ObjectiveTo compare effectiveness and safety of video-assisted thoracic surgery (VATS) and thoracotomy in lymph node (LN) dissection for lung cancer. MethodsA comprehensive search of PubMed, Ovid Medline, EMbase, Web of Science, ScienceDirect, the Cochrane Library, Scopus and Google Scholar was performed to identify studies (from January 1990 to July 2015) comparing VATS with thoracotomy in LN dissection. The data were analyzed by RevMan 5.3 software. Quality of literature was evaluated by Newcastle-Ottawa scale or Jadad scale. ResultsFifty-one articles met the inclusion criteria involved 7 127 patients in the VATS group and 9 217 patients in the thoracotomy group. Thirty-eight articles were of good quality and the remaining thirteen were medium. Meta-analysis showed that fewer N1 LN stations in the VATS group (95% CI -0.23 to -0.04, P=0.005), although VATS harvested more left-side LNs (95% CI 0.51 to 3.22, P=0.007). The number of total LNs (95% CI -1.81 to 0.28, P=0.15), total LN stations (95% CI -0.34 to 0.15, P=0.44), N2 LNs (95%CI -1.77 to 0.79, P=0.45), N2 LN stations (95% CI -0.22 to 0.16, P=0.78), N1 LNs (95% CI -0.95 to 0.11, P=0.12), and right-side LNs (95% CI -1.52 to 2.23, P=0.71) harvested in the two groups were not significantly different. ConclusionIn the surgical treatment of lung cancer, VATS can achieve the same efficacy of LN dissection as thoracotomy. This conclusion still needs to be further proved by more high-quality and large-scale RCTs.