Multicenter pre-operative assessment of pediatric ovarian malignancy

Purpose: The purpose of this study was to develop a pre-operative risk assessment tool for childhood and adolescent ovarian malignancy, in order to guide operative management of pediatric ovarian masses. Methods: We conducted a retrospective analysis of patients <18 years old who underwent ovarian surgery at two quaternary care pediatric centers over 4 years (1/1/13-12/31/16). Probability of malignancy was estimated based on imaging characteristics (simple cyst, heterogeneous, or solid), maximal diameter, and tumor markers (a-fetoprotein, beta-human chorionic gonadotropin). Results: Among 188 children with ovarian masses, 11% had malignancies. For simple cysts, there were no malignancies (0/24, 95% CI = 0-17%). Among solid lesions, 44% (15/34, 95% CI = 28-62%) were malignant. Among marker-elevated heterogeneous masses, 40% (2/5, 95% CI = 12-77%) were malignant. Conversely, small (<= 10 cm) and large (>10 cm) marker-negative heterogeneous lesions had malignancy proportions of 0% (0/39, 95% CI = 0-11%) and 5% (2/40, 95% CI = 1-18%), respectively. Conclusions: Given the malignancy estimates identified from these multi-institutional data, we recommend an attempt at ovarian-sparing resection for simple cysts or tumor marker-negative heterogeneous lesions <= 10 cm. Oophorectomy is recommended for solid masses or heterogeneous lesions with elevated markers. Finally, large (>10 cm) heterogeneous masses with non-elevated markers warrant a careful discussion of ovarian-sparing techniques. Complete surgical staging is mandatory regardless of operative procedure. Type of Study: Study of Diagnostic Test. (C) 2019 Elsevier Inc. All rights reserved.