Postoperative prolonged air leaks (PALs) occur after thoracic surgery in which lung parenchyma is resected, divided, or manipulated. These air leaks can place patients at risk for intensive care unit readmissions, longer hospital length of stay, and infectious complications. Studies have been conducted to identify patients who are at risk for air leak and several methods have been examined for the prevention and treatment of PALs. A standard method of air leak prevention or treatment has not been established. This article discusses the prophylactic measures that have been studied for the prevention of PALs following lung surgery.
There is no gold standard for the prevention of postoperative air leaks after thoracic surgery.
In patients with intraoperative air leaks, use of polyglycolic acid mesh and sealants is safe and may decrease the duration of air leak, chest drainage, and hospital stay.
Pleural tenting is a useful tool for the prevention of air leak in upper lobectomies.
There is no standard management for postoperative tube thoracostomy, although, conditionally, the use of continued suction is not superior to water seal in reducing the duration of air leak.
Postoperative prolonged air leaks (PALs) occur after thoracic surgery, most often in patients undergoing either oncologic lung resection or lung volume reduction surgery (LVRS). Air leaks occur via parenchymal defects, preventing sealing of the pleural space after resection. Between 8% and 26% of patients who undergo lobectomies and up to 46% of patients undergoing LVRS have PALs.
Although PALs can be managed in multiple ways, the sequelae of these leaks are increased intensive care unit readmissions, longer hospital stays, a higher incidence of pneumonia, and pleural space infections. Historically, the management of PALs following thoracic surgery has been limited to watchful waiting with continued tube thoracostomy for drainage. PALs secondary to alveolar-pleural fistulas rather than bronchopleural fistulas are likely to resolve over time with expectant management, although some controversy exists with regard to chest tube management. In addition, the use of 1-way valves averts the need for prolonged hospital stays for patients who can tolerate water sealing of their chest tubes.
Although there are several methods that have been used to treat PALs after they are diagnosed, populations at risk for developing an air leak have been identified and prophylactic measures can be taken at the index operation to prevent the onset of PALs. Female patients or those with chronic obstructive pulmonary disease, low forced expiratory volume in 1 second (FEV 1 ), a smoking history, diabetes, or chronic steroid use have been identified as high risk for developing PAL following lobectomy. Patients undergoing LVRS who have low diffusing capacity of the lungs for carbon monoxide and FEV 1 , pleural adhesions, or diffuse emphysema are also more prone to develop PALs. This article discusses the several prophylactic measures that have been studied for the prevention of PALs following lung resection, including the use of absorbable mesh, parenchymal sealants, pleural tenting, staple line buttresses, and prophylactic pneumoperitoneum.
Pleural mesh patch
The use of polyglycolic acid (PGA) mesh sheets for the coverage of pulmonary parenchyma has been studied in various contexts ( Table 1 ). Nakamura and colleagues evaluated the safety of PGA staple line buttress as well as PGA mesh coverage for stapled, sutured, and unsutured parenchyma following sublobar and lobar resections in a cohort of 344 of 1026 patients. No additional sealant was added, and there were no differences in surgical site infectious complications between patients with and without mesh use. Yoshimoto and colleagues examined mesh coverage of the parenchyma compared with suture closure of the visceral pleura in the setting of segmental resection. The purpose of mesh coverage of the parenchyma was to augment the benefit of lung preservation provided by performing sublobar resection; however, transection of the parenchyma resulted in exposed raw lung surface that predisposed patients to PAL. Although suture closure of the pleura had conventionally been used, this theoretically reduced lung expansion. Therefore, the utility of the PGA sheet was to cover the intersegmental plane, minimize PAL, and maximize lung function relative to suture closure. Although this study did not reveal a difference in postoperative lung function, it was not designed to compare PAL in mesh versus nonmesh coverage.
|Author, Year||Study Design (n)||Dissection and Closure||Duration Chest Tube Drainage||Mean Length of Stay (d)|
|Mesh||No Mesh||P Value||Mesh||No Mesh||P Value|
|Ueda et al, 2007||Prospective (45)||Stapler |
|Ueda et al, 2010||Retrospective (122)||Stapler |
PGA mesh a
|Saito et al, 2017||Retrospective (133)||Cautery b |
Since these initial studies, the utility of PGA mesh coverage for the prevention of PAL has been studied alone and in combination with other closure techniques for coverage of raw parenchymal surfaces following oncologic resection as well as lung resection for benign disease ( Fig. 1 ). Saito and colleagues reexamined the use of PGA mesh in addition to fibrin glue to prevent pulmonary complications after segmental resection for stage IA non–small cell lung cancer in a case-control study. Although the investigators confirmed no differences in pulmonary function for up to 6 months postoperatively, they noted an increased incidence of PAL associated with mesh use compared with suture closure of the pleura on univariate (8.7% vs 0%; P = .042) and regression analysis (odds ratio, 5.26; P = .047).
Okada and colleagues routinely used a PGA patch in combination with a fibrin sealant to cover the intersegmental plane following thermal dissection during segmentectomy, as described in a series of 52 patients who underwent segmentectomy for stage I cancers. There were 4 PALs (7.7%), with a 1-day median duration for all air leaks and thoracostomy drainage for a median 3 days for all patients. Kawai and colleagues applied low-voltage energy as well as PGA mesh and fibrin sealant to pleural defects in 40 patients with an intraoperative air leak out of a total cohort of 176 who had undergone thoracoscopic lobectomy for neoplasm. They compared patients with energy, mesh, and fibrin with those without intraoperative leaks for whom only mesh and fibrin were used. Intraoperative air leaks were defined as air leakage with a −20 cm H 2 O pressure load. Air leaks resolved in nearly 80% of patients and, for those with postoperative air leaks, no patients had leaks >7 days (median 3.5 days). Note that patients with intraoperative air leaks required 3 more days of chest drainage compared with those without intraoperative air leaks (5.3 days vs 2.2 days). These data suggested a possible utility for mesh pneumostasis in patients with an intraoperative air leak; however, this method of pneumostasis did not entirely avert the need for prolonged chest drainage.
Ueda and colleagues , , addressed this in a series of studies on patients with intraoperative air leaks discovered on intraoperative water-seal tests following resection for neoplastic disease. Ueda and colleagues attempted to prevent postoperative air leaks by identifying leaks intraoperatively and applying a fibrin sealant–soaked PGA mesh in 28 patients compared with 17 patients who had no intraoperative leak and were consequently not treated with mesh or sealant. There was no significant difference in the duration of chest drainage in patients who underwent mesh and fibrin closure relative to those who did not, and all intraoperative air leaks resolved immediately. Two patients developed delayed air leaks requiring redrainage. A follow-up study by Ueda and colleagues described results for 133 patients with intraoperative air leaks who underwent mesh and fibrin pneumostasis compared with 73 with no intraoperative air leak and no mesh pneumostasis. Similar to the initial study, delayed pneumothorax occurred in treated patients (3%) requiring redrainage, but there was no difference in median chest tube duration between those with and without an intraoperative air leak. Importantly, Ueda and colleagues described a safe and effective method of prophylactic mesh and fibrin application; however, these studies did not offer a direct comparison of mesh versus no mesh pneumostasis, because intraoperative air leaks may not have been clinically significant if left untreated. A separate study of 145 patients with intraoperative air leaks who underwent fibrin alone compared with mesh plus fibrin closure did reveal that patients with mesh and fibrin pneumostasis had 1 fewer day of chest tube drainage as well as a shorter hospital stay.
Lee and colleagues applied the use of PGA mesh pneumostasis for patients undergoing bullectomy and mechanical pleurodesis for spontaneous pneumothoraces. Notably in patients who underwent bullectomy, pleurodesis, and PGA mesh pneumostasis, there was a higher recurrence-free rate over a 24-month follow-up period (94.9% PGA mesh vs 89% no mesh; log-rank P = .047). Although 4 patients treated without PGA mesh pneumostasis required reoperation for recurrent pneumothoraces, all patients with recurrences following mesh pneumostasis were successfully treated nonoperatively. Saito and colleagues followed 11 patients after bullectomy with PGA mesh staple line reinforcement for a median of 11 months with no air leaks postoperatively and no recurrent pneumothoraces. Hirai and colleagues retrospectively reviewed 173 patients treated with bullectomy and PGA mesh and fibrin coverage to reinforce staple line pneumostasis and found that only 3% of patients had recurrent pneumothorax compared with almost a 20% recurrence rate in patients treated during the same period with only stapling.
Although these data for pleural patch pneumostasis in combination with several other techniques for closure are compelling, there have been no trials in the United States prospectively examining the role for prophylactic mesh pneumostasis in any patients undergoing open or thoracoscopic lung resection.
Polyethylene Glycol Hydrogel
Several sealants have been studied, including homologous and autologous fibrin-based sealants, and polyethylene glycol (PEG)–based absorbable sealants. This article reviews both fibrin-based and PEG-based sealants, because these have been studied extensively for the treatment of intraoperative air leak and the prevention of PALs ( Table 2 ). Although there are multiple formulations of homologous fibrin-based sealants, these are generally composed of human fibrinogen and thrombin. Autologous fibrin-based sealants are generated from patient blood samples taken intraoperatively.
|Author, Year||Study Design (n)||Experimental Technique||Duration Chest Tube Drainage||Mean Length of Stay (d)|
|Sealant||No Sealant||P Value||Sealant||No Sealant||P Value|
|Allen et al, 2004||Prospective (148)||PEG||6.8 b||6.2||.679||6 a||7||.028|
|Anegg et al, 2007||Prospective (152)||Homologous fibrin||5.1 b||6.3||.022||6.2||7.7||.01|
|Rena et al, 2009||Prospective (50)||Homologous fibrin||3.5 b||5.9||.002||5.9||7.5||.01|
|Dango et al, 2010||Retrospective (40)||PEG||2.1 a||3.9||.030||9.9||11.7||.178|
|Gonfiotti et al, 2011||Prospective (185)||Homologous fibrin||5.96 b||6.54||.992||7.64||7.58||.712|
|Tan et al, 2011||Prospective (119)||PEG||4 a||3||NR||7 a||6||NR|
|Lequaglie et al, 2012||Prospective (222)||PEG||—||—||—||4.3||8.4||.0001|
|Filosso et al, 2013||Prospective (24)||Homologous fibrin||6.1 b||10.8||<.001||6.9||9.5||<.001|
|Gologorsky et al, 2019||Retrospective (176)||PEG||1 a||1||.721||—||—||—|