Comparison of diagnostic performance of simple international ovarian tumor analysis rules versus subjective pattern recognition for triage of adnexal masses

Abstract

Background: Accurate and early diagnosis of adnexal masses is essential for optimal clinical decision-making. The aim of the study was to compare the diagnostic performance of simple international ovarian tumor analysis (IOTA) rules vs subjective pattern recognition, to discriminate between benign and malignant adnexal mass, and to establish the diagnostic utility of IOTA rules as a standardized examination tool in early diagnosis of ovarian malignancy. Methods: A prospective cohort study was conducted at a tertiary care hospital between November 2017 and March 2019 on 100 women with adnexal masses. All adnexal masses detected on ultrasound were classified according to IOTA rules by the trainee, followed by subjective pattern recognition by experts. These observations were further correlated with histopathology/intraoperative findings/ follow-up examination. Diagnostic efficacy was assessed by comparing sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy. Results: Among the 100 patients, 81 had benign, and 19 had malignant masses on final diagnosis. The sensitivity, specificity, PPV, NPV, and accuracy for the detection of malignancy using IOTA rules by trainee were 100%, 95.59%, 81.82%, 100%, and 95.65%, and by subjective evaluation by experts were 100%, 97.5%%, 90.5%, 100%, and 98% respectively. No statistically significant difference was found between the diagnostic accuracy of the two methods. Conclusions: Simple IOTA rules are as accurate as subjective evaluation by experts in the characterization of adnexal masses. Their inherent simplicity and reproducibility make them ideal for use by less experienced sonographers.