Premium
Performance of the abbreviated Reid colposcopic index in prediction of high‐grade lesions
Author(s) -
Boonlikit Sathone
Publication year - 2016
Publication title -
international journal of gynecology and obstetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.895
H-Index - 97
eISSN - 1879-3479
pISSN - 0020-7292
DOI - 10.1016/j.ijgo.2015.10.025
Subject(s) - medicine , cutoff , colposcopy , receiver operating characteristic , confidence interval , cervical intraepithelial neoplasia , predictive value , predictive value of tests , retrospective cohort study , radiology , surgery , cervical cancer , cancer , physics , quantum mechanics
Objective To assess the diagnostic performance of the Reid colposcopic index (RCI) and establish the optimal cutoff value to predict a histology of cervical intraepithelial neoplasia grade 2 or worse (CIN2 +). Methods A retrospective analysis was undertaken of the medical records of women who underwent colposcopy with RCI scoring at a center in Bangkok, Thailand, between 2003 and 2014. Only patients for whom histology reports were available were included. Cases had been scored according to three criteria: margin, color, and vascular pattern. The performance of this three‐criterion score (abbreviated RCI) was assessed for sensitivity, specificity, and positive and negative predictive values at every cutoff level. Receiver operation characteristics (ROC) curve analysis was performed to determine the optimal cutoff value to distinguish between women with CIN2 + and others. Results Among 349 included patients, 158 (45.3%) had CIN2 +. The most appropriate cutoff score was 3, which had a sensitivity of 72.7%, a specificity of 86.9%, and positive and negative predictive values of 82.1% and 79.4%, respectively. The area under ROC curve was 0.857 (95% confidence interval 0.815–0.898). Conclusion The performance of the abbreviated RCI seems satisfactory. The optimal cutoff value was 3.