A majority of observational studies on overall survival following thoracoscopic vs open lobectomy for early-stage non-small cell lung cancer did not demonstrate a significant difference, whereas several meta-analyses on this topic showed a significant difference. The PubMed, Scopus, and Web of Science databases were queried for studies published in the English language. We searched for meta-analyses and original studies comparing overall survival between thoracoscopic and open lobectomy for early-stage non-small cell lung cancer. Our meta-analysis, using random effect models and with a hazard ratio as a measure of effect, was performed on original studies. Publication bias was evaluated with funnel plots of precision and the Egger test. Seven meta-analyses on this topic were found and all of them have shown that thoracoscopic lobectomy is associated with significantly more favorable overall survival than open lobectomy, using odds ratio, risk ratio, or risk difference as measures of effect. Our meta-analysis of 11 observational studies demonstrated no significant difference in overall survival between thoracoscopic (n = 2386) and open lobectomy (n = 3494) for early-stage non-small cell lung cancer (pooled hazard ratio: 0.91, 95% confidence interval: 0.76-1.09, P = 0.30). Neither funnel plots of precision nor the Egger test suggested a publication bias. Our meta-analysis, using a hazard ratio as a measure of effect for a time-to-event outcome, did not demonstrate a significant difference in overall survival between thoracoscopic and open lobectomy with the current dataset available in the literature, as opposed to previous meta-analyses.
Keywords: lung cancer; open lobectomy; survival; thoracoscopy.
Copyright © 2017 Elsevier Inc. All rights reserved.