Objective: To evaluate whether external suction is more advantageous than water seal in patients undergoing selective pulmonary resection (SPR) for lung neoplasm.
Summary of background data: Whether external suction should be routinely applied in postoperative chest drainage is still unclear, particularly for lung neoplasm patients. To most surgeons, the decision is based on their clinical experience.
Methods: Randomized control trials were selected. The participants were patients undergoing SPR with lung neoplasm. Lung volume reduction surgery and pneumothorax were excluded. Suction versus non-suction for the intervention. The primary outcome was the incidence of persistent air leak (PAL). The definition of PAL was air leak for more than 3-7 days. The secondary outcomes included air leak duration, time of drainage, postoperative hospital stay and the incidence of postoperative pneumothorax. Studies were identified from literature collections through screening. Bias was analyzed and meta-analysis was used.
Results: From the 1824 potentially relevant trials, 6 randomized control trials involving 676 patients were included. There was no difference between external suction and water seal in decreasing the incidence of PAL [95% confidence interval (CI) 0.81-2.16; z = 1.10; P = 0.27]. Regarding secondary outcomes, there were no differences in time of drainage (95% CI-0.36-1.56, P = 0.22), postoperative hospital stay (95% CI -.31-.54, P = 0.87) or incidence of postoperative pneumothorax (95% CI 0.18-.02, P = 0.05) between external suction and water seal.
Conclusions: For participants, no differences are identified in terms of PAL incidence, drainage time, length of postoperative hospital stay or incidence of postoperative pneumothorax between external suction and water seal. The bias analysis should be emphasized. To the limitations of the bias and methodological differences among the included studies, we have no recommendation on whether external suction should be routinely applied after lung neoplasm SPR. More high-quality randomized controlled trials are needed.
Systematic review registration: None.