Prediction of cervical histologic results using an abbreviated Reid Colposcopic Index during ALTS

Abstract

The purpose of this study was to determine the accuracy of an abbreviated Reid Colposcopic Index (RCI) to detect cervical intraepithelial neoplasia (CIN) 2 and CIN 3 during the National Cancer Institute (NCI) ASCUS LSIL Triage Study (ALTS). Colposcopists from 4 ALTS clinical centers indicated an RCI score after colposcopically examining subjects. Colposcopy quality control reviewers also documented an RCI score for the same subjects after assessing their digitized cervical images. Associations of colposcopist and reviewer RCI scores with histologic results were evaluated. Agreement rates between colposcopists and reviewers were calculated for colposcopic impression. Overall RCI score agreement was poor (weighted Kappa = 0.17). The sensitivity, specificity, positive and negative predictive values of a colposcopist's RCI score of >3/6 to detect CIN 3 or worse were 37.3% (95% CI 32.5%-42.3%), 89.7% (88.6%-90.8%), 30.8% (26.7%-35.2%), and 92.1% (91.1%-93.0%), respectively. The 3 colposcopic signs were not individually sensitive in detecting CIN 3. Colposcopists using the RCI in the ALTS trial failed to detect CIN 2/3 at the levels expected.