Preoperative evaluation and triage of women with suspicious adnexal masses using risk of malignancy index

Aims:  To test the accuracy of risk of malignancy index (RMI) in preoperative prediction of malignancy and treatment of adnexal masses. Methods:  A total of 302 women with ultrasound diagnosed adnexal masses, and serum measurement of cancer-associated antigen CA-125 levels, were studied. They all had surgical exploration between October 2001 and September 2005 at the Friedrich-Alexander University Women's Hospital, Erlangen, Germany. The RMI was based on menopausal status, ultrasound morphology of adnexal masses and absolute level of serum CA-125. A cut-off of 250 was chosen as the threshold for determining the type of surgical operations (laparotomy versus laparoscopy) and the skill of the surgeons (gynecological oncologist versus general gynecologist). The data obtained were analyzed for baseline characteristics using χ2 test and analysis of variance (ANOVA). P < 0.05 were statistically significant. The various testing methods were evaluated for sensitivity, specificity, positive and negative predictive values. Results:  The best individual performance was found in RMI at a cut-off of 250 with a sensitivity of 88.2%, specificity of 74.3%, positive predictive value of 71.3% and negative predictive value of 90%. When RMI was used to triage patient treatment, 81.5% of patients who had laparoscopy had histological diagnosis of benign ovarian tumor and 7.5% had malignant tumor. In contrast, 74.4% of patients who had laparotomy had histological diagnosis of malignant ovarian tumor and 16% had benign tumor. Conclusion:  Risk of malignant index is a reliable, cheap, readily available and cost-effective method of preoperative discrimination of benign from malignant adnexal masses. It is also helpful in triaging patients to different treatment groups.