Objectives The purpose of this study was to investigate the value of PET/CT in the preoperative staging of non-small cell lung cancer in predicting long-term survival and diagnostic performance, validated by histopathology following surgical resection. Methods Between 02/2009 and 08/2011, 255 patients with non-small cell lung cancer were included in this single-center prospective study. All underwent 18F FDG-PET/CT for pre-operative staging, and in 243 patients complete surgical resection was possible. Regarding lymph node involvement and extrathoracic metastases, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using the histopathological staging as reference. Median follow-up for censored patients was 9.1 years. Results Overall 5-year survival rate of all patients was 55.6%, and of patients who had complete surgical resection it was 58.2%. In multivariate analysis of all surgically resected patients lymph node involvement (p=0.029) and age >61 years (p=<0.001) were significant independent prognostic factors. SUVmax and SUVmean cut-offs between SUV 2 and 11, however, were not associated with better or ;worse survival. The PET-CT sensitivity, specificity, positive predictive value and negative predictive value for predicting lymph node involvement were 57, 95, 88, and 76%, respectively. Furthermore, sensitivity, specificity, positive predictive value, and negative predictive value for detecting extrathoracic metastases were 100, 58, 98, and 100%, respectively. Conclusions In this study, tumor 18F FDG-uptake values did not provide additional prognostic information. Age>61 years and lymph node metastasis were associated with worse long-term survival in surgically resected patients. 18F FDG-PET/CT scans allow for improved patient selection. However, in staging mediastinal lymph nodes, there is a high rate of false positives and false negatives, suggesting that tissue biopsy is still indicated in many cases.