Prognostic Factors For Patients With Postoperative Metastasis From Non-Small-Cell Lung Cancer

Background: We sought to analyze prognostic factors for non-small-cell lung cancer (NSCLC) patients with postoperative metastasis to different organs. Methods: Three hundred and ninety-eight NSCLC patients who underwent complete resection and had recurrent cancer were included. Chi-square tests and Kaplan-Meier survival (log-rank test) analysis were used to investigate differences in survival rate by demographic factors, clinical characteristics, tumor characteristics, and surgical approaches. Cox proportional hazards models were used to evaluate the associations between these variables and overall survival time and between different sites of metastasis and survival time. Results: Cox proportional hazards regression models showed that liver metastasis predicted poor survival. Moreover, the following associations were found between metastatic site and initial surgical approach: brain-right upper lobectomy predicted poor survival times (P = 0.033); bone-right pneumonectomy predicted improved survival times (P = 0.042); lung-left pneumonectomy predicted increased survival time (P = 0.021); and polymetastasis-lobectomy and pulmonary resection with conduit reconstruction predicted better survival time (P = 0.014 and 0.017, respectively). Conclusions: Initial metastatic organ and initial surgical approaches in patients with postoperative metastasis were associated with survival time.