Interactive CardioVascular and Thoracic Surgery 24 (2017) 931–937 ORIGINAL ARTICLE

doi:10.1093/icvts/ivx002 Advance Access publication 8 March 2017

Cite this article as: Agostini P, Lugg ST, Adams K, Vartsaba N, Kalkat MS, Rajesh PB. Postoperative pulmonary complications and rehabilitation requirements following
lobectomy: a propensity score matched study of patients undergoing video-assisted thoracoscopic surgery versus thoracotomy. Interact CardioVasc Thorac Surg
2017;24:931–7.

Postoperative pulmonary complications and rehabilitation

requirements following lobectomy: a propensity score matched
study of patients undergoing video-assisted thoracoscopic surgery
versus thoracotomy†

Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019

Paula Agostinia,b,*, Sebastian T. Luggc, Kerry Adamsa, Nelia Vartsabab, Maninder S. Kalkata, Pala B. Rajesha,
Richard S. Steyna, Babu Naidua,c, Alison Rushtonb and Ehab Bishaya
a
Department of Thoracic Surgery, Heart of England NHS Foundation Trust, Bordesley Green East, Birmingham, UK
b
School of Sport, Exercise and Rehabilitation Sciences, University of Birmingham, Birmingham, UK

THORACIC
c
Institute of Inflammation and Ageing, Centre for Translational Inflammation Research, University of Birmingham, UK

* Corresponding author. Department of Thoracic Surgery, Heart of England NHS Foundation Trust, Bordesley Green East, Birmingham, B9 5SS, UK. Tel: +44-121-4242000;
e-mail: paula.agostini@heartofengland.nhs.uk (P. Agostini).

Received 1 September 2016; received in revised form 6 December 2016; accepted 17 December 2016

Abstract
OBJECTIVES: Video-assisted thoracoscopic surgical (VATS) lobectomy is increasingly used for curative intent lung cancer surgery com-
pared to open thoracotomy due to its minimally invasive approach and associated benefits. However, the effects of the VATS approach on
postoperative pulmonary complications (PPC), rehabilitation and physiotherapy requirements are unclear; our study aimed to use propen-
sity score matching to investigate this.
METHODS: Between January 2012 and January 2016 all consecutive patients undergoing lobectomy via thoracotomy or VATS were prospect-
ively observed. Exclusion criteria included VATS converted to thoracotomy, re-do thoracotomy, sleeve/bilobectomy and tumour size >7 cm
diameter (T3/T4). All patients received physiotherapy assessment on postoperative day 1 (POD1), and subsequent treatment as deemed appro-
priate. PPC frequency was measured daily using the Melbourne Group Scale. Postoperative length of stay (LOS), high dependency unit (HDU)
LOS, intensive therapy unit (ITU) admission and in-hospital mortality were observed. Propensity score matching (PSM) was performed using
previous PPC risk factors (age, ASA score, body mass index, chronic obstructive pulmonary disease, current smoking) and lung cancer staging.
RESULTS: Over 4 years 736 patients underwent lobectomy with 524 remaining after exclusions; 252 (48%) thoracotomy and 272 (52%)
VATS cases. PSM produced 215 matched pairs. VATS approach was associated with less PPC (7.4% vs 18.6%; P < 0.001), shorter median
LOS (4 days vs 6; P < 0.001), and a shorter median HDU LOS (1 day vs 2; P = 0.002). Patients undergoing VATS required less physiotherapy
contacts (3 vs 6; P < 0.001) and reduced therapy time (80 min vs 140; P < 0.001). More patients mobilized on POD1 (84% vs 81%; P = 0.018),
and significantly less physiotherapy to treat sputum retention and lung expansion was required (P < 0.05).
CONCLUSIONS: This study demonstrates that patients undergoing VATS lobectomy developed less PPC and had improved associated
outcomes compared to thoracotomy. Patients were more mobile earlier, and required half the physiotherapy resources having fewer pul-
monary and mobility issues.
Keywords: VATS • Thoracotomy • Pneumonia • Atelectasis • Rehabilitation • Physiotherapy

INTRODUCTION (VATS) has been increasingly performed as an alternative to re-

Lung cancer remains the most common cause of cancer death
within the UK [1], with potentially curative surgery generally ac-
cepted as the most effective treatment. Since first described in
1992 [2], lobectomy via video-assisted thorascopic surgery
†Presented at the 30th Annual Meeting of the European Association for
Cardio-Thoracic Surgery, Barcelona, Spain, 1-5 October 2016.
section by thoracotomy due to the minimally invasive nature of
the procedure and a perceived reduction in pain, morbidity and
length of stay (LOS) [3, 4]. For these reasons VATS lobectomy has
been recommended as the approach for early-stage non-small
cell lung cancer (NSCLC) patients [5].
Lung surgery is associated with a high incidence (13%) of post-
operative pulmonary complication (PPC), such as pneumonia
and clinically significant atelectasis, which has a profound effect

C The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
V
932 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery
on both short and long-term postoperative morbidity and mor-
tality [6]. Current evidence is conflicting as to whether a VATS ap-
proach to lobectomy reduces the incidence of PPC compared to
thoracotomy [2, 7, 8]. This is likely a combination of the variabil-
ity of definition of PPC between studies and the study method-
ology; none of which have used a prospective, standardized and
objective method to define a PPC. Furthermore, many of the
studies published precede the growth of the VATS lobectomy
procedure in current practice. With the purpose of preventing
and ameliorating PPCs, early physiotherapy and mobilization fol-
lowing thoracotomy and lung resection are recommended.
This study aimed to use a propensity score matched cohort
of patients undergoing either VATS lobectomy or thoracotomy
lobectomy procedures to determine the effect of the VATS ap-
proach on short-term outcomes, in particular the incidence of
PPC and the rehabilitation requirements. Propensity score
matching is considered to strengthen outcome measures mainly
by enabling investigators to assemble a study cohort in which
patients are balanced in all observed relevant covariates.
This makes the assessment of the intervention more accurate
by minimizing the potential bias between the comparative
groups [9].
MATERIALS AND METHODS
Patient selection
This prospective observational study was conducted at a single-
centre large regional thoracic surgery unit serving 6 million people.
Consecutive cancer patients who underwent open thoracotomy or
VATS for lobectomy between January 2012 and January 2016 were
included. Decisions regarding patient operability and resectability
were informed by the British Thoracic Society guidelines for lung
cancer resection [10]. Patients were admitted to hospital on the day
of surgery. All operations were performed with single lung ventilation
under general anaesthesia, and patients were subsequently scheduled
for extubation in the operating room. VATS was defined as the use
of a utility incision, without rib-spreading, two further port incisions
and use of a thoracoscope to visualize the anatomical hilar dissection
(as defined by Swanson et al. [11]). Thoracotomy incisions were post-
erolateral, entering the hemithorax usually between the fifth and
sixth ribs; the Serratus Anterior muscle was usually spared.
All five surgeons at our centre performed both thoracotomy
and VATS procedures. Two surgeons preferred to use the thor-
ocotomy approach to lobectomy in most of their patients, and
three performed VATS lobectomy based on guidance from
centres (national and international) with extensive experience.
Guidance included tumour size (<7 cm), avoidance of N1 in-
volvement where known preoperatively on PET scan, no neo-
adjuvant chemo/radio therapy, no visibility of the tumour at
bronchoscopy and no crossing of fissures by tumours. VATS
has since evolved as our centre’s experience has grown to in-
clude cases with preoperatively identified N1 involvement, pre-
vious neoadjuvant chemo/radio therapy, visibility of the
tumour at bronchoscopy requiring hand sewn bronchial stump
closure and tumours crossing fissures necessitating bi-
lobectomy. Patients were excluded if they underwent VATS
which was then converted to thoracotomy, re-do thoracotomy,
completion lobectomy, sleeve lobectomy, bilobectomy, or had
tumour status of T3/T4 defined by tumours larger than 7 cm in
diameter.
Postoperative care
Postoperatively patients were managed in a dedicated thoracic
high-dependency unit (HDU) (level 2) and subsequently the thor-
acic surgery ward, unless complications required admission to
the intensive therapy unit (ITU). Postoperative pain control was
achieved by continuous thoracic epidural analgesia, paraverte-
bral infusion, intrathecal morphine and/or intercostal blocks or
systemic opioids (parenteral administration or intravenous
patient-controlled administration). The choice of analgesic tech-
nique was made by the anaesthetist after discussion with the pa-
tient. Standard postoperative care also included nursing staff
sitting patients out of bed postoperative day 1 (POD1), and from
Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019
this point patients starting early mobilization as able, with assist-
ance as necessary for surgical attachments and safety.
All patients were assessed by specialist thoracic surgical physio-
therapists on POD1 to determine presence of issues amenable to
physiotherapy intervention, such as sputum retention, atelectasis,
reduced exercise tolerance or issues with mobility/physical activity.
Thoracic expansion exercises or incentive spirometry (Coach 2V,
R
Medimark Europe, Grenoble, France) were taught where deemed
necessary for identified lung volume loss. Specific airway clearance
techniques were utilized to aid and assist sputum clearance where
necessary. If pulmonary complication developed in either thora-
cotomy or VATS patients physiotherapy input was escalated as ap-
propriate for sputum clearance and/or lung re-expansion. All
patients received physiotherapy until resolution of pulmonary
issues, and/or independence, usual mobility and exercise tolerance
were restored. Hospital discharge criteria, agreed with investigators
in advance, included patients who were medically fit and who had
been discharged from physiotherapy.
Data collection
Data collected included demographics and preoperative record of
smoking status, body mass index (BMI), percentage predicted FEV1,
American Society of Anaesthetist (ASA) score, subjective preopera-
tive activity level and chronic obstructive pulmonary disease
(COPD) diagnosis defined by the referring clinician. Postoperative
data included underlying pathology and lung cancer staging.
Hospital LOS was defined as the LOS in hospital after the date of
surgery. The HDU LOS was also recorded, as well as ITU admission
and in-hospital mortality. PPC was identified using the Melbourne
Group Scale (MGS), which is a standardized scoring system vali-
dated by our group to define the presence of PPCs, such as pneu-
monia or clinically significant atelectasis likely to adversely affect
the patient’s clinical course [12, 13]. PPC is defined in those patients
presenting with four or more of the following eight dichotomous
factors: chest X-ray (CXR) findings of atelectasis or consolidation;
raised white cell count (WCC) (>11.2 x 109/l); temperature >38  C;
signs of infection on sputum microbiology; purulent sputum differ-
ing from preoperative status; oxygen saturations <90% on room air;
physician diagnosis of pneumonia; and prolonged HDU stay or re-
admission to HDU or ITU for respiratory complications. The MGS
variables were assessed daily by specialist physiotherapists whilst
performing postoperative pulmonary assessments.
Ethical considerations
This study was conducted with the approval of the National
Research Ethics Service (NRES) Committee West Midlands. This
 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery
study was registered with the Birmingham Heartlands Hospital
audit department (audit code 1672).
Statistical analysis
Analysis was performed using SPSS (IBM Corp. Released 2013.
IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY,
USA). Normally distributed continuous variables are expressed as
mean (±SD), skewed continuous variables as median (interquar-
tile range), and categorical variables as actual number (percent-
ages). Normality of distributions was assessed using the
Kolmogorov-Smirnov test. Differences in baseline characteristics
and postoperative outcomes were analysed using Chi-square
tests for categorical variables, Fisher’s exact test for categorical
variables where numbers per cell were 5 or less, independent
samples t-test for continuous variables and Mann-Whitney
U tests for continuous variables with skewed distributions.
A P-value <0.05 was considered significant.
Sample size
Prior to this study audit data at the research centre (2007–2012)
showed a frequency of 20.8% in PPC following thoracotomy lob-
ectomy (n = 582) using the MGS, and 9.4% following VATS lobec-
tomy (n = 64). Sample size calculation based on a difference of
10% PPC between thoracotomy and VATS groups (21% vs 11%
respectively), and based on a type I error of 5% and a type II
error of 20% (power of 80%) indicated that matched groups of
208 patients were required to assess the effect of the surgical ap-
proach on outcome. In order to ensure groups of this size a
4 year study was undertaken; over this period 544 thoracotomy
lobectomies were predicted and 260 VATS lobectomy. Following
application of exclusion criteria a loss of 38% to the thoracotomy
lobectomy group was expected (n = 377) and an unknown loss to
the VATS group. Also, further to this the propensity score match-
ing process would account for additional loss to the groups of
unknown quantity, hence a 4-year study period was necessary.
Propensity score matching
Propensity score matching (PSM) was performed; participants
defined by either surgical approach were matched in pairs based
on similar estimated propensity scores, in order to reduce the ef-
fect of confounding factors. Propensity scores were estimated
using a logistic regression model which included variables that
we had previously identified as factors independently associated
with the development of PPC following thoracotomy and lung
resection in our centre. As such they were considered the most
likely possible confounders, and included age, ASA score, BMI,
COPD, smoking status [12]. Lung cancer stage was also included
in the matching process as a potential confounder for treatment
assignment. Percentage predicted FEV1 was not included as our
previous published studies have shown this not to be an inde-
pendent risk factor for PPC development [6, 12].
Estimation of the propensity score specifying nearest-
neighbour 1:1 matching, and matching without replacement was
performed. This method finds matches based on a greedy match-
ing algorithm, sorting the observations in the thoracotomy cohort
by their estimated propensity score and matching each case to
one in the VATS cohort with the closest propensity score (nearest
933
neighbour). To ensure good matches, a calliper (maximum allow-
able difference between two participants) was imposed of 0.15 of
the standard deviation of the logit of the propensity score.
To determine the success of the matching process global imbal-
ance measures were performed using multivariate tests [14], the L1
measure, bounded by 0 for perfect balance and 1 for total separ-
ation, and significance tests [15] using an overall imbalance Chi-
square test. Balance of the individual covariates was tested using the
standardized mean differences and the variance ratio of the thora-
cotomy and VATS groups before and after matching. Univariate tests
provide a standardized mean difference, reported for the pre- and
post-matching samples. PSM analysis was performed using IBM
Statistics SPSS Version 22, having downloaded R, R essential plug-ins
Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019
and PSE to enable the SPSS propensity score matching tool.
RESULTS
Over the 4-year period a total of 736 cancer patients underwent
THORACIC
lobectomy. Two hundred and twelve cases were excluded from
the study. Reasons for exclusion (Fig. 1) included VATS converted
to thoracotomy in 89 patients (42%), bilobectomy in 35 patients
(16.5%), sleeve lobectomy in 27 patients (12.7%), re-do thoracot-
omy or completion lobectomy in 17 patients (8%) and tumour
size >7 cm in diameter in 29 patients (13.7%). A further 2 patients
(0.9%) were excluded as postoperative bleeding necessitated re-
turn to theatre for thoracotomy on POD 1. In 13 patients (6.1%)
data relating to covariates to be used for PSM were missing.
Propensity score matching
Of the 524 patients included, 252 patients (48.1%) underwent open
thoracotomy and 272 patients (51.9%) underwent VATS. The base-
line demographics of the prescore matched patients are shown in
Table 1. PSM was performed to reduce confounding for the follow-
ing covariates: age, ASA score >_3, BMI, COPD, current smoking, and
lung cancer staging/metastatic disease. The pair matching process
produced 215 matched pairs, characteristics of each group are
compared in Table 1. Balance of covariates was achieved as demon-
strated by a Chi-square statistic of P = 0.897, and a L1 value of 0.746
before matching and 0.729 after. The success of matching can be
seen in the dot-plot of individual propensity scores in Fig. 2 and the
standardized differences before and after matching in Fig. 3.
Outcomes
Compared to open thoracotomy, a VATS approach resulted in a
significant reduction in PPC development, hospital LOS and HDU
LOS (Table 2). There were no significant differences in ITU admis-
sion or hospital mortality.
Early mobility/physiotherapy
Two hundred and two (94%) patients who underwent thoracot-
omy were able to sit out of bed on POD1 and 206 (95.8%) follow-
ing VATS. Significantly more patients undergoing VATS
lobectomy were more mobile earlier (Table 3). All patients
undergoing thoracotomy received physiotherapy for pulmonary
or mobility issues, or to increase physical activity beyond that
associated with early mobilization. However, following VATS 54
934 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery

Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019

Figure 1: Reasons for exclusion of patients from thoracotomy and VATS groups.

Table 1: Demographics and risk factors of patients before propensity score matching, and thoractomy versus VATS groups after
matching

Baseline variables Prescore matched (n = 524) Matched thoracotomy (n = 215) Matched VATS (n = 215) P-value*

Gender (male) 263 (50.2%) 118 (54.9%) 96 (44.7%) 0.043

Age (years) median, IQR 69 (13) 69 (12) 69 (13) 0.920
BMI median, IQR 26 (7) 26.3 (6.8) 26 (5.9) 0.476
ASA score >_3 272 (51.9%) 112 (52.1%) 112 (52.1%) 1.000
Current smoker 105 (20%) 39 (18.1%) 35 (16.3%) 0.702
COPD 148 (28.2%) 57 (26.5%) 57 (26.5%) 1.000
FEV1%, mean (±SD) 88% (±20) 88.11% (±21.4) 87.08% (±19.73) 0.605
ppoFEV1, mean (±SD) 69.94 (±17.1) 70.26% (±18.33) 69.31% (±16.23) 0.569
Preop mobility <400 m 141 (26.9%) 58 (27%) 61 (28.4%) 0.829
NSCLC staging (%)
IA 175 (33.4%) 62 (28.8%) 72 (33.5%) 0.754
IB 151 (28.8%) 67 (31.2%) 71 (33%)
IIA 91 (17.4%) 46 (21.4%) 38 (17.7%)
IIB 18 (3.4%) 8 (3.7%) 5 (2.3%)
IIIA 50 (9.5%) 15 (7%) 12 (5.6%)
Secondary metastatic disease 39 (7.4%) 17 (7.9%) 17 (7.9%) 1.000

VATS: video-assisted thoracoscopic surgery; IQR: interquartile range; BMI: body mass index; ASA: American Society of Anaesthesiologist Score; COPD: chronic
obstructive pulmonary disease.
*P-value, difference between matched thoracotomy and VATS groups.

(25%) patients did not need specific physiotherapy either for first
time mobilization or to further promote early mobility; when first
mobilizing they did so with either a member of nursing staff or in-
dependently. In total 149 (69%) patients undergoing thoracotomy
required assistance of 2 staff to mobilize on POD1 compared to
103 (48%) following VATS. Overall VATS patients needed signifi-
cantly fewer physiotherapy sessions, less physiotherapists to provide
treatment and reduced median therapy time per patient (Table 3).
Also fewer specific chest physiotherapy interventions to treat spu-
tum retention and lung expansion were required (Table 3).
 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery
Figure 2: Dot-plot of individual propensity scores of units in the thoracotomy
and VATS groups (treated = thoracotomy, control = VATS).
Figure 3: Standardized differences before and after matching. BMI: body mass
index; ASA: American Score of Anaethesiology; COPD: chronic obstructive pul-
monary disease.
935
Table 2: Outcomes following open thoracotomy and VATS
Outcome Thoracotomy VATS P-value
(n = 215) (n = 215)
PPC 40 (18.6%) 16 (7.4%) <0.001
Median Hospital LOS (IQR) 6 (4) 4 (3) <0.001
Median HDU LOS (IQR) 2 (2) 1 (1) 0.002
ITU admission 9 (4.2%) 6 (2.8%) 0.599
Hospital mortality 5 (2.3%) 3 (1.4%) 0.724
VATS: video-assisted thoracoscopic surgery; PPC: postoperative pul-
monary complication; LOS: length of stay; HDU: high dependency unit;
ITU: intensive therapy unit; IQR: interquartile range.
Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019
DISCUSSION
THORACIC
We observed significantly less PPC following VATS lobectomy
compared to open thoracotomy; in our PSM groups the PPC in-
cidence following VATS was less than half that of open thoracot-
omy. Furthermore, patients undergoing VATS lobectomy had a
significant reduction in both hospital and HDU LOS, which are
both clinically and economically important outcomes. We have
also demonstrated that patients undergoing the VATS approach
required less postoperative physiotherapy and were capable of
engaging in mobilizing greater distances following surgery on
POD1.
The frequency of PPCs following thoracotomy at this regional
thoracic surgery unit is concurrent with other studies [16, 17].
Reporting of PPC following thoracic surgery varies predominantly
because there is no standard; it is dependent on the type of com-
plications included, the definition of pulmonary complications,
and the type of surgery. Use of the objective MGS to define a
PPC does not include rare but serious postoperative complica-
tions such as broncho-pleural fistulas and pulmonary embolism.
Despite this, the more common and potentially less severe PPCs
detected by the MGS have been shown to be associated with
increased hospital morbidity, increased 30-day re-admission fre-
quency, increased mortality at 30 days and 90 days and poorer
long-term overall survival [6, 12].
Our study showed that a VATS approach reduced incidence of
PPC compared to open thoracotomy, which is supported by
other studies [8]. A previously published meta-analysis which
included 4 PSM studies (n = 3634) found similarly that patients
undergoing VATS were significantly less likely to suffer pneumo-
nia (3.2 vs 5.0%; RR, 0.65; 95% CI 0.47–0.89; P = 0.008) [18].
A more recent large propensity matched study also showed sig-
nificant reduction in both minor and major complications within
30 days of surgery following VATS compared to thoracotomy,
including pneumonia (7.5% vs 10.6%; P = 0.045) [19]. Other stud-
ies however, have found no significant difference in PPC between
the surgical approaches, which may be due to the inclusion of
small studies, and those predating the widespread use of VATS
approach [20]. A USA database of 1054 VATS and 2907 patients
undergoing thoracotomy revealed no difference in pulmonary
infection (OR, 0.99; 95% CI –0.81–1.20; P = 0.914) [7]. Another
USA nationwide sample with propensity matching of 10 173 pa-
tients in the VATS group and 30 866 in the thoracotomy group
showed no significant difference in incidence of pneumonia
(7.3% vs 8.2%; P = 0.170) and of pulmonary collapse (13.6% vs
936 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery
Table 3: Physiotherapy requirements following open thoracotomy and VATS
Physiotherapy requirements Thoracotomy (n = 215)
Median number of sessions (IQR) 6 (4)
Median number of therapist (IQR) 9 (7)
Time (min) median (IQR) 140 (120)
Emergency call outs 21 (9.8%)
Manual techniques 30 (14%)
Incentive spirometry 112 (52.1%)
POD1: Sat out 202 (94%)
POD1: Mobility
No walking 41 (19.1%)
Mobility <10 meters 17 (7.9%)
Mobility >_10 meters 157 (73%)
Suction 26 (12.1%)
Deep breathing exercises 127 (59.1%)
VATS: video-assisted thoracoscopic surgery; IQR: interquartile range; POD1: postoperative day 1.
13.4%; P = 0.863) [21]. However, this type of study, although large,
retrospectively examines an administrative database and is lim-
ited by its dependence on the quality of the data entered, espe-
cially for postoperative morbidity.
The observed reduction in PPC following VATS lobectomy is
likely to be due to multifactorial components, such as reduction in
surgical attachments, ventilatory impairment [22], dyspnoea [23],
decreased exercise tolerance and pain, all of which may reduce
ability to mobilize, cough and clear secretions. Our study is novel
in that it clearly demonstrates that those undergoing VATS surgery
have less physical limitations in their postoperative rehabilitation
period; these patients require less intensive and less frequent post-
operative physiotherapy and engaged in more POD1 mobility
compared to thoracotomy counterparts. Indeed patients undergo-
ing VATS have shorter hospital LOS, which is a finding that is now
widely accepted [3, 8, 20, 24]. The importance of effective manage-
ment of postoperative pain in early mobilization after thoracic sur-
gery has been identified [25]; strategies to improve pain relief are
highly relevant in developing an enhanced recovery pathway. Less
invasive surgery involves fewer attachments such as intercostal
drains and a urinary catheter, and analgesic attachments. Studies
have previously demonstrated that a VATS approach reduces dur-
ation of intercostal drainage on average by 1.5 days [11]. Another
study has demonstrated less need for rehabilitation following dis-
charge in octogenarians undergoing VATS compared to thoracot-
omy (5% vs 22.5%, P = 0.015) [24]. The reduction in morbidity
including PPC and the reduced need for physiotherapy also has
cost implications; indeed, studies have shown VATS to be more
cost-effective than thoracotomy in both the early postoperative
period and one year post-surgery [7, 26].
We demonstrated no difference in hospital mortality between
the propensity score matched VATS and thoracotomy cases in
our study. Indeed, the initial VATS feasibility study showed no
difference in perioperative mortality [18]. Our study has not
looked at long-term survival in VATS compared to thoracotomy,
although other studies have focused on this and shown that
VATS offers acceptable survival rates [4, 27, 28].
Study limitations
This is a real-life study involving consecutive patients undergoing
thoracic surgery in a tertiary centre; our patient demographics
VATS (n = 215) P-value
3 (2) <0.001
5 (4) <0.001
80 (65) <0.001
8 (3.7%) 0.021
8 (3.7%) <0.001
52 (24.2%) <0.001
206 (95.8%) 0.511
36 (16.7%) 0.018
4 (1.9%)
175 (81.4%)
Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019
13 (6%) 0.044
101 (47%) 0.016
and staging of cancer is comparable to national data [29]. We
have used PSM in order to minimize the potential bias between
patients undergoing a VATS or open thoracotomy approach.
However, it should be acknowledged that PSM has its own limi-
tations and does not adequately substitute for well-designed,
randomized controlled trials. Also, as propensity matching
required full data for all covariates used, any patient where a full
data set was not collected could not be included, which was the
case in only a very small number of patients (n = 13) in our study.
Secondly, there was inevitably some variability in the surgical
techniques and skills of surgeons. This could be considered ad-
vantageous as this is more representative of VATS procedures in
the ‘real world’ than the skills and management of an individual
surgeon. However, it was a limitation to the PSM model that two
of the surgeons had a preference for the thoracotomy approach,
which limited the addition of true confounders (factors associ-
ated with both outcome and treatment assignment) to the
model, as treatment assignment factors were not possible to truly
capture for all patients. Lung cancer stage was included in the
model as it would have influenced surgeon choice regarding sur-
gical approach to some extent. COPD and smoking were
included as potentially having an impact on outcome, but could
also have an influence on the decision to use a less invasive ap-
proach. The confounders/covariates associated with outcome
were chosen primarily due to evidence that these 5 factors were
independently associated with PPC following thoracotomy in a
previous study at our centre [12]. We acknowledge that there
were 10% more males in the thoracotomy group, however, this is
not a factor associated with either treatment assignment or has
ever been associated with the development of PPC in the rele-
vant literature. Finally, our study only measured hospital morbid-
ity and mortality and therefore did not measure long-term
outcomes such as hospital readmissions, whether the VATS ap-
proach reduces this is conflicting [30]. An interest of future study
would be to assess the long-term outcome of VATS and thora-
cotomy; and whether reduced rate of PPC has an impact on pa-
tient survival.
CONCLUSION
Our prospective propensity matched study clearly demonstrates
that patients undergoing VATS lobectomy developed less PPC
 P. Agostini et al. / Interactive CardioVascular and Thoracic Surgery
and had improved associated outcomes compared to patients
undergoing thoracotomy lobectomy. Patients were significantly
more mobile earlier, and required half the physiotherapy having
developed fewer pulmonary and mobility issues. Although the
frequency of PPC following VATS lobectomy was relatively small,
it is important for future studies to identify associated risk factors
in order to maximize postoperative patient outcome following
VATS.
Conflict of interest: none declared.
REFERENCES
[1] Office for National Statistics. Mortality Statistics: Deaths Registered in
England and Wales (Series DR). 2013. http://www.ons.gov.uk/ons/rel/
vsob1/mortality-statistics–deaths-registered-in-england-and-wales–ser
ies-dr-/2013/index.html (accessed 30 October 2016).
[2] Roviaro G, Rebuffat C, Varoli F, Vergani C, Mariani C, Maciocco M.
Videoendoscopic pulmonary lobectomy for cancer. Surg Laparosc
Endosc 1992;2:244–7.
[3] Paul S, Altorki NK, Sheng S, Lee PC, Harpole DH, Onaitis MW et al.
Thoracoscopic lobectomy is associated with lower morbidity than open
lobectomy: a propensity-matched analysis from the STS database.
J Thorac Cardiovasc Surg 2010;139:366–78.
[4] Sugi K, Kaneda Y, Esato K. Video-assisted thoracoscopic lobectomy
achieves a satisfactory long-term prognosis in patients with clinical stage
IA lung cancer. World J Surg 2000;24:27–30; discussion 30-1.
[5] Howington JA, Blum MG, Chang AC, Balekian AA, Murthy SC. Treatment
of stage I and II non-small cell lung cancer: diagnosis and management
of lung cancer, 3rd ed: American college of chest physicians evidence-
based clinical practice guidelines. Chest 2013;143(5 Suppl):e278S–313S.
[6] Lugg ST, Agostini PJ, Tikka T, Kerr A, Adams K, Bishay E et al. Long-term
impact of developing a postoperative pulmonary complication after lung
surgery. Thorax 2016;71:171–6.
[7] Swanson SJ, Meyers BF, Gunnarsson CL, Moore M, Howington JA,
Maddaus MA et al. Video-assisted thoracoscopic lobectomy is less costly
and morbid than open lobectomy: a retrospective multiinstitutional
database analysis. Ann Thorac Surg 2012;93:1027–32.
[8] Villamizar NR, Darrabie MD, Burfeind WR, Petersen RP, Onaitis MW,
Toloza E et al. Thoracoscopic lobectomy is associated with lower mor-
bidity compared with thoracotomy. J Thorac Cardiovasc Surg
2009;138:419–25.
[9] Blackstone EH. Comparing apples and oranges. J Thorac Cardiovasc Surg
2002;123:8–15.
[10] Lim E, Baldwin D, Beckles M, Duffy J, Entwisle J, Faivre-Finn C et al.
Guidelines on the radical management of patients with lung cancer.
Thorax 2010;65(Suppl 3):iii1–27.
[11] Swanson SJ, Herndon JE II, D’Amico TA, Demmy TL, McKenna RJ Jr,
Green MR et al. Video-assisted thoracic surgery lobectomy: report of
CALGB 39802–a prospective, multi-institution feasibility study. J Clin
Oncol 2007;25:4993–7.
[12] Agostini P, Cieslik H, Rathinam S, Bishay E, Kalkat MS, Rajesh PB et al.
Postoperative pulmonary complications following thoracic surgery: are
there any modifiable risk factors? Thorax 2010;65:815–8.
937
[13] Agostini P, Naidu B, Cieslik H, Rathinam S, Bishay E, Kalkat MS et al.
Comparison of recognition tools for postoperative pulmonary complica-
tions following thoracotomy. Physiotherapy 2011;97:278–83.
[14] Iacus SM, King G, Porro G. CEM: coarsened exact matching software.
J Stat Softw 2009;30:1–27.
[15] Hansen BB, Bowers J. Covariate balance in simple, stratified and clustered
comparative studies. Stat Sci 2008;23:219–36.
[16] Novoa N, Ballesteros E, Jiménez MF, Aranda JL, Varela G. Chest physio-
therapy revisited: evaluation of its influence on the pulmonary morbidity
after pulmonary resection. Eur J Cardiothorac Surg 2011;40:130–4.
[17] Schussler O, Alifano M, Dermine H, Strano S, Casetta A, Sepulveda S et al.
Postoperative pneumonia after major lung resection. Am J Respir Crit
Care Med 2006;173:1161–9.
[18] Cao C, Manganas C, Ang SC, Peeceeyen S, Yan TD. Video-assisted thor-
acic surgery versus open thoracotomy for non-small cell lung cancer: a
meta-analysis of propensity score-matched patients. Interact Cardiovasc
Downloaded from https://academic.oup.com/icvts/article-abstract/24/6/931/3064188 by guest on 26 April 2019
Thorac Surg 2013;16:244–9.
[19] Laursen LØ, Petersen RH, Hansen HJ, Jensen TK, Ravn J, Konge L. Video-
assisted thoracoscopic surgery lobectomy for lung cancer is associated
with a lower 30-day morbidity compared with lobectomy by thoracot-
omy. Eur J Cardiothorac Surg 2016;49:870–5.
[20] Whitson BA, Groth SS, Duval SJ, Swanson SJ, Maddaus MA. Surgery for
THORACIC
early-stage non-small cell lung cancer: a systematic review of the video-
assisted thoracoscopic surgery versus thoracotomy approaches to lobec-
tomy. Ann Thorac Surg 2008;86:2008–16; discussion 2016-8.
[21] Paul S, Sedrakyan A, Chiu YL, Nasar A, Port JL, Lee PC et al. Outcomes
after lobectomy using thoracoscopy vs thoracotomy: a comparative ef-
fectiveness analysis utilizing the Nationwide Inpatient Sample database.
Eur J Cardiothorac Surg 2013;43:813–7.
[22] Miyoshi S, Yoshimasu T, Hirai T, Hirai I, Maebeya S, Bessho T et al.
Exercise capacity of thoracotomy patients in the early postoperative
period. Chest 2000;118:384–90.
[23] Bolliger CT, Jordan P, Solèr M, Stulz P, Tamm M, Wyser C et al.
Pulmonary function and exercise capacity after lung resection. Eur Respir
J 1996;9:415–21.
[24] Port JL, Mirza FM, Lee PC, Paul S, Stiles BM, Altorki NK. Lobectomy in
octogenarians with non-small cell lung cancer: ramifications of increasing
life expectancy and the benefits of minimally invasive surgery. Ann
Thorac Surg 2011;92:1951–7.
[25] Das-Neves-Pereira JC, Bagan P, Coimbra-Israel AP, Grimaillof-Junior A,
Cesar-Lopez G, Milanez-de-Campos JR et al. Fast-track rehabilitation for
lung cancer lobectomy: a five-year experience. Eur J Cardiothorac Surg
2009;36:383–91; discussion 391-2.
[26] Watson TJ, Qiu J. The impact of thoracoscopic surgery on payment and
health care utilization after lung resection. Ann Thorac Surg
2016;101:1271–80.
[27] Yang CF, Meyerhoff RR, Mayne NR, Singhapricha T, Toomey CB,
Speicher PJ et al. Long-term survival following open versus thoracoscopic
lobectomy after preoperative chemotherapy for non-small cell lung can-
cer. Eur J Cardiothorac Surg 2015;49:1615–23.
[28] Nakamura H, Taniguchi Y, Miwa K, Adachi Y, Fujioka S, Haruki T et al.
Comparison of the surgical outcomes of thoracoscopic lobectomy, seg-
mentectomy, and wedge resection for clinical stage I non-small cell lung
cancer. Thorac Cardiovasc Surg 2011;59:137–41.
[29] Powell HA, Tata LJ, Baldwin DR, Stanley RA, Khakwani A, Hubbard RB.
Early mortality after surgical resection for lung cancer: an analysis of the
english national lung cancer audit. Thorax 2013;68:826–34.
[30] Rajaram R, Ju MH, Bilimoria KY, Ko CY, DeCamp MM. National evalu-
ation of hospital readmission after pulmonary resection. J Thorac
Cardiovasc Surg 2015;150:1508–14.e2.