Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Kueian Chen
Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Yu-Ting Huang
Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Ren-Chin Wu
Department of Obstetrics and Gynecology and Gynecologic, Cancer Research Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Huei-Jean Huang
Department of Obstetrics and Gynecology and Gynecologic, Cancer Research Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Hung-Hsueh Chou
Department of Obstetrics and Gynecology and Gynecologic, Cancer Research Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Angel Chao
Department of Obstetrics and Gynecology and Gynecologic, Cancer Research Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Chyong-Huey Lai
Department of Obstetrics and Gynecology and Gynecologic, Cancer Research Center, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
Gigin Lin
Department of Medical Imaging and Intervention, Linkou Chang Gung Memorial Hospital, Chang Gung University, Taoyuan 33382, Taiwan
We aim to assess the additional value of diffusion-weighted imaging (DWI) and magnetic resonance spectroscopy (MRS) for the risk stratification of sonographically indeterminate ovarian neoplasms. A total of 21 patients with diagnosed adnexal masses between 2014 and 2017 were divided into malignant (four serous cystadenocarcinomas, four endometrioid carcinomas, three clear cell carcinomas, and one carcinosarcoma) and benign (four cystadenomas, two teratomas, one fibroma, one endometrioma, and one corpus luteal cyst) groups. An apparent diffusion coefficient (ADC) value of 1.27 × 10−3 mm2/s was considered as the optimal threshold in distinguishing malignant from benign ovarian tumors (sensitivity and specificity: 100% and 77.8%, respectively). Choline peaks were detected in six of seven O-RADS (Ovarian-Adnexal Imaging-Reporting Data System) 4 lesions and corrected all of the DWI false-negative clear cell carcinoma. Based on the presence of the choline peaks, the diagnostic performance of MRS showed a sensitivity of 77.8%, a specificity of 100%, and an accuracy of 85.7%, respectively. In conclusion, MRS could potentially play a complementary role for DWI in tumor characterization, particularly for O-RADS 4 tumors or clear cell carcinomas.
