MRI O-RADS: redefining diagnostic accuracy in characterization of ovarian lesions with histopathology as the...

Adnexal lesions are a common gynecological concern requiring accurate distinction between benign and malignant types for optimal clinical management. The Ovarian-Adnexal Reporting and Data System for magnetic resonance imaging (MRI) (O-RADS MRI) aim to standardize risk assessment and enhance diagnostic precision.

To evaluate the diagnostic accuracy of MRI using O-RADS MRI criteria, with histopathology as the gold standard.

This descriptive cross-sectional study was conducted from January to June 2025 at the Dow Institute of Radiology, Dow University of Health Sciences (DUHS), Karachi, Pakistan. Women aged 18-65 years undergoing contrast-enhanced MRI for ultrasound-detected ovarian lesions, with available histopathology, were included. Patients with prior surgery, chemotherapy, radiotherapy, pregnancy, lactation, other malignancies, or inconclusive histopathology were excluded. O-RADS MRI scores were assigned by an experienced radiologist. Diagnostic metrics including sensitivity, specificity, predictive values, likelihood ratios, overall accuracy, and area under the curve (AUC) were calculated.

Among 145 patients (62.76% premenopausal, 37.24% postmenopausal), O-RADS categories 3 (32.41%) and 4 (31.03%) were most frequent. Benign lesions predominated in premenopausal women (60.44%), while malignant lesions were more common in postmenopausal women (66.67%). Common benign lesions included hemorrhagic cysts (14.48%), endometriomas (9.66%), and serous cystadenomas (6.90%), while serous cystadenocarcinoma (14.48%) was the most frequent malignancy. Using O-RADS ≥4 as the threshold for malignancy, sensitivity was 93.10%, specificity 79.31%, positive predictive value 75.00%, negative predictive value 94.52%, overall accuracy 84.83%, and AUC 0.921. Redefining O-RADS categories 3-5 as positive markedly reduced diagnostic accuracy to 56.42% and sensitivity to 66.67.

The O-RADS MRI scoring system demonstrated excellent diagnostic accuracy in distinguishing benign from malignant ovarian lesions, supporting its role in clinical decision-making and risk stratification.