Early stage lung cancer is routinely treated by lobectomy whenever clinically feasible, whereas the role of segmentectomy is controversial. The purpose of this study was to investigate the benefits of segmentectomy vs lobectomy for early stage lung cancer through a meta-analysis of published data. Eligible studies were identified from MEDLINE through February 2013. The manual selection of relevant studies was based on the summary analysis. We used published hazard ratios (HRs) if available or estimates from the published survival data. Lobectomy was chosen as the reference in all HR calculations. We compared the effect of segmentectomy and lobectomy for Stage I, Stage IA, Stage IA with tumours larger than 2 cm but smaller than 3 cm in size and Stage IA with tumours of 2 cm or smaller in 22 observational studies. The HRs of overall and cancer-specific survival indicated significant benefits of lobectomy for Stage I, Stage IA and Stage IA with tumours larger than 2 cm but smaller than 3 cm at 1.20 (95% confidence interval [CI] 1.04-1.38; P = 0.011), 1.24 (95% CI 1.08-1.42; P = 0.002) and 1.41 (95% CI 1.14-1.71; P = 0.001), respectively. For tumours 2 cm or smaller, segmentectomy provided an effect equivalent to that of lobectomy (HR 1.05; 95% CI 0.89-1.24; P = 0.550). No significant publication bias was detected in any part of the analysis. These findings should be interpreted in the context of the inherent limitations of meta-analyses of retrospective studies, including the heterogeneity of patient characteristics.