Indeterminate pleural metastasis on contrast-enhanced chest CT in non-small cell lung cancer: improved differential diagnosis with 18F-FDG PET/CT

Abstract

To evaluate the diagnostic performance of (18)F-fluorodeoxyglucose (FDG) PET/CT (PET/CT) for determining the presence of pleural metastasis in patients with indeterminate findings on a contrast-enhanced chest CT (CECT) for non-small cell lung cancer (NSCLC). This is a retrospective study. NSCLC patients (n=63) who underwent thoracentesis and/or pleural biopsy were enrolled. CECT and PET/CT reports of pleural metastasis were analyzed based on comparison with cytological or histological confirmation. Negative cytologic results were re-confirmed with follow-up study prior to cancer-related therapy. CECT results were classified into 3 categories: negative, indeterminate, and positive for pleural metastasis. PET/CT results were classified into 2 categories (negative and positive for pleural metastasis) based on FDG uptake visual grading. The level of max SUV of pleura was also analyzed. ROC analysis was done for establishing the max SUV cut-off value. PET/CT could differentiate pleural metastasis with 70.8% diagnostic accuracy when the CECT finding was indeterminate (n=24). Optimal cut-off value to predict pleural metastasis was 2.8 for max SUV. Diagnosis by max SUV 2.8 had lower sensitivity (86.3 vs. 92.2%), but higher specificity (66.7 vs. 58.3%) than PET/CT by FDG visual grading criteria. PET/CT showed better diagnostic performance than CECT for detecting pleural metastasis in NSCLC patients. When the finding of CECT is controversial, PET/CT can differentiate the metastatic pleural lesion. Both FDG uptake visual grading and max SUG cut-off value can be used as diagnostic criteria for pleural metastasis.