BACKGROUND: Accurate staging of lung cancer is essential in determining the most appropriate management plan, as detection of occult metastasis can significantly alter management. AIMS: The aims of this study are to determine the prevalence of occult metastasis in patients undergoing 2-(18)F-fluoro-2-deoxy-D-glucose ((18)F-FDG) positron emission tomography (PET) for evaluation of suspected/proven lung carcinoma and correlate pre-PET TNM stage with prevalence of metastasis. METHODS: FDG-PET, which identified patients with metastasis on institutional database, was re-evaluated by a nuclear medicine physician blinded to clinical information. The confidence level of metastasis was scored on a 5-point scale, with a score of >/=4 considered positive. RESULTS: There were 67 of 645 (10%) patients identified with suspected occult metastasis on FDG-PET. Twelve patients scoring =3 were excluded. Prevalence of occult metastasis was 10/156 (6%) in solitary pulmonary nodules (SPN); 22/319 (7%) and 23/170 (14%) in proven and suspected lung cancer, respectively. Positive predictive value of FDG-PET for metastasis was 8/10 (80%) in solitary pulmonary nodules, 14/20 (70%) and 17/21 (81%) in proven and suspected lung cancer, respectively. (18)F-FDG-avid lesions classified as false positives were patients with cholelithiasis, rib fractures and those with equivocal/negative bone scans or computed tomography on follow up. There was a higher incidence of true positive occult metastasis in patients in all stages of disease, particularly stage III disease. CONCLUSION: (18)F-FDG PET is predictive for occult metastatic disease in patients with solitary pulmonary nodules and proven or suspected lung cancer and is more likely to be present in all stages, particularly in stage III. PET findings should be actively pursued with correlative investigation to identify benign pathology in patients who remain candidates for curative treatment.