Diagnostic value of the 2011 International Federation for Cervical Pathology and Colposcopy Terminology in predicting cervical lesions

Objective To evaluate the diagnostic accuracy of the 2011 International Federation for Cervical Pathology and Colposcopy (IFCPC) colposcopic terminology. Methods The clinicopathological data of 2262 patients who underwent colposcopy from September 2012 to September 2016 were reviewed. The colposcopic findings, colposcopic impression, and cervical histopathology of the patients were analyzed. Correlations between variables were evaluated using cervical histopathology as the gold standard. Results Colposcopic diagnosis matched biopsy histopathology in 1482 patients (65.5%), and the weighted kappa strength of agreement was 0.480 (P<0.01). Colposcopic diagnoses more often underestimated (22.1%) than overestimated (12.3%) cervical pathology. There was no significant difference between the colposcopic diagnosis and cervical pathology agreement among the various grades of lesions (P=0.282). The sensitivity, specificity for detecting high-grade lesions/carcinoma was 71.6% and 98.0%, respectively. Multivariate analysis showed that major changes were independent factors in predicting high-grade lesion/carcinoma, whereas transformation zone, lesion size, and non-stained were not statistically related to high-grade lesion/carcinoma. Conclusions The 2011 IFCPC terminology can improve the diagnostic accuracy for all lesion severities. The categorization of major changes and minor changes is appropriate. However, colposcopic diagnosis remains unsatisfactory. Poor reproducibility of type 2 transformation zone and the significance of leukoplakia require further study.