Mediastinal lymph node staging by FDG-PET in patients with non-small cell lung cancer: Analysis of false-positive FDG-PET findings

Background: Accurate staging of mediastinal and hilar lymph nodes is a critical factor determining operability in patients with non-small cell lung cancer (NSCLC). Positron emission tomography with 2-[F-18]fluoro-2-deoxy-D-glucose as a tracer (FDG-PET) has recently been reported to be more effective in detecting tumor involvement in mediastinal and hilar lymph nodes than computed tomography (CT). Objective: In this study, we analyzed the accuracy of FDG-PET in mediastinal and hilar lymph node staging in patients with NSCLC and the factors associated with false-positive or false-negative FDG-PET findings in mediastinal and hilar lymph node staging. Methods: Fifty-four patients with NSCLC who underwent preoperative analysis including chest CT and whole-body FDG-PET were evaluated retrospectively. Using FDG-PET, lesions were considered to be positive if a definite, localized area of higher uptake, excluding physiologic uptake, than in surrounding normal tissue was present. On CT findings, lymph nodes were considered to be positive if they were 110 mm in short-axis diameter, except subcarinal lymph nodes (#7), which were considered to be positive if they were >15 mm in short-axis diameter. All patients underwent surgical resection of primary tumors and mediastinal and hilar lymph nodes between 1999 and 2001 in our institute. Resected lymph nodes were histologically examined for the existence of tumor cells. Results: A total of 306 lymph nodes were resected and used for analysis. The sensitivity, specificity, positive predictive value and negative predictive value of FDG-PET were 73, 98, 70 and 98%, while those of CT were 55, 96, 55 and 96%, respectively. When preoperative nodal staging was compared with post-operative histopathological staging, 44 patients (81%) were correctly staged, 7 (13%) were overstaged and 3 (6%) were understaged by FDG-PET, while 39 patients (72%) were correctly staged, 8 (15%) were overstaged and 7 (13%) were understaged by CT. All 7 overstaged patients by FDG-PET had other pulmonary complications, including interstitial pneumonitis (n = 2), previous pulmonary tuberculosis (n = 3), silicosis (n = 1) and emphysema ( n = 1), although they were not in the active stage. In 3 understaged patients by FDG-PET, lymph nodes were also undetectable by CT. Conclusion: FDG-PET is superior to CT in mediastinal and hilar lymph node staging of patients with NSCLC. However, care should be taken in lymph node staging for patients who have other pulmonary complications, including interstitial pneumonitis, previous pulmonary tuberculosis and silicosis. Copyright (C) 2003 S. Karger AG, Basel.