The prognostic nutritional index (PNI) has been reported to be a prognostic indicator in some malignant tumors. However, its prognostic value in nonsmall cell lung cancer (NSCLC) has not been fully investigated. A retrospective review of 1416 patients with NSCLC who underwent radical surgery between January 2006 and December 2011 was conducted. To obtain optimal cutoff levels of PNI, running log-rank statistics was applied. Survival was calculated by the Kaplan-Meier method. The prognostic significance of PNI, together with various clinicopathological factors, was evaluated by multivariate analysis. The optimal cutoff point for PNI was 52. The 1-, 3-, and 5-yr survival rates in patients with PNI of less than 52 were 80.0%, 61.3%, and 50.4%, respectively, and were significantly more unfavorable than those in patients with PNI 52 or higher (84.7%, 71.5%, and 60.3%, respectively, P < 0.001). Multivariate analysis suggested that gender (P = 0.026), age (P < 0.001), PNI (P = 0.005), differentiation (P = 0.024), pathology T category (P = 0.003), and pathology N category (P < 0.001) were revealed to be independent prognostic factors. Our results indicate that PNI is an independent predictor of survival for patients undergoing radical surgery with NSCLC.