Adding apparent diffusion coefficient to the ovarian-adnexal reporting and data system MRI: a trade-off between sensitivity and specificity

Background The Ovarian-Adnexal Reporting and Data System Magnetic Resonance Imaging (O-RADS MRI) can assess the behavior of ovarian masses through high-resolution anatomical sequences, diffusion-weighted imaging, and dynamic contrast-enhanced studies using plotted time-intensity curves. Accurate O-RADS MRI categorization requires experience with the O-RADS lexicon and associated descriptors. Objectives To assess the effects of adding the mean apparent diffusion coefficient (ADC) to the O-RADS MRI score for adnexal lesion characterization. Methods Sixty-six patients with seventy-two ovarian lesions, identified on preliminary pelvic ultrasound examination and categorized as O-RADS 3, 4, or 5, underwent dynamic MRI with diffusion-weighted imaging and ADC mapping. The reported results were compared with either the histopathology of the lesions or clinical follow-up. Results O-RADS MRI alone achieved high sensitivity (95.1%) but moderate specificity (61.5%). ADC values improved sensitivity (97.6%) but reduced specificity to 38.5%, creating a diagnostic trade-off between detecting more malignancies and misclassifying benign lesions. Conclusions Apparent diffusion coefficient integration with O-RADS MRI improves sensitivity at the expense of specificity, representing a key clinical trade-off. This study advances prior ADC-O-RADS literature by explicitly quantifying this effect in a real-world cohort and clarifying the limited reliability of ADC in benign and borderline adnexal lesions.