Results of Video-Assisted Techniques for Resection of Lung Cancer

Summary of Key Points

•

Many meta-analyses, outcomes and matched cohort studies demonstrate equal long-term outcomes for video-assisted thoracoscopic surgery (VATS) and open lobectomy.
•

Matched comparisons generally demonstrate equal long-term survival for VATS versus open lobectomy for lung cancer, suggesting that the improved survival seen in unmatched studies is due to confounding factors.
•

Many meta-analyses, outcomes, and matched cohort studies demonstrate similar operative mortality for VATS and open lobectomy.
•

Pain, hospital length of stay, and complications are lower after VATS than after open lobectomy.
•

There is no difference in the incidence of N2 nodal upstaging; the effect on N1 upstaging is unclear.
•

The ability to deliver adjuvant chemotherapy may be better after VATS lobectomy.
•

VATS lobectomy is associated with a learning curve of about 50 cases.

Over 100 years ago, Jacobaeus first reported the diagnosis and treatment of pleural effusions using a thoracoscope. Since then, the application of thoracoscopy to pulmonary resection has advanced as a useful adjunct for surgeons, perhaps most prominently for wedge resection or pleural procedures. From the late 1980s to the 1990s, surgeons began using VATS for lobectomies to treat patients with early stage nonsmall cell lung cancer (NSCLC). In 1992, VATS simultaneously stapled lobectomy without rib spreading was reported by Lewis et al. and with individual vessel and bronchial ligation was reported by Roviaro et al. During the following year, the outcomes of VATS lobectomy were published by Walker et al., Coosemans et al., Kirby et al., and Hazelrigg et al. The surgical approach has become far less invasive as the instrumentation has gradually improved over the past two decades and VATS has evolved into a basic and vital thoracic surgical technique.

However, the penetration of VATS anatomic resections (lobectomy and segmentectomy) has been slow to occur, and currently about 30% of anatomic resections in the United States are performed using VATS. These VATS lobectomies in the United States are performed primarily by dedicated thoracic surgeons. As reported in the database of the Society of Thoracic Surgeons (STS), which represents predominantly dedicated thoracic surgeons in the United States, the proportion of lobectomies done via VATS has increased from 19% in 2005 to 44% in 2009 and is currently 66%. A similar increase from 20% to 54% from 2007 to 2011 has been reported in Denmark, which has a highly centralized health-care system. Robust data about the overall proportion of lobectomies that are performed using VATS in Europe or Asia are not available, but estimates put this well below the penetration currently seen in the United States.

Definitions

Video-assisted thoracoscopic surgery (VATS) is used for wedge resection, segmentectomy, lobectomy, pneumonectomy, sleeve lobectomy, lobectomy with chest wall resection, and extrapleural pneumonectomy. The approaches vary substantially: reports on VATS lobectomy alone describe the use of one to six incisions from 4 cm to 10 cm in length, with and without rib spreading. In general, however, VATS lobectomy is interpreted to mean an anatomic lobectomy with ligation of individual bronchi and vessels and lymph node dissection or sampling using a minimal number of ports without a retractor or rib spreading.

A clear definition of VATS lobectomy is needed. The Cancer and Leukemia Group B established a definition of VATS lobectomy for a prospective, multi-institutional trial, and this definition has been widely adopted. A VATS lobectomy is defined as involving no rib spreading; maximum incisions of 8 cm; dissection of individual veins, arteries, and airways for the lobe in question; and node sampling or dissection identical to open thoracotomy. This definition was endorsed by most of the 55 experts participating in the 20th Anniversary of VATS Lobectomy Conference: The Consensus Meeting, organized by the Scientific Secretariat and the International Scientific Committee of the International VATS Lobectomy Consensus Group. However, several participants thought that a small retractor should be acceptable in specific circumstances, for example, when performing complex procedures such as bronchoplasty or when delivering a large specimen.

Some surgeons have suggested that VATS should include only procedures done exclusively with visualization on a monitor. This criterion should be abandoned for several reasons. The most important issues for patients with malignant diseases are the incisional trauma, achievement of curative surgery, and subsequent oncologic outcomes. Occasional viewing through a 4-cm, non–rib-spreading incision does not change the nature of the procedure. A hybrid VATS approach using both a monitor and direct vision without rib spreading or a robotic approach can provide a three-dimensional understanding of anatomy as well as magnified proximity visualization and would surely clinically facilitate complex procedures such as sleeve lobectomy or segmentectomy.

Technical aspects of how to accomplish a VATS lobectomy are beyond the scope of this chapter and are the subject of surgical atlases. Styles vary in how surgeons perform different steps of the operation. The essential feature, however, is a hilar dissection and individual division of the vessels and bronchus of the lobe. Most often, these structures are divided using an endostapler, but ligation, division between clips, or sealing with energy devices is also feasible for smaller vessels.

Patient Selection

Patient selection ultimately depends on the surgeon’s judgment about the ability to accomplish an oncologically appropriate resection of a given patient’s tumor. However, some general guidance can be obtained from the opinions of experts at the recent Consensus Meeting ( Table 27.1 ). Most of these experts consider VATS to be appropriate for stage I and II tumors unless there is substantial hilar involvement or the need for chest wall, bronchial, or vascular sleeve resection.

TABLE 27.1

Video-Assisted Thoracoscopic Surgery (VATS) Lobectomy Recommendations Derived From Consensus Statement of the International VATS Lobectomy Consensus Group

Data taken from Yan T, Cao C, D’Amico TA, et al. Video-assisted thoracoscopic surgery lobectomy at 20 years: a consensus statement. Eur J Cardiothorac Surg. 2014;45(4):633–639.

Indications for VATS Lobectomy
≤7 cm (T1, T2a, and T2b)	Recommended
N0 or N1 status	Recommended
Patients with previous thoracic surgery or pleurisy	Highly recommended
Contraindications for VATS Lobectomy
Chest wall involvement including rib(s)	Contraindicated
Central tumor invading hilar structure(s)	Contraindicated
FEV ₁<30%	Contraindicated
DLCO <30%	Contraindicated
Preoperative Investigations
PET/CT and sampling of positive mediastinal lymph nodes	Highly recommended
Sampling of positive lymph nodes by EBUS/EUS	Recommended
VATS assessment at the time of surgery	Highly recommended
Total ipsilateral lymph node dissection in all patients	Recommended
Indications for Conversion to Open Thoracotomy
Major bleeding	Highly recommended
Significant chest wall involvement	Recommended
Vascular sleeve	Highly recommended
Bronchial sleeve	Highly recommended
Bronchovascular sleeve	Highly recommended
Training
Number of cases to overcome steep learning curve: 50	Highly recommended
Resident case volume of a training center: >50/y	Recommended
Minimum case volume to maintain VATS skills: >20/y	Recommended
Proctoring should be necessary in all new VATS surgeons	Highly recommended
Future Directions
Establishment of multi-institutional database	Recommended
Increased exposure of VATS lobectomy to trainees	Highly recommended
Establishment of standardized VATS lobectomy workshops	Highly recommended

CT, computed tomography; DLCO, diffusing capacity of the lungs for carbon monoxide; EBUS, endobronchial ultrasound; EUS, endoscopic ultrasound; FEV ₁, forced expiratory volume in 1 second; PET, positron emission tomography.

Outcomes: Comparison of Vats and Open Lobectomy

The scope of what can be successfully performed thoracoscopically includes pneumonectomy, segmentectomy, sleeve lobectomy, and lobectomy with chest wall resection. However, the literature is too limited for most of these procedures to define outcomes, and the results are confounded by patient selection, making assessment of the impact of the VATS approach per se difficult at best.

By contrast, the literature on VATS lobectomy for cancer is so abundant that we have chosen to focus our review on meta-analyses and large outcomes studies. Only a few randomized controlled trials were conducted; these were early in the VATS experience and were small, underpowered studies. As the procedure matured and experience grew, increasing amounts of nonrandomized data showed that VATS lobectomy for cancer was safe, with outcomes similar to or better than those of open lobectomy. In the United States, the decision was made that a randomized controlled trial of VATS compared with open lobectomy would require a design to prove equivalence, which would necessitate too many patients and would not be worth the expense. Hence, the strongest data come from large outcomes and propensity-matched studies and not randomized controlled trials.

To provide an evidence base for this chapter, a comprehensive literature search was carried out for studies comparing VATS with open lobectomy or segmentectomy for lung cancer. We included reports that involved a meta-analysis, a randomized controlled trial, a propensity-matched or otherwise case-matched study, or an outcomes study using a large multi-institutional database. We did not include reports of individual, single-institution comparative series without case matching. Two systematic reviews were not included because a quality assessment judged them to be poor, and they included studies that were otherwise already captured. One meta-analysis was excluded because it focused only on three propensity-matched studies, which were already included individually. Another was excluded because it used an erroneous code for VATS lobectomy.

Short-Term Outcomes

The results of many studies have demonstrated that VATS lobectomy is a safe procedure ( Table 27.2 ). These studies, including meta-analyses, propensity-matched series, case-matched series, and randomized controlled trials, demonstrate that the conversion rate from VATS to open procedure is about 5% to 10%. Most of these conversions resulted from oncologic or technical factors; bleeding was the reason for conversion in only a few studies. Furthermore, conversion from VATS to an open procedure is not associated with increased perioperative mortality or complications when compared with planned open thoracotomy.

TABLE 27.2

Short-Term Outcomes of VATS Compared With Open Procedures

First Author	Year	N (Total)	Inclusion Criteria, Comments	Conversion Rate %	Operative Mortality %			Complications (Overall) %			Hospital Stay (Days, Median)
First Author	Year	N (Total)	Inclusion Criteria, Comments	Conversion Rate %	VATS	Open	p	VATS	Open	p	VATS	Open	p
Meta-Analyses
Cheng	2007	3589	About 20% rib spr	6	1.2	1.7	NS	13	20	0.0002	Lower	—	0.007
Chen	2013	3457	Stage I	—	—	—	—	20	29	<0.0001	Lower	—	<0.01
Yan	2009	2641	20% Rib spr	8	0.4	0.7	NS	—	—	—	12	12	—
Cai	2013	1564	Stage I	—	—	—	—	Lower	—	0.013	—	—	—
Propensity-Matched Series
Paul	2013	41,039	NIS	—	1.6	2.3	NS	41	45	<0.001	5	7	<0.001
Yang	2016	18,780	NCDB	—	1.5	1.8	NS	—	—	—	5	6	<0.01
Falcoz	2016	5442	ESTS DB	—	1.0	1.9	0.02	29	32	<0.04	6	8	0.0003
Cao	2013	3634		—	1.3	1.8	NS	25	35	0.0001	Lower	—	<0.00001
Cao	2013	2916	Chinese DB	8	0.8	1.1	—	—	—	—	—	—	—
Paul	2010	2562	STS	—	0.9	1	NS	26	35	<0.0001	4	6	<0.0001
Scott	2010	752	Stage cI	—	0	1.6	NS	27	48	NS	5	7	<0.001
Flores	2009	741	Stage cIa	18	0.3	0.3	NS	24	30	0.05	5	7	<0.001
Villamizar	2009	568	Prospective DB	5	3	5	0.02	31	49	—	4	5	<0.0001
Lee	2013	416	Cornell U	2	1	3	NS	15	18	—	4	5	0.02
Ilonen	2011	232	Stage cI	14	2.6	3.4	NS	16	27	<0.03	8	11	0.001
Jeon	2013	182	COPD, Stage cI	11	0	3.3	NS	22	33	NS	6	9	0.04
Scott	2010	136	Stage cI	7	1.4	1.6	NS	34	39	NS	4	7	<0.0001
Yang	2016	60	Preop chemo	—	3	7	NS	40	57	NS	4	5	0.007
Case-Matched Series
Cattaneo	2008	164	Elderly	1	0	3.6	NS	28	45	0.04	5	6	0.001
Jones	2008	78	Converted	11	0	2	NS	50	48	NS	8	8	NS
Demmy	1999	38	Old, frail	14	16	5	—	32	32	—	5	12	0.02
Outcomes Studies (Adjusted Data) ^a
Ceppa	2012	12,970	STS	—	—	—	—	Lower		0.001	—	—	—
Ceppa	2012	—	Hi pulm risk	—	—	—	—	Lower		0.02	—	—	—
Farjah	2009	12,958	SEER Medicare	—	Lower	—	NS	—	—	—	4	8	<0.001
Park	2012	6292	NIS	—	—	—	NS	Lower		0.004	Lower	—	0.001
Swanson	2012	3961	Premiere DB	—	—	—	—	Lower		0.02	6	8	<0.0001
Licht	2013	1513	Stage cI. DLCR	—	1.1 ^b	2.9 ^b	0.02 ^b	—	—	—	—	—	—
Randomized Controlled Trials
Craig	2001	110	—	—	0	0	NS	3	8	—	(9) ^c	(8) ^c	(NS) ^c
Kirby	1995	55	Stage cI	10	—	—	—	24	53	<0.05	7	8	NS

COPD, chronic obstructive pulmonary disease; DB, database; DLCR, Danish Lung Cancer Registry; ESTS DB, European Society of Thoracic Surgery Database; Hi pulm risk, high pulmonary risk; NCDB, National Cancer Database (US); NIS, National Inpatient Sample (a representative large sample of US hospital admissions); NS, not significant; NS (italics) ; not significant, but a trend (i.e., p ≤ 0.1 but > 0.05); preop chemo, preoperative chemotherapy; rib spr, rib spreading; SEER, Surveillance, Epidemiology and End-Result database; STS, Society of Thoracic Surgeons Database; U, university; VATS, video-assisted thoracoscopic surgery.

a Reported data are that adjusted for multiple predictive factors (multivariate analysis).

b Unadjusted data.

c Study protocol demanded a minimum 7-day hospitalization.

The studies included in this review ( Table 27.2 ) almost universally show no significant difference in perioperative mortality between VATS and open lobectomy ( Fig. 27.1 ). Closer examination shows that, in fact, the trend to lower perioperative mortality rates with VATS is consistent. This lower mortality rate is true even among special populations of patients (e.g., older, frail, high pulmonary risk) in case-matched series. The length of hospital stay has been significantly lower for VATS resection in most studies ( Fig. 27.2 ). The difference is approximately 1 day to 3 days, on average. Comparison between studies is difficult, however, because the average length of stay varies markedly, most likely reflecting differences in regional standards and the structure of the health-care system.

The overall rate of complications is also significantly lower in VATS resection in most studies ( Table 27.2 , Fig. 27.3 ). Comparison of complications among studies is difficult because of the varying definitions of a complication and the degree of severity of complications that were counted. Certain specific relevant complications have been reported in some studies ( Table 27.3 ). A fairly consistent trend toward lower rates of pneumonia, prolonged air leak, arrhythmia, and need for postoperative mechanical ventilation has been demonstrated, with the difference being significant in less than half of the studies.

TABLE 27.3

Short-Term Outcomes of VATS Compared With Open Procedures: Specific Complications

First Author	Year	N (Total)	Inclusion Criteria, Comments	% Pneumonia			% Prolonged Air Leak			% Arrhythmia			% Mechanical Ventilation
First Author	Year	N (Total)	Inclusion Criteria, Comments	VATS	Open	p	VATS	Open	p	VATS	Open	p	VATS	Open	p
Meta-Analyses
Cheng	2007	3589	~20% Rib Spr	Lower		NS	Higher ^a		NS	Same	Same	NS	Lower		NS
Chen	2013	3457	Stage I	2	5	0.03	5	7	NS	10	12	NS	—	—	—
Yan (all)	2009	2641	20% Rib Spr	2	10	NS	5	6	NS	10	10	NS	—	—	—
Cai	2013	1564	Stage I	Lower	—	NS	Lower	—	NS	Lower	—	0.05	—	—	—
Yan (no Rib Spr)	2009	925	No Rib Spr	0.5	2	NS	2	2	NS	4	4	NS	—	—	—
Propensity-Matched Series
Paul	2013	41,039	NIS	7	8	NS	—	—	—	14	18	<0.001	5	6	NS
Falcoz	2016	5442	ESTS	6	6	NS	10	9	NS	5	5	NS	0.7	1.4	<0.008
Cao	2013	3634		3	5	0.008	8	10	0.02	7	12	<0.00001	—	—	—
Paul	2010	2562	STS	3	4	NS	8	9	NS	7	12	0.0004	0.5	0.6	NS
Scott	2010	752	Stage cI	—	—	—	2	7	NS	9	13	NS	0	4	—
Flores	2009	741	Stage cIa	—	—	—	4	4	NS	10	11	NS	—	—	—
Villamizar	2009	568	Prospective DB	5	10	0.05	13	19	0.05	13	21	0.01	—	—	—
Lee	2013	416	Cornell U	1	3	—	6	6	—	6	6	NS	1	3	—
Ilonen	2011	232	Stage cI	4	3	—	4	10	—	1	3	—	0	1	—
Jeon	2013	182	COPD, Stage cI	1	11	0.01	11	15	NS	8	9	NS	—	—	—
Yang	2015	60	Preop chemo	7	13	NS	10	20	NS	23	23	NS	0	3	NS
Outcomes Studies (Adjusted Data) ^b
Ceppa	2012	12,970	STS	3 ^c	5 ^c	<0.001	—	—	—	—	—	—	0.4 ^c	0.8 ^c	0.002
Swanson	2012	3961	Premiere DB	Same	Same	NS	Lower	—	NS	Lower ^c	—	—	—	—	—
Randomized Controlled Trials
Craig	2001	110	—	0	6	—	—	—	—	0	2	—	2	0	—
Kirby	1995	55	Stage cI	—	—	—	12	27	NS	—	—	—	—	—	—