External validation of the adapted Risk of Malignancy Index incorporating tumor size in the preoperative evaluation of adnexal masses

Abstract

Objective: The Risk of Malignancy Index (RMI) is a simple scoring system to standardize and improve the preoperative evaluation of adnexal masses. Since 1990, three versions of the RMI have been validated in different clinical studies. Recently, a fourth version of the RMI (RMI-4) was introduced that includes tumor size as an additional parameter. The aim of this study was to validate the ability of RMI-4 to discriminate between non-invasive lesions and invasive malignant adnexal masses, and to compare its performance with RMI-3.

Study design: Women scheduled for surgery for an adnexal mass between 2005 and 2009 in 11 hospitals were included. Ultrasonographic characteristics, menopausal status and serum CA 125 level were registered preoperatively, and combined into the RMI. The performances of RMI-3 and RMI-4 were assessed and statistically tested for differences.

Results: A total of 643 patients were included: 469 benign, 73 borderline and 101 malignant tumors. The RMI-3 had a sensitivity of 76%, specificity of 82%, positive and negative predictive values (PPV and NPV) of 45% and 95%, and an accuracy of 81%. The RMI-4 had a sensitivity of 74%, specificity of 79%, PPV of 40%, NPV of 94%, and an accuracy of 78%. The accuracy of RMI-3 was significantly higher than the accuracy of RMI-4 (p=.001). Both models had an area under the curve of 0.86.

Conclusion: Both RMI-3 and RMI-4 were able to discriminate between non-invasive lesions and invasive malignant adnexal masses, with similar performances. Including tumor size in the RMI does not improve its performance.