Purpose or Case Report: Pediatric ovarian malignancies account for 3–8% of childhood ovarian masses. Accurate preoperative risk stratification is vital to balance fertility preservation against timely oncologic treatment. While the American College of Radiology Ovarian-Adnexal Reporting and Data System (ACR O-RADS) is validated in adults, there are few studies investigating its diagnostic accuracy in the pediatric population. This study assessed the diagnostic accuracy and inter-observer agreement of ultrasound O-RADS in pediatric ovarian and adnexal masses.
Methods & Materials: We retrospectively reviewed consecutive pediatric patients who underwent surgical resection of ovarian or adnexal masses from May 2015 to October 2022 at our country’s largest tertiary pediatric hospital. Their pelvic ultrasound images were reviewed by a senior consultant radiologist (>30 years’ experience in pediatric/women’s imaging) and a senior radiology resident (4 years’ experience), who independently assigned O-RADS categories using the ACR O-RADS mobile application. The reference standard was histology. Diagnostic performance was assessed with standard metrics, and inter-observer agreement with Cohen’s κ.
Results: We evaluated 119 ovarian/adnexal lesions in 114 patients (mean age 12.8 years, range 0.35–18.8 years). Histology showed 2 malignant (1.7%), 6 borderline (5.0%), and 111 benign (93.3%) lesions. One patient with benign histology was scored O-RADS 0. Using O-RADS ≥4 as threshold for malignancy, sensitivity was 100%, specificity 77.5%, PPV 24.2% (positive likelihood ratio 4.44), NPV 100% (negative likelihood ratio 0), with overall diagnostic accuracy of 79% (AUROC=0.601). Inter-observer agreement between both radiologists was moderate (κ=0.61, p<0.001), with most discrepancies between O-RADS 3 and 4.
Conclusions: Preliminary application of the adult US O-RADS system in our local cohort of pediatric ovarian and adnexal masses demonstrates acceptable diagnostic accuracy and substantial inter-observer agreement. As such, O-RADS may be valuable in risk stratification in children, potentially reducing unnecessary radical surgery while facilitating appropriate oncologic referral. Larger multicenter prospective studies are warranted to validate these results and refine O-RADS for pediatric use.