OBJECTIVE: Many patients with non-small cell lung cancer have positive mediastinal lymph nodes on preoperative positron emission tomography (PET) but do not have mediastinal involvement after surgery. The prognostic significance of this discordance was assessed. METHODS: This Institutional Review Board-approved study evaluated patients treated with upfront surgery at Duke Cancer Institute (Durham, NC) for non-small cell lung cancer from 1995 to 2008. Those staged with PET with pN0-1 disease after negative invasive mediastinal assessment were included. Mediastinal lymph nodes were scored as positive or negative based on visual analysis of the preoperative PET. Clinical outcomes of the PET-positive and PET-negative cohorts were estimated using the Kaplan-Meier method and compared using a log-rank test. Prognostic factors were assessed using a multivariate analysis. RESULTS: A total of 547 patients were assessed, of whom 105 (19%) were PET positive in the mediastinum. The median number of mediastinal lymph node stations sampled was 4 (range, 1-9). The 5-year risk of local recurrence was 26% in PET-positive versus 21% in PET-negative patients (P = .50). Patterns of local failure were similar between the 2 groups. Distant recurrence (35% vs 29%; P = .63) and overall survival (44% vs 54%; P = .52) were comparable for PET-positive and PET-negative patients. On multivariate analysis, a positive PET was not significant for local recurrence (hazard ratio [HR], 1; P = 1), distant recurrence (HR, 0.82; P = .42), or overall survival (HR, 1.08; P = .62). CONCLUSIONS: Patients with positive mediastinal lymph nodes on preoperative PET, but negative on histologic analysis, are not at increased risk of disease recurrence. Pathologic staging remains the standard.