EVALUATION OF PULMONARY-LESIONS WITH FDG-PET - COMPARISON OF FINDINGSIN PATIENTS WITH AND WITHOUT A HISTORY OF PRIOR MALIGNANCY

Study objective: The purpose of this study was to evaluate the accurac y of positron emission tomography (PET) using F-18-fluorodeoxyglucose (FDG) in differentiating benign from malignant pulmonary lesions both in patients with and without a history of prior malignancy. Design: Fo rty-eight consecutive patients with pulmonary lesions suspicious for m alignancy underwent FDG-PET scanning. Group 1 included 27 patients wit hout and group 2 included 21 patients with a history of malignancy. Pa thologic proof of diagnosis was obtained for 32 patients and 16 patien ts were followed up clinically and radiographically for at least 6 mon ths. The standard uptake ratio (SUR) and the lesion to background (L/B ) ratio were determined in 45 patients. Setting: Vanderbilt University Medical Center. Results: In group 1, the average SUR and L/B ratio fo r malignant lesions (n = 14) were 8.9 +/- 4.9 and 20.6 +/- 14.2, respe ctively, For benign lesions (n = 12), the average SUR was 3.3 +/- 3.2 and L/B ratio was 5.2 +/- 5.5. In group 2, the average SUR and L/B rat io for malignant lesions were not significantly different from group 1 . Using either a SUR greater than 2.5 or L/B ratio greater than 5 as a n cutoff level to differentiate benign and malignant lesions, the sens itivity and negative predictive value in both groups were 100%, There were five false-positive studies in group 1 and one in group 2, includ ing tuberculosis (n = 2), a granulomatous lesion (n = 1), an inflammat ory lesion (n = 1), a schwannoma (n = 1), and a fibrous mesothelioma ( n = 1). The overall accuracy was 88%, 81% in group 1, and 95% in group 2. Conclusion: FDG-PET can identify malignant pulmonary lesions both in patients without and with a history of prior malignancy with a high sensitivity and negative predictive value for lesions greater than 1 cm (100% in this study), High FDG uptake by some inflammatory processe s and benign tumors may cause false-positive results. Semiquantitative evaluation using SUR or L/B ratio provides similar accuracy.