Does the division of the inferior pulmonary ligament in upper lobectomy result in improved short-term clinical outcomes and long-term survival?

Summary

A best evidence topic in thoracic surgery was written according to a structured protocol. The question addressed was ‘Does the division of the inferior pulmonary ligament (IPL) in upper lobectomy result in improved short-term clinical outcomes and long-term survival?’. Altogether 43 papers were found using the reported search, of which 6 studies represented the best evidence to answer the clinical question, including a previous best evidence topic study, a meta-analysis and 4 retrospective cohort studies. The author, journal, date and country of publication, patient group studied, study type and relevant outcomes and results of these papers are tabulated. Most of the enrolled studies reported that there is no significant difference between the division groups and the preservation groups in terms of drainage time, drainage volume, postoperative dead space and complications. While 3 cohort studies revealed unfavoured postoperative pulmonary function in the division groups, including lung volume, forced vital capacity and forced expiratory volume in 1 s. The previous meta-analysis and a recent cohort study also found that the division of IPL might lead to increased bronchus angle change or torsion. Moreover, 2 cohort studies found that the division of IPL could not improve the long-term survival of patients undergoing upper lobectomy. Current evidence showed that dividing the IPL could not result in clinical benefits but might lead to decreased pulmonary function instead. Therefore, we recommended not dissecting the IPL routinely during upper lobectomy.

A best evidence topic was constructed according to a structured protocol.

INTRODUCTION

A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].

THREE-PART QUESTION

In [patients undergoing upper lobectomy], does [the division of inferior pulmonary ligament (IPL)] result in [improved short-term clinical outcomes and long-term survival]?

CLINICAL SCENARIO

A 65-year-old male patient was planning to receive upper lobectomy for lung carcinoma by your surgical team. The IPL is to be divided routinely during the surgery to prevent postoperative complications caused by the free space, including pleural effusion, empyema and dead space. However, you are unsure wondering whether the division of the IPL could benefit the postoperative pulmonary function recovery, such as lung volume, forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV1). You resolved to check the literature for evidence.

SEARCH STRATEGY

We searched the Medline database using the PubMed interface from January 1982 to July 2022 with the following search term: (lobectomy) AND (pulmonary ligament).

SEARCH OUTCOME

A total of 43 papers were found using the reported search. In total, 6 papers were identified that provided the best evidence to answer the question (Table 1), after excluding irrelevant studies, noncomparative studies and the studies of which all available data were already included in the previous meta-analysis.

Table 1:

Best evidence papers

Author, date, journal and country Study type (level of evidence)	Patient group	Outcomesb	Key results	Comments
Khanbhai et al. (2013), Interact Cardiovasc Thorac Surg, UK [2] Best evidence topic (level II)	Included 9 papers (a survey, an RCT, a case–control study and 5 case series) Ovid databases Embase from 1974 to December 2012 and Medline from 1948 to December 2012 Division, n = 553 Preservation, n = 1872	Perioperative mortality	Division versus preservation (D versus P), 2–3.7% vs 0.26–0.8%	Bias caused by noncomparative studies (survey and case series) Inconsistency in indicators of outcomes reported in the cited papers
		Conversion	D versus P, 1.8–14.01% vs 2.5–17.6%
		Postoperative morbidity	D versus P , 2–19.1% vs 2.15–15.2%
		Prolonged air leak	D versus P , 4.22–11.33% vs 4.72–5.1%
Bu et al. (2016), J Surg Res, China [3] Cohort study (level III)	72 patients with stage I–III NSCLC who underwent VATS superior lobectomy March 2012 to August 2013 No preoperatively severe pleural adhesion or chronic obstructive pulmonary disease history Excluded cases with incomplete follow-up Division, n = 33 (left, n = 12; right, n = 21) Preservation, n = 39 (left, n = 15; right, n = 24)	Perioperative mortality	None	Limited sample size in both groups
		Delayed postoperative complications	D versus P, n = 2 vs n = 0, P = 0.119
		Drainage time (days, mean ± SD)	D versus P, 4.3 ± 3.1 vs 5.0 ± 3.9, P = 0.528
		Drainage volume (ml, mean ± SD)	D versus P, 1433 ± 405 vs 1302 ± 454, P = 0.115
		Change of the main bronchus anglec (°, mean ± SD)	3-month postop (D versus P): all, 64.00 ± 16.80 vs 56.85 ± 11.88, P = 0.144 6-month postop (D versus P): left, 82.5 ± 9.9 vs 72.0 ± 8.2, P = 0.038; right, 48.9 ± 9.4 vs 47.6 ± 7.1, P = 0.541; all, 65.98 ± 18.61 vs 57.37 ± 14.33, P = 0.126
		Lung capacity (ml, mean ± SD)	D versus P: 3-month postop, 3102 ± 415 vs 3553 ± 1107, P = 0.218; 6-month postop, 2810 ± 372 vs 3898 ± 913, P = 0.002
		FVC (l, mean ± SD)	D versus P: 3-month postop, 2.20 ± 0.64 vs 1.91 ± 0.43, P = 0.117; 6-month postop, 2.28 ± 0.63 vs 2.04 ± 0.59, P = 0.261
		FEV1 (l, mean ± SD)	D versus P: 3-month postop, 1.73 ± 0.46 vs 1.89 ± 0.52, P = 0.917; 6-month postop, 1.76 ± 0.55 vs 1.92 ± 0.58, P = 0.040
		FEV1% (mean ± SD)	D versus P: 3-month postop 79.43 ± 13.13 vs 91.06 ± 12.35, P = 0.015; 6-month postop, 82.38 ± 12.93 vs 90.49 ± 11.26, P = 0.073
		MVV (l, mean ± SD)	D versus P: 3-month postop, 61.97 ± 12.24 vs 54.69 ± 22.42, P = 0.321; 6-month postop, 67.82 ± 13.54 vs 56.24 ± 20.40, P = 0.085
Kim et al. (2019), Interact Cardiovasc Thorac Surg, South Korea [4] Cohort study (level III)	52 patients who underwent right upper lobectomy by VATS for lung cancera January 2011 to April 2014 Excluded patients with emphysematous destruction or interstitial lung disease, any cancer-related pneumonia, partial or diffuse pleural adhesion and incomplete pre- and postoperative CT images Division, n = 31 Preservation, n = 21	Atelectasis	D versus P, n = 11 vs n = 8, P ˃ 0.99	Limited sample sizes in both groups Study confined to right upper lobectomy only Incomplete pulmonary function data due to 5 follow-up loss
		Postoperative dead space	D versus P: 3–6-month postop, n = 7 vs n = 1, P = 0.12
		Delayed pleural effusion	D versus P, n = 11 vs n = 8, P ˃ 0.99
		Change of the main bronchus angle (°)	P = 0.93
		Lung volume (ml, mean ± SD)	1-Year postop (D versus P): RML, 367 ± 188 vs 355 ± 167, P = 0.81; RLL, 1477 ± 402 vs 11 432 ± 322, P = 0.67; RML + RLL, 1844 ± 478 vs 1787 ± 382, P = 0.65 Pre- and postoperative difference (D versus P): RML, −29 ± 111 vs −27 ± 97, P = 0.95; RLL, 335 ± 294 vs 397 ± 293, P = 0.46; RML + RLL, 306 ± 350 vs 369 ± 355, P = 0.53
		FVC (l)	1-Year postop (D versus P): 3.05 ± 0.63 vs 3.28 ± 0.63, P = 0.23 Pre- and postoperative loss (D versus P): −0.42 ± 0.33 vs. −0.16 ± 0.26, P = 0.007
		FEV1 (l)	1-Year postop (D versus P): 2.26 ± 0.46 vs 2.26 ± 0.32, P = 0.68 Pre- and postoperative loss (D versus P): −0.26 ± 0.22 vs −0.33 ± 0.21, P = 0.33
		5-Year OS	D versus P: 75.4% vs 90.5%, P = 0.25
Lv et al. (2020), World J Surg Oncol, China [5] Meta-analyses (level I)	Included 3 RCTs and 5 CCTs January 2001 to February 2019 630 patients who underwent upper lobectomy by either thoracotomy or MISa Excluded letters, editorials, case reports, reviews and articles failing to extract original data Division, n = 315 Preservation, n = 295	Drainage time	MD = 0.14, 95% CI −0.05 to 0.33, P = 0.15	All the included trails had limited sample sizes and were come from Asia Potential bias caused by different operative techniques and strategies used among the studies
		Drainage volume	First day: MD = 6.83, 95% CI 3.57–10.08, P < 0.0001 Second day: MD = −1.70, 95% CI −4.01 to 0.61, P = 0.15 Third day: MD = 4.01, 95% CI 2.25–5.77, P < 0.00001
		Postoperative dead space rate	OR = 1.33, 95% CI 0.72–2.46, P = 0.36
		Postoperative dead space ratio	Left: MD = 0.76, 95% CI 0.13–1.64, P = 0.09 Right: MD = −1.92, 95% CI −4.07 to 0.23, P = 0.08
		Postoperative complications	OR = 1.20, 95% CI 0.66–2.19, P = 0.56
		Change of the main bronchus angle (°)	Left: MD = 3.96, 95% CI −9.40 to 17.33, P = 0.56 Right: MD = 5.00, 95% CI 1.68–8.33, P = 0.003
Moon et al. (2021), J Clin Med, South Korea [6] Cohort study (level III)	80 patients who underwent left upper lobectomy by VATS for early-stage lung cancer January 2014 to January 2017 Excluded patients with severe pleural adhesion, emphysematous destruction or interstitial lung disease, preoperative cancer-related pneumonia and inadequate CT images Division, n = 40 Preservation, n = 40	Drainage time (days)	D versus P: 4.8 ± 3.4 vs 4.5 ± 3.8, P = 0.598	Limited sample sizes in both groups Study confined to left upper lobectomy only Lacking comparison of postoperative clinical symptoms and pulmonary function
		Change of the main bronchus angle (°)	D versus P: axial angulation, −5.7 ± 7.7 vs −2.9 ± 5.5, P = 0.065; vertical angulation, −19.2 ± 11.0 vs −15.7 ± 8.4, P = 0.114
		Main bronchus curvature index (mean ± SD)	1-month postop (D versus P): 1.11 ± 0.05 vs 1.08 ± 0.03, P = 0.003 Pre- and postoperative difference (D versus P): 0.06 ± 0.03 vs 0.03 ± 0.03, P < 0.001
		Lung volume (ml)	1-Year postop (D versus P): 1325.6 ± 383.3 vs 1412.8 ± 420.6, P = 0.335 Pre- and postoperative difference (D versus P): −736.3 ± 382.7 vs −558.1 ± 410.0, P = 0.043
Yazgan et al. (2021), Acta Chir Belg, Japan [7] Cohort study (level III)	840 NSCLC patients who underwent upper lobectomy and mediastinal lymph node dissection by VATS or thoracotomy January 2007 to June 2020 Excluded resections other than lobectomy, bilateral surgeries, rethoracotomies, incomplete resections, stage IV tumours, carcinoid tumours, benign or metastatic tumours, neoadjuvant therapies and operative mortalities Division (station 9 lymph node and IPL), n = 697 (left, n = 325; right, n = 372) Preservation, n = 143 (left, n = 40; right, n = 103)	Prolonged air leak	D versus P: n = 271 vs n = 60, P = 0.51	Significant difference in patients who underwent VATS and thoracotomy in the division group (P = 0.03) Lower survival rate in patients who underwent sleeve resection in the preservation group (P = 0.09) Different surgical teams and the decision of the division of the IPL made by surgeon’s own discretion may cause bias
		Postoperative dead space	D versus P: n = 323 vs n = 72, P = 0.41
		Drainage time (days)	D versus P: 9.7 ± 7.5 vs 9.9 ± 7.7, P = 0.80
		5-Year OS	D versus P: left, 65.0% vs 59.1%, P = 0.44; right, 64.8% vs 62.1%, P = 0.85; all, 64.9% vs 61.3%, P = 0.56

^aNo other detailed eligible criteria were mentioned in the study, such as histologic type and clinical or pathological staging.

^bAll indicators in this column refer to the operative laterality, unless otherwise specified (annotated with ‘all’, ‘left’ or ‘right’).

^cAngles formed either by the main bronchus and the truncus intermedius on the right operated side or by the main bronchus and the lower bronchus on the left operated side were measured by chest CT preoperatively and at least one month postoperatively.

CCT: clinically controlled trails; CI: confidence intervals; CT: computed tomography; FEV1: forced expiratory volume in 1 s; FVC: forced vital capacity; IPL: inferior pulmonary ligament; MD: mean difference; MIS: minimally invasive surgery; MVV: maximal ventilator volume; NSCLC: non-small-cell lung cancer; OR: odds ratio; OS: overall survival; postop: postoperatively; RCT: randomized controlled trial; RLL: right lower lobe; RML: right middle lobe; SD: standard deviation; VATS: video-assisted thoracic surgery.

RESULTS

In 2013, a previous best evidence topic study by Khanbhai et al. [2] reviewed the relevant evidence up to 2012 in the Ovid database Embase and Medline to investigate the necessity of dissecting the IPL during upper lobectomy. The authors analysed a survey conducted in the Japanese Society of Thoracic Surgery, 2 control studies with small sample sizes and 5 case series (3 routinely dissected the IPL while the other 2 did not). They observed no difference in clinical outcomes between the 2 groups and concluded that there was no convincing evidence for routinely dividing the IPL in upper lobectomy because it seems not to improve outcomes or reduce complications as theoretically believed. However, except for the randomized controlled trial, the other included papers were of low quality due to either the absence of direct comparison or the lack of important outcomes such as survival rate and direct indicators of bronchial deformation or postoperative pulmonary function, which might limit the validity of their conclusions.

Bu et al. [3] retrospectively studied 72 patients with stage I–III non-small-cell lung cancer who underwent VATS upper lobectomy by the same surgical team, and no perioperative death was observed in both the IPL division group and the IPL preservation group. There was no significant difference between the 2 groups for the delayed postoperative complications, drainage volume and drainage time. The primary bronchus angle, FVC (P = 0.117 at 3 months, P = 0.261 at 6 months) and maximal ventilator volume (P = 0.321 at 3 months, P = 0.085 at 6 months) were all comparable postoperatively between the 2 groups. However, the preservation group showed improved lung capacity (division versus preservation, 2810 ± 372 vs 3898 ± 913 ml, P = 0.002) and FEV1 (division versus preservation, 1.76 ± 0.55 l vs 1.92 ± 0.58 l, P = 0.040) after 6 months postoperatively and favoured FEV1% (division versus preservation, 79.43 ± 13.13% vs 91.06 ± 12.35%, P = 0.015) at 3 months postoperatively. In addition, the change in bronchus angle was significantly smaller for the surgical laterality in the preservation group for left lung resection either at 3 months (P = 0.046) or 6 months (P = 0.038) postoperatively, while it was comparable between the groups for right lung resection. Thus, the preservation of IPL might reduce the displacement of remnant bronchus and improve the recovery of the postoperative pulmonary function.

Kim et al. [4] retrospectively analysed 52 patients who underwent VATS right upper lobectomy for lung carcinoma by the same surgeon. They found the 2 groups comparable in the postoperative occurrence of atelectasis, dead space, delayed pleural effusion and extent of angle change. The postoperative FVC, FEV1 and the RML and RLL volume were consistent between the 2 groups (P = 0.23, 0.68, 0.81 and 0.67, respectively). The postoperative FEV1 decrease was comparable (P = 0.33), while the FVC loss was significantly lower in the preservation group (division versus preservation, −0.42 ± 0.33 vs −0.16 ± 0.26, P = 0.007). Moreover, there was also no significant difference of the 5-year overall survival between the 2 groups (division, 75.4%; preservation, 90.5%; P = 0.25). As a result, they considered that there is no need to divide the IPL in the upper lobectomy routinely.

In 2020, Lv et al. [5] only explored the impact of the division of IPL in short-term outcomes (except for pulmonary function as well as long-term survival) of upper lobectomy by performing a meta-analysis consisting of 3 randomized controlled trials and 5 prospective or retrospective clinically case–control trials, which included the 2 studies [3, 4] that we identified above. They reviewed the Medline, Embase, Ovid, Cochrane Library, China Biology Medicine disc and China National Knowledge Infrastructure databases for studies both in English and Chinese. There was no significant difference between the 2 groups in the bronchial angle change (only for the left lung resection), drainage time, drainage volume on the second postoperative day, postoperative dead space and postoperative complications. However, the drainage volume on the first and third days postoperatively and the bronchial angle change for the right lobectomy were significantly increased after dissecting IPL. These results indicated that dissecting IPL might not provide effective improvement to postoperative rehabilitation. Nevertheless, this study lacked the analysis of important parameters of pulmonary function and long-term survival in support of the authors’ conclusions, despite Bu et al. [3] and Kim et al. [4] reported relevant outcomes.

Moon et al. [6] retrospectively reviewed 80 patients who underwent VATS left upper lobectomy for early-stage lung cancer in the National Health Insurance Service database and 2 institutions in which a single surgeon was appointed to perform the surgery in each institution. There was no significant difference in the chest tube duration, change in left bronchial axial or vertical angulation. The postoperative left lung volume was comparable (P = 0.335), but the preservation group revealed significantly reduced perioperative left lung volume loss (division versus preservation, −736.3 ± 382.7 vs −558.1 ± 410.0 ml, P = 0.043). The postoperative main bronchus curvature index (P = 0.003) and changes in the main bronchus curvature index (P < 0.001) were also significantly decreased in the preservation group. Thus, preserving the IPL seemed to benefit the recovery and lead to less distortion of the left remaining lung.

Yazgan et al. [7] conducted a retrospective study of 840 patients who underwent upper lobectomy and mediastinal lymph node dissection from the hospital and national databases. Patients were categorized by whether their IPL and station 9 lymph nodes were dissected or not. No significant difference was observed in postoperative dead space, prolonged air leak and drainage time between the 2 groups. The 5-year overall survival rate was comparable regardless of the operative laterality (division versus preservation: left, 65.0% vs 59.1%, P = 0.44; right, 64.8% vs 62.1%, P = 0.85; all, 64.9% vs 61.3%, P = 0.56). Among the patients who received sleeve resections during the surgery, the preservation group presented superior prognosis (P = 0.009). The male gender, thoracotomy approach, tumours located on the right side and tumours larger than 3 cm were significant factors of postoperative dead space and/or prolonged air leak in the multivariable analyses (P < 0.001, P = 0.002, P < 0.001 and P = 0.02, respectively) other than the dissection of IPL. Therefore, the dissection of station 9 lymph nodes and IPL might not be a necessary and effective procedure in the upper lobectomy.

CLINICAL BOTTOM LINE

The division of IPL during upper lobectomy could not reduce the postoperative complications. Instead, it might result in decreased postoperative pulmonary function (such as lung volume, FVC and FEV1) by increasing the displacement of remnant bronchus compared to the preservation of IPL. Furthermore, patients undergoing upper lobectomy might not benefit from the division of IPL in terms of long-term survival. Thus, we recommend not dissecting the IPL routinely during upper lobectomy due to the lack of significant clinical benefits.

Funding

This work was supported by the Department of Science and Technology of Sichuan Province (2022JDKP0009).

Conflict of interest: None declared.

Reviewer information

Interactive CardioVascular and Thoracic Surgery thanks Nuria M Novoa, Noriyoshi Sawabata and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.