PURPOSE: This study aimed to evaluate the diagnostic accuracy of contrast-enhanced FDG PET/CT for unexpected ovarian lesions in patients with nongynecologic cancers. PATIENTS AND METHODS: The present retrospective study was approved by the hospital institutional review board. We enrolled 72 nongynecologic cancer patients (33 colorectal cancer, 20 breast cancer, and 19 gastric cancer) who showed unexpected ovarian lesions on contrast-enhanced FDG PET/CT and subsequently underwent surgical resection of the ovarian lesions. The SUVmax values of the ovarian lesions were compared according to the histopathological results, and the accuracy of PET/CT was evaluated. RESULTS: A total of 103 ovarian lesions were surgically resected from 72 patients (64 ovarian lesions were metastases, 21 were benign, and 18 were malignant primary ovarian cancer). SUVmax among the 3 groups were different (P < 0.0001). Using a cutoff SUVmax of 2.5, the sensitivity, specificity, and accuracy of PET/CT for detecting malignant ovarian lesions were 80.5%, 81.0%, and 80.6%, respectively. Detection rates of PET/CT for colorectal, breast, and gastric cancer metastases, as well as primary ovarian cancers were 91.4%, 83.3%, 47.8%, and 100.0%, respectively. Using contrast-enhanced CT findings with SUVmax, the sensitivity, specificity, and accuracy of contrast-enhanced FDG PET/CT were improved to 95.1%, 76.2%, and 91.3%, respectively (P = 0.003). CONCLUSIONS: Conventional FDG PET/CT showed moderate diagnostic accuracy for the detection of malignant ovarian lesions mainly due to low detection rate in gastric cancer metastases. The PET/CT detection rate of malignant ovarian lesions can be improved using contrast-enhanced CT findings with FDG uptake measurement.
Bibliographical notePublisher Copyright:
© 2015 Wolters Kluwer Health, Inc.
All Science Journal Classification (ASJC) codes
- Radiology Nuclear Medicine and imaging