[Extent of lymphadenectomy in stage I-IIIA non-small cell lung cancer: a randomized clinical trial]

To study the role of radical systematic mediastinal lymphadenectomy for non-small cell lung cancer (NSCLC).

All 504 operable eligible cases with NSCLC were randomly divided to a radical lymphadenectomy (RL) group and a conventional lymph node dissection group (control) treated between Aug. 1989 and Dec. 1995. For patients postoperatively eligible, thirteen parameters (operation type, pathological type and grade, tumor size, total number of dissected lymph nodes, number of metastatic lymph nodes, metastasis ratio of lymph nodes, postoperative TNM staging, adjuvant therapy, recurrence or metastasis, morbidity, survival and life quality) were evaluated. The end point of follow-up was Dec. 31, 1998. Lost follow-up rate was 1.9%. The results were analyzed with soft were SPSS7.5. The cumulative survival was calculated by the Kaplan-Meier method and compared by the log rank test. The prognostic factors were analyzed by the Cox model.

There were 320 cases, 160 cases in each group, who entered the study. The mean numbers of dissected lymph nodes was 9.49 in the RL group and 3.63 in the control group. For stage I NSCLC patients, the 1,3,5,9-year survival rate was 91.8%, 86.9%, 81.4%, 74.2% respectively in the RL group and 88.7%, 72.5%, 58.5%, 52.1% respectively in the control group (P < 0.014). However, no statistically significant difference in survival rates between RL and control groups of patients with stage II and IIIA NSCLC. The postoperative TNM staging, metastasis ratio of lymph nodes, extent of lymphadenectomy were the factors influencing long term survival upon multivariable analysis.

Classical lobectomy or pneumonectomy with radical systematic mediastinal lymphadenectomy is the surgical treatment of choice for NSCLC.