No evidence that manual closure of the bronchial stump has a lower failure rate than mechanical stapler closure following anatomical lung resection

Abstract

A best evidence topic in cardiothoracic surgery was written according to a structured protocol. The question addressed was whether manual closure of the bronchial stump is safer with lower failure rates than mechanical closure using a stapling device following anatomical lung resection. One hundred and twenty-nine papers were identified using the search below. Eight papers presented the best evidence to answer the clinical question as they included sufficient number of patients to reach conclusions regarding the issues of interest for this review. Complications, complication rates and operation time were included in the assessment. The author, journal, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of the papers are tabulated. When looking at manual vs mechanical staples, it was noted that stapler failure can occur in around 4% of cases. The rate of bronchopleural fistula (BPF) development varied more in patients who underwent manual closure (1.5–12.5%) than in patients who underwent mechanical closure (1–5.7%). Although most of the studies reviewed showed no statistical differences between manual and mechanical closure in terms of BPF development, one study, however, showed that manual closure was significantly associated with lower numbers of postoperative BPF, while another study showed that mechanical closure is significantly associated with lower incidence of BPF. When looking at the role of the learning curve and training opportunities, it seems that the surgeon's inexperience when using mechanical staples can contribute to BPF development. A surgeon's experience can play a major role in the prevention of BPF development in patients having manual closure. Manual closure can provide a cheap and reliable technique when compared with costs incurred from using staplers, it is applicable in all situations and can be taught to surgeons in training with an acceptable risk. However, there is a lack of evidence to suggest that manual closure is better than mechanical stapler closure following anatomical lung resection.

INTRODUCTION

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

CLINICAL SCENARIO

A newly appointed thoracic surgeon discovers that the thoracic unit does not use bronchial staplers. The new surgeon feels that the use of bronchial staples to close the stump following lung resection is safer, with lower complication rates and can provide better training opportunities for junior surgeons; but is there any evidence?

THREE-PART QUESTION

In patients undergoing anatomical lung resection, does mechanical stump closure reduce the incidence of [bronchopleural fistula development], increase [effectiveness] and improve [training] opportunities when compared with manual closure?

SEARCH STRATEGY

Medline from 1948 to September 2012 using the PubMed interface (‘Wound closure techniques’ [Mesh]) AND (‘Respiratory tract fistula’ [Mesh]) AND (‘bronchial fistula’ [Mesh]). This was complemented by search using PubMed interface (‘Bronchial fistula’ [Mesh] AND ‘Closure of bronchial stump’). Related articles and references were screened for suitable articles.

SEARCH OUTCOME

One hundred and twenty-nine articles were found using the reported search strategy. From these, 8 articles were identified that provided the best evidence to answer the question. These are presented in Table 1.

Table 1:

Summary table

Authors, date, journal, country Study type (level of evidence)	Patient group	Outcomes	Key result	Comments
Asamura et al. (2000), Eur J Cardiothorac Surg, Japan [2] Retrospective study (level B)	Between1995 and 1997: 527 consecutive patients (533 bronchial stumps closed) Patients underwent pulmonary resections for different conditions (non-anatomical pulmonary resections and bronchoplastic procedures, which did not include closure of the bronchial stump, were excluded) 483 (91%) patients underwent mechanical closure, and 50 (9%) patients had manual closure 431 patients underwent lobectomy (397 had mechanical closure); 38 patients had bilobectomy (33 had mechanical closure) 42 patients had pneumonectomy (6 had mechanical closure) 22 patient had segmentectomy (17 had mechanical closure) The bronchial stump was reinforced in 45 patients due to the presence of risk factors	Endostaplers can be used safely for various types of bronchial closure and have a wide range of advantages such as saving surgical time	BPFs and effectiveness 7 BPFs developed postoperatively (1.3%): 2 after manual suturing (4%) and 5 after stapling (1%) (P = ns) 4 patients (0.7%) died of BPF-related complications Stapler failure occurred in 18 patients (3.7%). 1 patient had stapler failure in the theatre Training No specific data are presented but it is suggested that learning curves can play an important role in outcomes	The group's policy is to use mechanical staplers routinely except in cases when it is difficult to apply staplers The group of patients presented is heterogeneous and underwent surgery for benign or malignant conditions There are large differences between the two groups compared, which can lead to statistical bias The group used different types of mechanical staples in the study and the results can reflect the lack of experience in the use of certain type of staples (learning curve) Some of the patients who underwent mechanical closure had manual reinforcement but there is no specific information regarding the complication rate in this subgroup of patients No data regarding how many of the patients who developed BPFs had been ventilated postoperatively or whether these patients had clear postoperative margins
Sonobe et al. (2000), Eur J Cardiothorac Surg, Japan [3] Retrospective study (level B)	Between 1989 and 1998 547 consecutive patients (557 bronchial stumps closure) Patients underwent lung resection for lung cancer Different types of bronchial closure used: (A): manual closure, (B): mechanical closure, (C): combination of mechanical closure and manual reinforcement 507 patients underwent lobectomy (41 bilobectomy) [(A) 155, (B) 36 and (C) 326] 40 patients underwent pneumonectomy [(A) 9, (B) 4 and (C) 27] 12 patients received induction therapy	Both manual and mechanical closure are suitable methods for stump closure after lung resection; however, preventive coverage should be considered for stumps in patients with high risk of BPF development	BPFs and effectiveness 10 patients (1.8%) developed BPFs [method (A) 3 patients (1.8%), method (B) 2 patients (5%), method (C) 3 patients (1.9%)] with no statistical significance. None of the patients with BPFs were ventilated postoperatively 2 patients (20%) died of BPF-related complications No data regarding stapler failure Training No data presented	The group compared stumps that had dehiscence to uneventful ones using contingency table analysis and looked at different variables No data regarding stapler failure and how this was managed There were large differences in the number of patients between the two groups as method (A) was applied in 29.44% to total stump closure compared with only 7.2% in method (B) No data regarding the type of staplers used No data regarding margin checking by frozen section
Sirbu et al. (2001), Ann Thorac Cardiovasc Surg, Germany [4] Retrospective study (level B)	Between 1990 and 1999 565 consecutive patients (490 patients included) Patients underwent surgery for lung cancer Manual closure was performed in 317 patients (64.7%) and mechanical closure in 173 patients (35.3%) 295 patients underwent lobectomy (197 had manual closure) 15 patients had sleeve resections 5 patients underwent wedge resections 175 patients underwent pneumonectomy (103 had manual closure)	Manual or mechanical closure after lung resection can be suitable methods with no differences in terms of BPF	BPFs and effectiveness BPFs occurred in 22 patients (4.4%); 12 patients (54.6%) had pneumonectomy [8 patients had manual closure (7.6%), and 4 patients (5.7%) had mechanical closure] (P = ns) 9 patients (40.9%) underwent lobectomy [6 patients (3%) had manual closure, and 3 patients (2.9%) had mechanical closure] (P = ns) 22 patients (27.2%) had BPFs-related mortality No data available regarding stapler failure Training: No data presented	All pneumonectomy stumps on the right side were reinforced using different types of cover There are no data regarding stapler failure and whether this contributed to the development of BPFs The group included patients who had wedge resection in the analysis despite not having any bronchial closure which could influence outcomes The group presented specific data regarding risk factors contributing to the development of BPFs, but it was not linked to the type of closure used The group carried out frozen section to confirm clear margins before stump closure
Uçvet et al. (2011), Interact CardioVasc Thorac Surg, Turkey [5] Retrospective study (level B)	Between 2002 and 2007 625 consecutive patients Patients underwent surgery for malignant and benign lung conditions (excluding non-anatomical pulmonary resections, and bronchoplastic procedures, which did not include closure of the bronchial stump) 204 patients (32.6%) had manual closure; 421 patients (67.4%) had mechanical closure [regions that became weak because of free staples at the end points of the stump and recognized during stapler use were supported by suturing (170 patients)] Lobectomy was carried out in 419 (67%) patients [manual closure in 159 (37.9%)]. Pneumonectomy was carried out in 206 (33%) [45 (21.8%)] had manual closure 512 (81.9%) had risk factors for development of BPFs (171 patients had adjuvant or neoadjuvant therapy)	Closure with a stapler causes more BPFs than manual closure	BPFs and effectiveness BPFs occurred in 24 (3.8%) patients. The incidence of BPF was 5.0% (21/421) in the stapling group and 1.5% (3/204) in the manual group (P = 0.04) 6 patients (0.96%) had BPF-related mortality No data available regarding stapler failure Theatre effectiveness No data presented Training No data presented	The study included a large heterogeneous cohort of patients with different malignant and benign conditions that required lung resection There are no data regarding the stapler failure and whether it contributed to the development of BPFs; however, the group commented on the fact that when the edges became weak because of free staples at the end points of the stump, additional support was used The group also used a wide variation of supporting methods in the mechanical closure group in those considered to be high-risk patients, which did not appear to add any extra protection but probably affected the length of the operation. These patients were included in the statistical analysis and then analysed as subgroups No data regarding the policy of margin checking by frozen section
Al-Kattan et al. (1995), Eur J Cardiothorac Surg, UK [6] Retrospective study (level B)	Between 1980 and 1993 530 consecutive patients 471 patients underwent pneumonectomy for lung malignancy included 374 patients operated on by the consultant and 97 operated on by the training surgeons using manual closure	Manual closure of the bronchial stump after pneumonectomy provides a cheap and reliable technique, it is applicable in all situations and can be taught to surgeons in training with an acceptable risk	BPFs and effectiveness BPFs occurred in 7 (1.5%) patients 1 patient had a risk factor Identified 2 patients operated on by the consultant (0.5%) and 5 patients operated on by the training surgeons (5.1%) 2 patients of the 7 who developed BPFs had fatal complications Training Manual closure can be taught to surgeons in training with an acceptable risk The experience of the surgeon appeared to play a role in the results	The group identified the risk factors for BPFs The group reported that in the patients with BPFs, there was no evidence of avascular necrosis at the site of the stump, and no evidence of histological residual tumour at the revised stump The margins of the stump were only sent for histology if they looked suspicious or the tumour was proximal
Deschamps et al. (2001), Ann Thorac Surg, USA [7] Retrospective study (level B)	Between 1985 and 1998 713 consecutive patients Patients had pneumonectomy for primary malignancy in 607 (85.1%) patients, metastatic disease in 32 (4.5%) patients and benign disease in 74 (10.4%) 656 (92.0%) patients had mechanical closure. The stump was reinforced with viable tissue in 175 patients (24.5%) 360 (50.5%) patients had preoperative predisposing factors for the development of BPFs	Univariate analysis demonstrated that bronchial stump closure with staples had a protective effect against BPF compared with suture closure	BPFs and effectiveness BPFs occurred in 32 (4.5%) patients 25 (3.8%) in the stapled stump had fistula 7 (12.5%) had fistula in the manual closure (P = 0.009) BPF was significantly associated with bronchial stump reinforcement (P = 0.03) No data regarding mortality No data regarding stapler failure Training No data presented	Study included a large number of heterogeneous patients who underwent pneumonectomy for different conditions but excluding surgery for malignant mesothelioma The paper looked at the risk factors associated with the development of empyema and BPFs post-surgery Although most of the stumps were closed by mechanical staples, there are no data regarding the incidence of stapler failure and if it contributed to BPFs The group excluded patients who died within 24 h of surgery from analysis, which could have created bias No data regarding the policy of margin checking by frozen section and the group suggested that BPFs developed more in benign conditions The group presented no multivariate analysis
Wright et al. (1996), J Thorac Cardiovasc Surg, USA [8] Retrospective (level B)	Between 1980 and 1995 256 consecutive patients. Patients underwent pneumonectomy for different conditions Manual closure was used in all patients 133 patients had risk factors for BPFs	Carefully sutured closure of the main bronchus with a tissue buttress after pneumonectomy can yield excellent results	BPFs and effectiveness BPFs occurred in 8 (3.1%) patients 2 (25%) patients with BPF died Training The group reported that most of the BPFs noted in theatre were technically related. The group commented on the difficulties that can be encountered during surgery when using manual closure and suggested that it can contribute to the development of BPFs. This can give some indication to the difficulties in terms of training and contribution of the learning curve to the development of BPFs	The study included a heterogeneous cohort of patients undergoing surgery for different conditions No data regarding checking margins with frozen section
Panagopoulos et al. (2009), Interact CardioVasc Thorac Surg, Greece [9] Retrospective (level B)	Between 1999 and 2005 243 consecutive patients (221 were included) Patients underwent pneumonectomy for lung malignancy 14 (6.3%) patients underwent preoperative chemo/radio therapy 192 (86.9%) patients had mechanical closure Coverage of the bronchial stump was routinely carried out in 91 cases of right pneumonectomy, regardless of the type of closure	Mechanical stapling is superior to manual closure, although not an independent factor for the development of BPF	BPFs and effectiveness 5 (2.3%) patients developed BPF; 2/192 (1%) patients had mechanical closure and 3/29 (10.3%) manual closure A total of 7 (3.2%) patients died including 1 patient with BPF No data regarding stapler failure Training No data presented	Most of the patients had mechanical closure and the group did not report the reason behind using manual closure in the remaining small cohort of patients The results were obtained by univariate analysis only The group reported that the margin free of disease was histologically certified intraoperatively

COMMENTS

Asamura et al. [2] looked at bronchial closure in 533 consecutive stumps following anatomical pulmonary resection. This was reviewed in terms of the incidence of complications related to mechanical stapling (staple failure) and to bronchopleural fistula (BPF) formation. In this study, 50 (9%) stumps were closed by manual suturing and 483 (91%) by mechanical stapling. Different staplers were used in the mechanical closure group. This study revealed a 3.7% overall incidence of stapling failures (18 patients), but only 1 patient in this group developed BPF after surgery. Seven BPFs developed postoperatively (overall incidence, 1.3%) which included two after manual suturing (4%) and five after stapling (1%) (not statistically significant) leading to 4 cases of BPF-related mortality. The group concluded that although bronchial closure by stapling was accompanied by failure, its incidence was acceptable and was not directly associated with the development of BPF postoperatively and there is no difference in the incidence of BPFs when using manual or mechanical staples.

Uçvet et al. [5] assessed the effect of bronchial closure methods on BPF occurrence following pulmonary resections in 625 consecutive patients. Stumps were closed by manual suturing in 204 and by mechanical stapling in 421 cases. In the mechanical stapling group, stapling supported by manual suture was performed in 170 cases. BPFs occurred in 3.8% of patients, stapling was used in 5.0%, whereas manual suturing was used in 1.5% (P = 0.04). The study concluded that optimum bronchial closure method has to be chosen by considering the patient- and bronchus-based characteristics. This has to be assessed carefully, especially in pneumonectomy and co-factors. The manual closure seems to be the more preferable method in high-risk patients. An additive support suture following stapling on the bronchial stump did not decrease the risk of BPF.

Deschamps et al. [7] analysed factors affecting the incidence of empyema and BPF after pneumonectomy in 713 patients who underwent pneumonectomy for different conditions. Univariate analysis demonstrated that the development of BPF was significantly associated with bronchial stump reinforcement (P = 0.03). Bronchial stump closure with staples had a protective effect against BPF compared with suture closure (P = 0.009). The group concluded that multiple perioperative factors were associated with an increased incidence of BPF after pneumonectomy including the prophylactic reinforcement of the bronchial stump with viable tissue; however, mechanical stump closure had a protective effect against the development of BPFs. It is important to note, however, that although this study did not demonstrate a reduced incidence of BPFs with tissue reinforcement, the group practice included the routine reinforcement of stumps in patients deemed most likely to develop BPFs and therefore represent a selection bias.

AI-Kattan et al. [6] studied the outcomes in 471 pneumonectomies that were performed by one surgical team for primary lung cancer. In this study, all operations were performed using a uniform hand suture technique. There were 7 cases of BPF (incidence of 1.5%). This study demonstrated clearly the importance of surgeons’ experience as the senior author performed 374 pneumonectomies with 2 fistulas (0.5%), while other surgeons in training performed 97 pneumonectomies with 5 fistulas (5.1%). The group concluded that suture closure to the bronchial stump after pneumonectomy provides a cheap and reliable technique that is applicable in all situations and can be taught to surgeons in training with an acceptable risk.

CLINICAL BOTTOM LINE

The risk of BPF development post-lung resection can be multifactorial. There is actually very little objective evidence of the superiority of manual stump closure over mechanical stapling closure in anatomical lung resection. Furthermore, there is very little objective evidence regarding the superiority of either technique in terms of training opportunities. Further studies are needed to assess safety, effectiveness and training variation between the two approaches.

Conflict of interest: none declared.