Roma Index and Adnex Model: which is more Superior in Predicting Epithelial Ovarian Malignancy?

Abstract

Objective: To compare the accuracy of ROMA index and ADNEX model in predicting the risk of malignancy in ovarian tumour. Methods: This was a prospective analytic study. A total of 37 samples were acquired from women of all ages diagnosed with an ovarian cystic tumour in the Central General Hospital Prof. Dr. R. D. Kandou. A CA-125 marker, HE4 marker, menopausal status and ultrasonography (USG) examination were obtained, and subsequently compared with the fi nal histopathological results. The data were analysed by using the SPSS statistics software. Results: Thirty-seven women participated in this study. The mean age of participants was 43 years old. The Area Under Curve (AUC) of the ADNEX was 0.979 with a sensitivity of 90.0%, specifi city of 88.2%, negative predictive value of 89.8%, and positive predictive value of 80.5%. The AUC of the ROMA model was 0.734 with the sensitivity, specifi city, negative predictive value, and positive predictive value of 65.0%, 64.7%, 64.8%, and 64.8%, respectively. Both models showed AUC values > 0.50 (p-value < 0.05). Conclusions: The IOTA ADNEX had better accuracy than the ROMA model in predicting ovarian epithelial malignancy. The ADNEX model had higher sensitivity and specifi city than the ROMA model.