[Prognostic effect of mediastinal lymph node dissection in patients with stage I non-small cell lung cancer].

OBJECTIVE: To investigate the prognostic effect of mediastinal lymph node dissection in patients with stage I non-small cell lung cancer (NSCLC). METHODS: The clinical data of 330 patients with stage I NSCLC who were treated with curative resection from January 1994 to December 2003 were reviewed retrospectively. According to the extent of mediastinal lymph node dissection and the pathology report, the patients were assigned to lung resection combined with mediastinal lymph node dissection (LND) group or with lymph node sampling (LNS) group. The Kaplan-Meier method was used for survival analysis. COX proportional hazards model was used for multivariate analysis. RESULTS: There were 233 (70.6%) male patients and 97 (29.4%) female patients. The median age was 60 years old. Ninety-eight patients were in stage IA and 233 in stage IB. One hundred and forty patents were in group LND and 190 in group LNS. The mean number of removed lymph nodes in group LND and group LNS were (13.3 +/- 4.7) and (5.2 +/- 3.0) (P < 0.01), respectively. The mean of mediastinal lymph node station sampled in group LND and group LNS were (3.7 +/- 0.9) and (1.3 +/- 1.1) (P < 0.01), respectively. The 5-year and 10-year survival rates of patients in group LND were 72.0% and 66.1%, while in group LNS were 65.9% and 43.0% (P < 0.05), respectively. Other prognostic factors included symptom, staging, visceral pleura invasion and tumor size. LND was disclosed as a favourable prognostic factor at COX multivariate analysis, together with absence of symptom at diagnosis. CONCLUSION: As compared with LNS, LND can improve survival in stage I NSCLC.