supplemental Table 2.
3.3. Blood test outcomes
As shown in Table 3, the analysis of blood tests suggests a trend
in increased prevalence of “in range” values after PN therapy,
compared with prevalence before PN therapy. However, the

Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8
improvement was confirmed as statistically significant (p < 0.05)
only for total blood proteins (p ¼ 0.001) and blood calcium
(p ¼ 0.011).
We also compared blood parameters' changes in different
conditions in order to see if any could have had an impact on
clinical outcomes at the end of PN therapy (Supplemental Tables 3,
4, 5 and 6).
First, we divided our patients into two groups, one received
adjustments to their bag, while the other didn't. The mean values of
the blood test however were not significantly different between the
two groups. In a similar way, we compared blood test outcomes
between central and peripheral vein access, finding out a statisti-
cally significant difference in mean transferrin and C reactive pro-
tein values: data unexpectedly showed a better outcome for
peripheral access (lower CRP and higher transferrin with p ¼ 0.018
and p ¼ 0.042 respectively). The LOS was not able to impact any of
the measured blood tests, while for duration of PN therapy a sta-
tistically significant difference in mean transferrin is present, with
higher transferrin in the shorter duration group (p ¼ 0.013).
3.4. Consumption of PN bags in different wards
The total amount of PN bags prescribed in geriatrics and surgery
wards was respectively 863 and 988 in the study period.
The NST prescribed 126 bags in the geriatrics ward (18.0%),
confirming the pre-existing therapy for four out of seven patients,
modifying it in two patients and starting the PN in only one patient.
In the surgery ward, the NST prescribed 475 bags (48.1%). For ten
patients, pre-existing therapy was confirmed, nine were modified,
and in the remaining twelve patients, PN was started from the
Team.
When considering the total number of patients admitted into
each ward during the year of observation (Table 5), the higher
prevalence of nutritional consult has been made in the Emergency
Medicine and General Surgery wards (2.8% and 2.7% respectively). A
prevalence between 1% and 2% has been found in the Geriatric,
Internal Medicine A and B, Acute Coronary Care Unit and Vascular
Surgery. Less than 1% of patients received the NST consultation in
the Psychiatry, Cardiology and Urology wards.
4. Discussion
4.1. Demographics and clinical characteristics
If we consider the prevalence of all subjects over 65 years of age,
the result is that they represent 67.3% of all our medical advices.
This may be due to the higher prevalence of malnutrition seen in
the elderly compared to younger people or either to the fact that
the elderly are hospitalized more often [11]. In addition, almost all
subjects (86.3%) had at least two comorbidities. Considering the
median age of our cohort it is not surprising to have many different
diseases associated among the elderly. When considering the
admitting ward, the higher prevalence of subjects was hosted in the
general surgery ward (32.5%), but if we consider all the internist
wards, the cumulative prevalence is higher (54.8%) compared to the
surgery wards (45.2%).
We do have a reference body weight for each single patient as
we used this data to obtain the estimated nutritional requirements,
however we did not have the actual body weight for more than half
of the patients (data missing in 58% of patients). This was due to the
fact that it was not possible for the patient to stand on a scale, and a
scale-bed was not available in the ward. This lack of equipment
represents a big issue, since not only nutritional therapy needs a
precise body weight, but also simple hydration and many drug
regimens are based on the actual weight of patients.
Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
e5
In more than half of the patients, the NST evaluated the
appropriateness of a pre-existing PN regimen prescribed by the
attending physician. Of the fifty-two PN regimen already in place,
more than half have been modified. Possible explanations of this
high percentage could be that, attending physicians may not have a
deep knowledge of the composition of different nutrition bags and
they end up not prescribing the one that better fulfill patients' re-
quirements or because a consultation has been required for more
complex patients where basic knowledge might not be enough.
Despite the intervention of the NST, the estimated requirements
have not been reached in all patients (see next paragraph “Patients
and nutritional outcomes”) and one of the explanations is the route
of administration: peripheral access rarely allows a patient to get all
the nutrients they require.
A second medical advice has been performed in 55% of patients.
Based on our experience, we think that this data might be under-
estimated. A possible explanation, is that we recorded only visits
following a referral involving the physician of the NST, with or
without a dietician consultation. More often, what is requested, as
the first follow-up visit, is a consultation solely with the Registered
Dietitian part of the NST if the patient is not in critical conditions. A
follow-up visit with a physician is requested as a third follow-up or
for complex patients. As for internal protocol, if the dietician no-
tices any condition requiring prescription modification while
seeing a patient in a certain ward, they will suggest a follow-up
made by the NST physician. Therefore, what happens in reality is
that, patients are followed more closely than what appears from
our data. Further advices shown here are hence related to major
therapy modification such as bag type or administered volume
changes and do not represent in full the attentions that NST
members give to their patients.
The median duration of PN therapy in our cohort is thirteen
days. An interesting data is that in thirty-three subjects PN lasted
for more than fifteen days. Among them, almost 40% of subjects
received PN via a peripheral catheter despite the ASPEN guide-
lines, that affirm that, for a duration greater than two weeks, it is
necessary to use a central route of administration. Ten out of
these thirteen patients whose PN lasted for more than 15 days via
PVC, underwent a second medical consult; three of them had it at
14, 21 and 22 days after the PN start but no change of catheter
type had been performed. We can state that more attention
should be paid to the relation between vein access and duration
of PN therapy.
When considering patients with peripheral access, re-
quirements are difficult to reach with bags for peripheral use.
Because we had only two types of peripheral bags, we expected a
higher prevalence of Nutriperilipid, which supplies more calories
and proteins. Data shows instead the prevalence as being quite
similar between the two bags (19.4% and 22.5% of total bags pre-
scribed), resulting in lack of nutrients compared to estimated
needs. Instead, when considering bags for central administration,
Smofkabiven 1477 was the most prescribed, representing 37%
(more than one third) in patients with a central catheter.
4.2. Patients and nutritional outcomes
The prevalence of patients who underwent surgery is quite high,
and this is certainly due to the fact that two large surgery wards
were included in the study. In most patients, the NST performed
medical consult after surgery in order to correct malnutrition
resulting from the surgery itself or the baseline health condition. In
a smaller percentage (16.3%) medical advice has been done pre-
operatively. In some cases (but we could not extrapolate this data)
surgery was not planned in advance but was performed as an
emergency life-saving treatment.
e6 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8
A state of malnutrition is a complex condition that involves
different apparatus, and is reflected in altered blood tests, even
though none of them are specific. We decided to use total blood
proteins as a surrogate marker of malnutrition since it is one of the
parameters used to evaluate nutritional status (together with
transferrin, albumin and lymphocytes for example). It is also the
parameter mostly used by the NST when one must decide on which
therapeutic approach to follow. For the reason that, beyond the
clinical visit, it is the most frequently available data, being dosed in
90.5% of patients (albumin only in 37.9% and transferrin in 18.9% of
patients, prealbumin measurement not available in the Hospital).
Thus, considering total blood proteins as a surrogate index of
malnutrition, we observed a quite high prevalence before PN
therapy (88% of subjects) of malnourished subjects. Of those pa-
tients, only 23.8% improved their nutritional status, reaching a
value of total blood proteins that was in range (the improvement
was statistically significant). This can be partially explained by the
following observations.
Comparing required and administered calories and proteins
we found a statistically significant difference, meaning that nu-
trients provided were less than those needed by the patients. The
median lack of 3.1 calories/kg/day, multiplied for the median
reference weight and for the median LOS resulted in a median
lack of 5158 kcal per hospitalization (equal to three full days of
starvation) while for the protein, we found a lack of 383 g per
hospitalization (the amount that would cover the protein re-
quirements of almost five days). This gap between requirements
and administration could be due to the limited amount of nu-
trients that can be administered with the peripheral veins when
this route was used, as confirmed by median amounts of PVC and
CVC that we observed in our cohort (the former being signifi-
cantly lower). The acknowledgement that almost half venous
accesses are peripheral catheters could sustain these findings.
The prospective interventional study from Soguel et al. [12]
showed that the additional presence of an intensive care unit
dietitian in an ICU ward helped in achieving overall better early
energy balance (from 5870 kcal to 3950 kcal (p < 0.001), thus
confirming that personnel with a strong background in nutrition
could have a positive impact on provision of the right amounts of
nutrients.
The wide difference in bag adjustments could be due to the
fact that attending physicians prefer to administer electrolytes
separately, diluted into saline rather than into nutritional bags.
This may be explained by the lack of experience in managing
nutrition bags or to the fact that they prefer to adjust electrolytes
dosage day by day without modifying the nutrition bag
prescription.
4.3. Blood test outcomes
We observed a statistically significant improvement in total
blood proteins after PN therapy, however we did not observe an
increase in other proteins such as transferrin and albumin. With a
mean duration of PN therapy of 13 days we expected to see an
improvement also in other parameters. For data shown in
Supplemental Tables 3e6, only a few of them reached a statistical
significance. This might be due to the lack of power of the
analysis caused by an inadequate number of subjects. Since dif-
ferences were quite small, we probably needed a higher number
of subjects in order to reach a significant result. However, when
considering transferrin and CRP, we expected to see an
improvement in the group of subjects with CVC and with longer
PN duration rather than in PVC and shorter PN duration groups.
These are unexpected results that need further investigation.
Only a few patients (seven before and four after PN) were tested
Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
for phosphorus and this is why data on this element are not
available in Supplemental Tables 3e6. These practices are not
following the ESPEN guidelines, that states, among others, that
magnesium and phosphate should be monitored weekly or twice
a week in PN patients [13,14]. Overall, our findings are aligned
with those from a study performed by Chong et al. at Hospital
Sungai Buloh, a large public hospital in Kuala Lumpur, Malaysia
[15]. They evaluated the effectiveness of the NST in optimizing PN
outcomes. Results showed that, despite a significant reduction in
metabolic abnormalities and an improvement in adherence to
biochemical monitoring guidelines, the NST did not improve
patient mortality rates and LOS. However, the NST lead to im-
provements in treatment outcomes, including reduced incidences
of metabolic complications, especially electrolyte imbalances,
which are strongly associated with improvements in adherence
to biochemical monitoring guidelines. These are positive results
that, even if not linked to major outcomes like length of stay and
mortality, are still a key to good patient management. In regards
to the partial compliance with the PN guidelines, this is an issue
common to several hospitals. Regular updates are recommended
for NST members, as described by Shiroma et al. [16]. In their
paper they performed a longitudinal study comparing ASPEN
guidelines with NST practices followed in a specific hospital. They
concluded there was a moderate level of PN inadequacy in in-
dications, administration, and monitoring. We believe our find-
ings could confirm this recommendation, in particular in regards
to monitoring and route of venous access.
4.4. Consumption of PN bags in different wards
Geriatrics and surgery wards consumed comparable quantities
of nutrition bags. Despite this, the NST prescribed almost half of
bags in the general surgery ward and only one fifth of the bags in
the geriatric ward. An explanation for this could be that physi-
cians working in the geriatrics ward feel more comfortable in
managing PN therapy. Surgeons, instead, beyond the prescription
of PN bags for established protocols, prefer to ask for a medical
consult from a specialist in nutrition when PN is needed. When
considering medical advice made by NST, instead of simply the
number of bags, the above-mentioned trend is confirmed. The
availability of a NST seems to influence the number of bags pre-
scribed in some wards, but not in others. A study comparing use
and complications of hospital PN before and after the imple-
mentation of an NST has been performed by Hvas et al. [17]
showed that introduction of a NST increased the total PN use
and safety of PN catheters. In particular they observed that this
increase was restricted to teams that infrequently referred for PN,
and enteral nutrition could replace PN. To better understand and
explain the above-mentioned data that we collected, we could
administer to attending physicians a questionnaire about PN
therapy knowledge and their confidence in managing it. We
could also use results of such questionnaires in order to build a
nutrition-focused training that could be performed in different
wards with the aim of helping improve attending physician
knowledge on parenteral nutrition and empowering their pre-
scription adequacy. The importance of a training program is
supported by a study from Boitano et al. [18]. They implemented
process improvement strategies that lead to quality improvement
and cost saving. Clinicians with advanced certifications in nutri-
tion support were pivotal to the success of the project. A great
knowledge in nutrition is crucial for PN prescription. The
Parenteral Nutrition Safety Task Force of the American Society for
Parenteral and Enteral Nutrition (A.S.P.E.N.) states that specific
competencies for prescribers from all disciplines are a require-
ment in order to ensure safe prescribing [19]. In 2013, DeLegge
 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8
and Kelly [20] discussed about the benefits of NST, and of training
and implementation strategies describing how a number of in-
stitutions introduced strong nutrition training, demonstrating
both clinical and economic benefit. The cost effectiveness of NSTs,
this time dealing with Home PN, is also discussed by Pietka et al.
[21]. Several guidelines describe how these multidisciplinary
groups should work in order to provide real improvement of this
complex therapy. Recommendations exist on how to set up
teams, and Larsen et al. outlines the NST method of practice
giving the interesting example of an Hospital in New Zealand
[22]. In our data shown above, we underlined the high number of
prescription changes made by the NTS to pre-existing TPN regi-
mens. Even if in our analysis we did not focus the attention to the
comparison of effectiveness between NTS and attending physi-
cians, we observed that more than 50% of prescriptions made by
attending physicians were not adequate. We want to underline
here the valuable role of NST. In support of the need of a
specialized team we can mention one last paper by Jensen et al.
[23] where a multidisciplinary team approach is recommended to
promote improved communication and quality of care.
5. Conclusions
This is an overview of the PN prescription at the San Giovanni
Bosco Hospital in Turin. It is necessary to use the correct route of
administration based on the requirements of each single patient
and more attention has to be paid to the duration of PN therapy.
More attention to blood tests is also needed, especially for elec-
trolytes, such as, phosphorus and magnesium, and blood pro-
teins, for instance, transferrin. Proper equipment should be
available in each ward since measuring body weight is as
important as getting blood pressure or body temperature or other
parameters.
Nutrition-focused training performed in different wards could
help in improving attending physician knowledge on parenteral
nutrition.
Improvements can be made in order to maximize efficacy and
effectiveness of PN therapy. Cooperation between nutrition spe-
cialists, Registered Dietitians, attending physicians and nurses is
essentials in order to reach this goal.
Disclosure of conflict
All authors state that they have no financial or personal conflict
of interest.
Sources of support
This work was supported by a scholarship awarded by Baxter
S.p.A. to L Rabbione and ML Ponta.
Acknowledgements
We would like to thank Dr. Pezzana and Dr. Zanardi for the
conception and design of the study, interpretation of data and for
the approval of the final version of the manuscript. Dr Ponta and Dr
Rabbione for generation and collection of data, and in addition Dr.
Ponta for assembly and interpretation of data, contribution to
conception and design of the study, and for drafting and revising
the manuscript; all the Registered Dietitians of the NST for gener-
ation and collection of data; Dr. Geninatti for statistical data
Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072
e7
analysis. A special thank to Dr. Eleanor Ann Peterson for reviewing
the manuscript.
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.clnesp.2018.01.072.
References
[1] Chima CS, Barco K, Dewitt ML, Maeda M, Teran JC, Mullen KD. Relationship of
nutritional status to length of stay, hospital costs, and discharge status of
patients hospitalized in the medicine service. J Am Diet Assoc 1997 Sep;97(9):
975e8. https://doi.org/10.1016/S0002-8223(97)00235-6.
[2] Cereda E, Lucchin L, Pedrolli C, D'Amicis A, Gentile MG, Battistini NC, et al.
Nutritional care routines in Italy: results from the PIMAI (Project: Iatrogenic
MAlnutrition in Italy) study. Eur J Clin Nutr 2010 Aug;64(8):894e8. https://
doi.org/10.1038/ejcn.2010.85.
[3] Lucchin L, Kob M, Lando L. La malnutrizione in Europa e in Italia. ADI Mag
2006;4:544e6.
[4] Somanchi M, Tao X, Mullin GE. The facilitated early enteral and dietary
management effectiveness trial in hospitalized patients with malnutrition.
J Parenter Enteral Nutr March 2011;35(2):209e16. https://doi.org/10.1177/
0148607110392234.
[5] Braunschweig C, Gomez S, Sheean PM. Impact of declines in nutritional status
on outcomes in adult patients hospitalized for more than 7 days. J Am Diet
Assoc Nov. 2000;100(11):1316e22. https://doi.org/10.1016/S0002-8223(00)
00373-4.
[6] Corkins MR, Guenter P, DiMaria-Ghalili RA, Jensen GL, Malone A, Miller S,
et al., American Society for Parenteral and Enteral Nutrition. Malnutrition
diagnoses in hospitalized patients: United States, 2010. J Parenter Enteral Nutr
February 2014;38:186e95. https://doi.org/10.1177/0148607113512154. first
published on November 18, 2013.
[7] Allard JP, Keller H, Jeejeebhoy KN, Laporte M, Duerksen DR, Gramlich L,
et al. Malnutrition at hospital admission-contributors and effect on length
of stay: a prospective cohort study from the Canadian malnutrition task
force. J Parenter Enteral Nutr 2016 May;40(4):487e97. https://doi.org/
10.1177/0148607114567902. Epub 2015 Jan 26.
[8] Gallagher-Allred CR, Voss AC, Finn SC, McCamish MA. Malnutrition and clin-
ical outcomes: the case for medical nutrition therapy. J Am Diet Assoc 1996
Apr;96(4):361e6. 369. PMID: 8598437.
[9] A.S.P.E.N. Board of Directors and the clinical guidelines task force. Guidelines
for the use of parenteral and enteral nutrition in adult and pediatric patients.
J Parenter Enteral Nutr 2002;26(1 Suppl.):1SAe138SA.
[10] Berlana D, Sabin P, Gimeno-Ballester V, Romero-Jime nez R, Zapata-Rojas A,
Marquez E, et al. Cost analysis of adult parenteral nutrition systems: three-
compartment bag versus customized. Nutr Hosp 2013 Nov 1;28(6):
2135e41. https://doi.org/10.3305/nutr.hosp.v28in06.6862.
[11] Sarria Cabrera MA, Gomes Dellaroza MS, Trelha CS, Cecilio CH, Souza SE. One-
year follow-up of non-institutionalized dependent older adults: mortality,
hospitalization, and mobility. Can J Aging 2012 Sep;31(3):357e61. https://
doi.org/10.1017/S0714980812000244.
[12] Soguel L, Revelly JP, Schaller MD, Longchamp C, Berger MM. Energy deficit
and length of hospital stay can be reduced by a two-step quality improve-
ment of nutrition therapy: the intensive care unit dietitian can make the
difference. Crit Care Med 2012 Feb;40(2):412e9. https://doi.org/10.1097/
CCM.0b013e31822f0ad7.
[13] ESPEN Guidelines. Guidelines and position papers from ESPEN. http://www.
espen.org/education/espen-guidelines.
[14] Staun M, Pironi L, Bozzetti F, Baxter J, Forbes A, Joly F, et al. ESPEN guidelines
on parenteral nutrition: home parenteral nutrition (HPN) in adult patients.
Clin Nutr 2009;28:359e479.
[15] Chong PF, Paraidathathu T. Effects of a nutrition support team on clinical
outcomes, metabolic complications and electrolyte abnormalities in patients
receiving parenteral nutrition. Asia Pac J Clin Nutr 2013;22(4):548e56.
https://doi.org/10.6133/apjcn.2013.22.4.15.
[16] Shiroma GM, Horie LM, Castro MG, Martins JR, Bittencourt AF, Logullo L, et al.
Nutrition quality control in the prescription and administration of parenteral
nutrition therapy for hospitalized patients. Nutr Clin Pract 2015 Jun;30(3):
406e13. https://doi.org/10.1177/0884533614567540. Epub 2015 Feb 13.
[17] Hvas CL, Farrer K, Donaldson E, Blackett B, Lloyd H, Forde C, et al. Quality and
safety impact on the provision of parenteral nutrition through introduction of
a Nutrition Support Team. Eur J Clin Nutr 2014 Dec;68(12):1294e9. https://
doi.org/10.1038/ejcn.2014.186. Epub 2014 Sep 24.
[18] Boitano M, Bojak S, McCloskey S, McCaul DS, McDonough M. Improving the
safety and effectiveness of parenteral nutrition: results of a quality
improvement collaboration. Nutr Clin Pract 2010 Dec;25(6):663e71. https://
doi.org/10.1177/0884533610385349.
e8 M.L. Ponta et al. / Clinical Nutrition ESPEN xxx (2018) e1ee8

[19] Guenter P, Boullata JI, Ayers P, Gervasio J, Malone A, Raymond E, et al.,
Parenteral Nutrition Safety Task Force, American Society for Parenteral and
Enteral Nutrition (A.S.P.E.N.). Standardized competencies for parenteral
nutrition prescribing: the American Society for Parenteral and Enteral
Nutrition Model. Nutr Clin Pract 2015 Aug;30(4):570e6. https://doi.org/
10.1177/0884533615591167. Epub 2015 Jun 15.
[20] DeLegge MH, Kelly AT. State of nutrition support teams. Nutr Clin Pract 2013
Dec;28(6):691e7. https://doi.org/10.1177/0884533613507455. Epub 2013 Oct 29.
[21] Pietka M, Watrobska-Swietlikowska D, Szczepanek K, Szybinski P, Sznitowska M,
Kłe˛ k S. Nutritional support teams: the cooperation among physicians and
pharmacists helps improve cost-effectiveness of home parenteral nutrition
(HPN). Nutr Hosp 2014 Sep 12;31(1):251e9. https://doi.org/10.3305/nh.2015.31.
1.8012.
[22] Larsen S, Baskett R, Hill S, Walmsley RS. Support and practical advice for the
use of nutrition support teams. Br J Nurs 2014 Jun 26eJul 9;23(12):636e40.
https://doi.org/10.12968/bjon.2014.23.12.636.
[23] Jensen GL, Compher C, Sullivan DH, Mullin GE. Recognizing malnutrition in
adults: definitions and characteristics, screening, assessment, and team
approach. J Parenter Enteral Nutr 2013 Nov;37(6):802e7. https://doi.org/
10.1177/0148607113492338. Epub 2013 Aug 22.

Please cite this article in press as: Ponta ML, et al., Assessing the appropriateness of parenteral nutrition use in hospitalized patients. A
comparison on parenteral nutrition bag prescription in different wards and nutritional outcomes, Clinical Nutrition ESPEN (2018), https://
doi.org/10.1016/j.clnesp.2018.01.072