Updates Surg
DOI 10.1007/s13304-015-0334-z
ORIGINAL ARTICLE
Conversion during thoracoscopic lobectomy: related factors
and learning curve impact
David E. Smith1 • Agustin Dietrich1 • Matias Nicolas1 • Alejandro Da Lozzo1
Enrique Beveraggi1
Received: 18 July 2015 / Accepted: 18 October 2015
Ó Italian Society of Surgery (SIC) 2015
Abstract Video-assisted thoracoscopic surgery (VATS)
lobectomy has become a standard procedure for lung
cancer treatment. Conversion-related factors and learning
curve impacts, were poorly described. The aim of this
study was to review the reasons and related factor for
conversion in VATS lobectomy and the impact on this of
the surgeon’s learning curve. From June 2009 to May 2014,
154 patients who underwent a VATS lobectomy were
included in our study. Patients’ characteristics, pathology
background, operative times, overall length of stay, overall
morbidity and type of major complications were recorded
for all patients and compared between non converted
(n = 133) and converted (n = 21) patients. To evaluate
surgeon’s learning curve, we analyzed rates and causes of
conversion in the first period (first 77 patients) and in the
last period (78–154 patients). Patients characteristics were
similar between converted and non-converted groups.
Patients who were converted to open thoracotomy pre-
sented more frecuently tumors [3 cms (P = 0.02). The
average of operative times and the length of stay were not
significantly different between groups. Overall morbidity
and major complications were also similar in both groups.
There were no impact of surgeon’s learning curve in
overall rate conversion in both groups. Emergency con-
version was always secondary to vascular accidents, all in
& Agustin Dietrich
agustin.dietrich@hospitalitaliano.org.ar
1
Hospital Italiano de Buenos Aires, Thoracic Surgery and
Lung Transplant, Perón 4190, 1181 Buenos Aires, Argentina
•
the first group (p = 0.059). Surgeons should be expecting
to perform a conversion to a thoracotomy in patients who
present in preoperative studies, tumors greater than 3 cms.
Learning curve only affected the emergency conversion,
occurred all in the first half of our series.
Keywords Vats lobectomy
Conversion
Learning
curve
Related factors
Thoracotomy
Introduction
Since the introduction of video-assisted thoracoscopic
surgery (VATS) lobectomy in the early 1990s [1–4], it has
become a standard procedure worldwide for lung cancer
treatment. This surgical technique minimize trauma and
improves patient recovery with shorter length of stay
without compromising surgical outcomes, like oncological
prognosis or surgical morbidity. However, most of pul-
monary lobectomies are still performed via a thoracotomy.
Perceived technical challenges comparing to an open
approach and surgeon’s learning curve probably explain
that only approximately 10 % of lobectomies are per-
formed thoracoscopically [5]. Moreover, almost 20 % of
VATS lobectomies are converted to open thoracotomy.
Causes of conversion are clearly described such as intra-
operative complications, technical problems or anatomical
problems. However, conversion-related factors are not
clearly described. In the same way, to the present, learning
curve association with intraoperatively complications and
conversion to open surgery was poorly described.
The aim of our study is to review the reasons and related
factors for conversion in VATS lobectomy and the impact
on this of the surgeon’s learning curve in a unique surgical
team.
123

Methods
From June 2009 to May 2014, clinical data of 285
consecutive patients who underwent pulmonary lobec-
tomy were collected retrospectively. There was a unique
surgical team which performed all procedures. Patients
who underwent a thoracotomy (n = 121) or a uniportal
VATS lobectomy (n = 10) were excluded. The remain-
ing 154 patients (54 %) were considered for further
analysis. This group was divided into patients who
underwent a Complete Vats Lobectomy (Group 1,
n = 133) and those who were converted to thoracotomy
(Group 2, n = 21). Patient demographics were similar
between groups (see Table 1).
All patients with lung tumours with a diameter minor of
5 cms, with preoperative knowledge about absence of
involvement of great vessels, chest wall or diaphragm
underwent a VATS lobectomy. In patients with history of
pleural pleurodesis, thoracic radiotherapy or diagnoses of
bronchiectasis or lung abscess, we performed a lung
lobectomy through a thoracotomy due to the adherences
related to these etiologies.
We analyzed histopathological characteristics in the two
groups, like etiology, limph node metastases and tumor
size (Table 1). Operative times, overall length of stay,
overall morbidity and type of major complications were
recorded for all patients and compared between both
groups. Postoperative morbidity was defined as any post-
operative adverse event, which occurred during the first 30
postoperative days. Severe complication was defined as a
complication grade CIIIb of Dindo Clavien classification.
Perioperative mortality was defined as death during the
resection hospitalization or within 90 days of discharge
after resection.
To analyze surgeon’s learning curve, we divided our
series into two periods: 18 period which includes the first
77 patients (1–77) and the 28 period which includes the
last 77 patients (78–154). Patient demographics were
similar between first and second periods. We analyzed
rates and causes of conversion in the two periods (see
Table 2).
Statistical analysis
Data were analyzed by STATA version 12 (StataCorp LP,
Texas, USA). Results were given as percentages, mean and
standard deviations or median and ranges. Quantitative and
qualitative variables were compared with student’s t test
and Fisher test, respectively. Odds ratios were calculated
using logistic regression. A p value \0.05 was considered
significant.
123
Updates Surg
Surgical procedures
Our surgical strategy in lung tumours includes two surgical
times. We performed in the first surgery a mediastinoscopy
to perform a lung cancer stadiation in all patients who
presented lung tumours with more of 2 cm size. After that,
our patients underwent the VATS lobectomy. We per-
formed a 5 cm incision in the anterior axillary line in the
fourth or fifth intercostals space, depending on the tumour
location without ribs spreading. We added two 10 mm
trocars which are placed in seventh to eighth intercostals
space. We performed an anterior approach opening the
mediastinal pleura to achieve an anatomical dissection of
the pulmonary hilum. After that we dissected the pul-
monary ligament and performed a complete limphadenec-
tomy of the mediastinal limph node groups: 7, 9, 10 and 11.
We made an individual stapling of the pulmonary vein,
artery, bronchus and fissure by using a 45-mm long linear
endostapler (Ethicon Endo-Surgery) and the resected lobe
is extracted with and Endo Catch (Autosuture, Covidien).
Finally we placed one or two chest tubes depending on
intraoperative surgeon decision.
The patients who underwent VATS conversion to open
lobectomy, the incisions were enlarged through a pos-
terolateral thoracotomy.
Results
From June 2009 to may 2014, 154 VATS lobectomies were
performed. One hundred and thirty-three patients (87 %,
Group 1) underwent a complete VATS lobectomy and 21
patients (13 %, Group 2) were converted to open surgery.
There were no differences in age and gender between
groups. We also analyzed patient’s comorbidities with
ASA scores and pulmonary reserve (FEV1, DLCO) which
were similar in both groups. There were also no differences
in prior treatments, chemo and radiotherapy and previous
surgeries (Table 1).
The pathological findings revealed that 87 % of the
patients in group 1 and all patients of group 2 presented
malignancy tumors (p = 0.08). Types of cancer are shown
in Table 1, with no differences between groups. In the
same way, rates of positive lymph nodes were similar in
both groups (3 and 10 % in group 1 and 2, respectively).
Regarding tumor size, 38 % of patients who were con-
verted to open thoracotomy and 20 % who underwent a
complete VATS lobectomy, presented tumors greater than
3 cm (P = 0.02).
The average of operative times (185 and 198 min) and
the length of stay were not significantly different between
Updates Surg

Table 1 Baseline characteristics of the study population and postoperative outcomes

Variable Vats (n = 133) Vats conversion (n = 21) P value

Age [65 years, n (%) 76 (57) 11 (52) 0.68

Sex/male, n (%) 64 (48) 6 (28) 0.095
ASA operative risk n (%)
C3 48 (36) 7 (33) 0.80
COPD n (%)
FEV1 \50 % 8 (6) 0 0.24
CVF \50 % 9 (7) 0 0.21
Pathologic background of the disease

Ethiology, n (%)
Malignancy 116 (87) 21 (100) 0.08
Adenocarcinoma 84 13 0.41
Squamous 11 2 0.31
Metastases 15 4 0.79
Others 6 2 0.70
Benign 17 0 0.131
Tumor size, [3 cms, n (%) 23 (20 %) 8 (38 %) 0.039
Lymph node metastases N2, n (%) 5 (3) 2 (10) 0.21

Surgical and short term outcomes

Operative times (min) mean (SD) 185 (49) 198 (46) 0.28
Surgical side left n (%) 71 (53) 11 (52) 0.91
Associated resections n (%) 4 (3) 2 (10) 0.21
R0 resections n (%) 116 (97) 19 (90) 0.11
Postoperative complications n (%)
Overall morbidity 33 (24.8) 2 (9) 0.12
Major complications 10 (7) 1(4) 0.19
Air leak 5 (3) 0
Atrial fibrillation 7 (5) 1(4)
Empyema 2 (1) 0
Mortality 0 0
Hospital Stay (days) mean (SD) 7.9 (4.7) 6.6 (3) 0.14

COPD chronic obstructive pulmonary disease

groups (7.9 vs 6.6 days). Overall morbidity (24 vs 9 %)
and major complications (7 vs 4 %) were also similar in
both groups. There was no mortality in our series (see
Table 1).
In the analysis of surgeon’s learning curve impact, there
were no preoperative differences in the characteristics of
the study population between the first 77 patients (Period 1)
and the second period (78–154) (see Table 2). Operative
times and blood loss were greater in period 1, without
significantly statistically differences between these values.
There were no impact in overall rate conversion in both
groups (14 vs 13 %), the presence of tumor adhesions
being the first reason for conversion. Emergency
conversion was always secondary to vascular accidents and
happened in five patients (see Table 3), all of the first
group (p = 0.059).
Discussion
Vats lobectomy is a surgical approach which becomes
popular around the world in the last 20 years. Since the first
descriptions, series of patients who underwent lung resec-
tion through this procedure were published worldwide [6,
7]. Although there is a lack of information from random-
ized studies comparing with the open approach, VATS

123
 Updates Surg

Table 2 Learning curve and conversion rate

Variable First period (1–77) n = 77 Second period (78–154) n = 77 P

Age [65 years, n (%) 43 (56) 44 (57) 0.87

Sex/male n (%) 36 (46) 34 (44) 0.74
ASA operative risk n (%)
C3 25 (32) 30 (39) 0.41
COPD n (%)
VEF1 \50 % 4 (5.2) 4 (5.2) 0.98
DLCO \50 % 6 (7.8) 3 (4) 0.31
Operative times (min) mean (range) 230 (190–410) 175 (153–370) 0.09
Bleeding median (range, ml) 420 (100–1750) 305 (100–850) 0.078
Vats conversion n (%) 11 (14) 10 (13) 0.41
Pleural adhesion 2 5 0.32
Compromise of adjacent organs 1 2
Determine the optimal resection 1 2
Difficult to perform a single-lung ventilation 2 1
Emergency conversion 5 (6.5) 0 0.059

Table 3 Emergency conversions

Variable Pulmonary vein injuries (n = 2) Pulmonary artery injuries (n = 3)

Technical problems (stapler malfunction) 2 –

Vascular injury during:
Lymph node dissection – 1
Pulmonary artery branch manipulation 1
Bronchial dissection 1

techniques seems to offer to the patients significantly less
pain in the postoperative course [8, 9], improvement of
pulmonary functions than after a thoracotomy [10], fewer
complications [11] and similar operative’s times with
similar rates of survival [12].
These evidences justified that rates of patients who
underwent VATS were increased the last years, extending
many times the indication of surgery to locally advanced
tumors which could also increase rates of patients that
underwent a conversion to open thoracotomy [13].
Regarding this, conversion rates range from 2 to 20 % [14,
15]; however, there is a lack of information in the literature
about factors related to conversion and the impact of
learning curve [16]. Similar to these series, in our study we
present rates of conversion of 13 % (21 of 154 patients).
Krasna et al. [17]., in his series of 321 patients, described
that pleural adhesions and oncological reasons, such as
unexpected tumor’s which infiltrate organs like the chest
wall, diaphragm, or great vessels, were the most common
causes of conversion to open lobectomy. McKenna pre-
sented in 2006, 1100 cases with 2.8 % of conversion to
thoracotomy with similar causes [18]. In our series, we
divided those conversions that were not performed due to
an emergency situation, where the most frequent causes of
conversion are in line with others publications (see
Table 2). Pleural adhesions were our most common causes
of conversion in seven patients. Difficult to perform a
single lung ventilation (3), compromise of adjacent organs
(3) and evaluation of the tumor to determine the optimal
resection (3) were the others causes of conversion. These
life-threatening situations to the patient who required an
emergency thoracotomy were presented in five patients due
to vascular injuries of branches of pulmonary artery (3) and
the pulmonary vena (2). We present also a univariated
statistically analysis of those preoperative conversion’s
related factor which could predict these events. In our
study, there were no difference regarding age, gender,
preoperative treatments, ASA score or pulmonary function
which could allow predict an intraoperative conversion.
With respect to histopathological findings neither type of
cancer, malignancy or not, or limph node metastases (N2)
presented differences between converted and non’-con-
verted patients. However, we found that those patients who
presented in preoperative images of tumors greater than
3 cm, presented higher risk of VATS conversion to
thoracotomy.
Surgeon’s training in VATS lobectomy and its impact in
patient’s outcomes, was previously analyzed in some

123
Updates Surg
reports [19, 20]. In line with other centres around the
world, our experience comes from a unique surgical team
who change the traditional posterior approach trough a
posterolateral thoracotomy to an anterior approach trough a
miniinvasivally incision. The complete VATS lobectomy is
done through a mini-thoracotomy (4–5 cm) with two
additional ports of 10 mm, and performed under indirect
vision looking on the monitor without rib spreading. This
represents a major challenge to the surgeons, which must
be experienced only in selected hospital with high volume
of patients [21].
The impact of learning curve of VATS lobectomy
conversion was also poorly defined. The number of patients
needed to learn surgical skills is variable. It was described
that surgeons feel comfortable with this technique after the
first 50 VATS, and VATS lobectomy should be only per-
formed after 100 cases of other VATS procedures like lung
wedge resections or pleural biopsies [22]. Li et al. in his
study of 200 patients who underwent VATS lobectomy
[20] evaluated learning curves for operative time, blood
loss, and postoperative length of stay, concluding that
between 100 and 200 cases are required to achieve effi-
ciency in these variables, and consistency requires even
more cases. Regarding VATS conversion, they presented
higher rates of conversion to open surgery in the first cases
of the study.
We propose a statistical analysis dividing our whole
series in two periods, according to the possible impact of
the number of operated patients and also the achieved
surgeon’s expertise and confidence, to understand the real
impact of learning curve in rates of conversion. In the first
period which includes the first half of operated patients,
there was an overall rate conversion of 14 % (11 patients).
Comparing with the last half of patients (13 % of rate
conversion (10 patients), there were no difference between
both periods. We understand these results, as the intention
to treat patients with a mini-invasively approach that would
be performed always, unless contraindications exist to
VATS procedures, because of the information that the
thoracoscopical vision gives about the tumor and its
extension and the anatomy of the thoracic cavity of every
patient is crucial for intraoperative decisions. In line with
others publications, pleural adhesion was the most common
cause of controlled conversion and that did not change
through the times.
In contrast, in our experience rates of emergency con-
version were changing through the times. In our study,
there were statistically differences (p = 0.05) between the
first and the second period in VATS conversion due to
intraoperative accidents. In the first 77 patients, we present
five emergency conversions due to vascular accidents. Two
cases due to technical problems (stapler malfunction) with
injury of pulmonary vein and in three patients due to injury
of pulmonar artery or its branches. We did not present
emergency conversion in the last series of patients, which
demonstrates the positive impact of learning curve in this
part of our surgical experience. Surgeon’s confidence and
the achievement of surgical techniques on critical steps in
VATS lobectomy surgery, like vascular control during
artery and vena pulmonary dissections, were important
keys in our experience to conduct 77 VATS lobectomy
without emergency conversion due to vascular accidents.
Regarding morbidity and mortality in those converted
patients, there are still controversies in the literature about
the risk of VATS conversion. Winter et al. [23] in a mul-
tivariate analysis of 1008 patients who underwent a VATS
surgery, describe that patients who presented a conversion
to open thoracotomy had a significantly correlation
(P \ 0.001) with the incidence of complications. On the
contrary, there are few studies about failed VATS surgery
who presents no differences about postoperative morbidity
and mortality in converted patients in compares with
complete VATS lobectomy [24, 25]. In line with these
studies we did not find any differences between those
patients who underwent conversion to open lobectomy and
those with complete VATS surgery, presenting in our
series, an overall morbidity in converted and non-converted
group of 9 and 24 %, respectively, with also similar rates
of major complications (4 vs 7 %).
A weakness of the present study is mainly its retro-
spective design, and, as such, it is unable to assess the true
role of the learning curve of this surgical strategy. On the
other hand, as mentioned above, prospective data collec-
tion may somehow mitigate this issue.
Conclusion
We present in this study the results of our initial experience
in VATS lobectomy. We emphasized in our analyses the
impact of the learning curve and the preoperative factors
which could determine an unplanned VATS conversion.
We concluded that surgeons should be expecting to
perform a conversion to a thoracotomy in patients who
present in preoperative studies, tumors [3 cm. Learning
curve only affected the emergency conversion, where only
the first half of our series of patients, presented these type
of conversion. Although each learning curve is individual
and multifactorial, and also the number of patients is
variable to develop proficiency in this issue, we showed
that a unique surgical team with more than 75 VATS
lobectomies performed, could reduce these kinds of intra-
operative accidents and also the rates of emergency
conversion.
In those patients who did not presents vascular acci-
dents, rates of conversion did not change through times.
123

Overall conversion remains similar in the whole series,
with no impact in converted patients outcomes. Regarding
this, we are convinced that those experienced surgeons in
VATS procedures, should start all pulmonary lobectomies
through a thoracoscopic approach, because these tech-
niques present several intra-operative advantages, and the
rates of morbidity of those converted patients are not
higher than those in whom a complete VATS lobectomy
could be achieved.
Compliance with ethical standards
Conflict of interest The authors declare no conflict of interests.
Ethical standard The study protocol was approved by the Ethics
Committee of the Department and conformed to the ethical guidelines
of the Helsinki Declaration (as revised in Tokyo 2004).
Research involving human participants and/or animals No ani-
mals but only human participants were engaged in our review.
Informed consent Since this study is a systemic review, no
informed written consent was needed. Patients anonymity was
maintained. The study was not advertised and no remuneration was
offered.
References
1. Roviaro G, Rebuffat C, Varoli FC, Mariani C, Maciocco M
(1992) Videoendoscopic pulmonary lobectomy for cancer. Surg
Laparosc Endosc Percutan 4:244–247
2. Landreneau RJ, Hazelrigg SR, Ferson PF (1992) Thoroscopic
resection of 85 pulmonary lesions. Ann Thorac Surg 5:415–419
3. Stantley DG (1992) Thorascopic lobectomy. Tenn Med
2:463–464
4. Lewis RJ, Sisler GE, Caccavale RJ (1992) Imaged thoracic
lobectomy: should it be done? Ann Thorac Surg 4:80–83
5. McKenna RJ Jr, Houck W, Fuller CB (2006) Video-assisted
thoracic surgery lobectomy: experience with 1,100 cases. Ann
Thorac Surg 81(2):421–425 (discussion 425–6)
6. Yim APC, Izzat MB, Liu HP, Ma CC (1998) Thoracoscopic
major lung resections: an Asian perspective. Semin Thorac Car-
diovasc Surg 10:326
7. Kaseda S, Aoki T, Hangai N (1998) Video-assisted thoracic
surgery (VATS) lobectomy: the Japanese experience. Semin
Thorac Cardiovasc Surg 10:300
8. Giudicelli R, Thomas P, Lonjon T et al (1994) Major pulmonary
resection by video assisted mini-thoracotomy. Initial experience
in 35 patients. Eur J Cardiothorac Surg 8:254–258
9. Kirby TJ, Mack MJ, Landreneau RJ, Rice TW (1993) Initial
experience with video-assisted thoracoscopic lobectomy. Ann
Thorac Surg 56:1248–1253
10. Nakata M, Saeki H, Yokoyama N, Kurita A, Takiyama W,
Takashima S (2000) Pulmonary function after lobectomy:
123
Updates Surg
videoassisted thoracic surgery versus thoracotomy. Ann Thorac
Surg 70:938–941
11. Hoksch B, Ablassmaier B, Walter M, Muller JM (2003) Com-
plication rate after thoracoscopic and conventional lobectomy.
Zentralbl Chir 128:106–110
12. Sugiura H, Morikawa T, Kaji M, Sasamura Y, Kondo S, Katoh H
(1999) Long-term benefits for the quality of life after videoas-
sisted thoracoscopic lobectomy in patients with lung cancer. Surg
Laparosc Endosc 9:403–410
13. Hanna JM, Berry MF, D’Amico TA (2013) Contraindications of
video-assisted thoracoscopic surgical lobectomy and determi-
nants of conversion to open. J Thorac Dis 5(Suppl 3):S182–S189.
doi:10.3978/j.issn.2072-1439.2013.07.08
14. Hennon M, Sahai RK, Yendamuri S et al (2011) Safety of tho-
racoscopic lobectomy in locally advanced lung cancer. Ann Surg
Oncol 18:3732–3736
15. Roviaro G, Varoli F, Vergani C et al (2004) Video-assisted
thoracoscopic major pulmonary resections: technical aspects,
personal series of 259 patients, and review of the literature. Surg
Endosc 18:1551–1558
16. Gazala S, Hunt I, Valji A, Stewart K, Bédard ER (2011) A
method of assessing reasons for conversion during video-assisted
thoracoscopic lobectomy. Interact CardioVasc Thorac Surg
12(6):962–964
17. Krasna MJ, Deshmukh S, McLaughlin JS (1996) Complications
of thoracoscopy. Ann Thorac Surg 61:1066–1069
18. McKenna RJ Jr, Houck W, Fuller CB (2006) Video-assisted
thoracic surgery lobectomy: experience with 1,100 cases. Ann
Thorac Surg 81(2):421–425 (discussion 425–6)
19. Jensen K, Ringsted C, Hansen HJ, Petersen RH, Konge L (2014)
Simulation-based training for thoracoscopic lobectomy: a ran-
domized controlled trial: virtual-reality versus black-box simu-
lation. Surg Endosc 28(6):1821–1829
20. Li X, Wang J, Ferguson MK (2014) Competence versus mastery:
the time course for developing proficiency in video-assisted
thoracoscopic lobectomy. J Thorac Cardiovasc Surg
147(4):1150–1154
21. Belgersa EHJ, Siebengaa J, Boscha AM, van Harenb EHJ, Bol-
lena ECM (2010) Complete video-assisted thoracoscopic surgery
lobectomy and its learning curve. A single center study intro-
ducing the technique in The Netherlands. Interact CardioVasc
Thorac Surg 10:176–180
22. McKenna RJ Jr (2008) Complications and learning curves for
video-assisted thoracic surgery lobectomy. Thorac Surg Clin
18(3):275–280. doi:10.1016/j.thorsurg.2008.04.004
23. Winter H, Meimarakis G, Pirker M, Spelsberg F, Kopp R, Rüt-
tinger D, Loehe F, Jauch KW, Hatz R (2008) Predictors of
general complications after video-assisted thoracoscopic surgical
procedures. Surg Endosc 22(3):640–645
24. Jones RO, Casali G, Walker WS (2008) Does failed video-as-
sisted lobectomy for lung cancer prejudice immediate and long-
term outcomes? Ann Thorac Surg 86:235–239
25. Sawada S, Komori E, Yamashita M (2009) Evaluation of video-
assisted thoracoscopic surgery lobectomy requiring emergency
conversion to thoracotomy. Eur J Cardiothorac Surg 36:487