The aim of this study was to assess the value of fluorodeoxyglucose positron emission tomography (FDG PET) imaging of small pulmonary nodules incidentally detected by spiral computed tomography (CT) in a high-risk population. Ten patients (five females, five males, aged 54-72 years) were recruited from an ongoing 4-year placebo controlled intervention study of the effect of inhaled steroids in 300 smokers with moderate to severe chronic obstructive pulmonary disease. The participants received yearly CT scans of the chest. Patients with a negative chest radiograph at the time of inclusion, but with pulmonary nodules indeterminate for malignancy detected by conventional spiral CT on a subsequent scan, were referred for FDG PET. Histological diagnoses were sought for all nodules with FDG uptake or where CT showed that they had grown. Ten patients had pulmonary nodules indeterminate for malignancy (approx. 3.3% of the entire study population). The prevalence of malignancy in this group was 50%. The accuracy of PET was high, in spite of the fact that seven patients had nodules smaller than 15 mm and two patients had bronchoalveolar cell carcinoma. This small prospective study indicates that subsequent assessment with FDG PET of small pulmonary nodules incidentally detected by CT has the potential to minimize the numbers of invasive procedures performed in individuals with a benign pulmonary lesion. FDG PET also increases the possibility of an early diagnosis as compared to the strategy of watchful waiting.