Colposcopic analysis of cervical lesions using popular scoring systems: Reid versus Swede.

Abstract

To assess the effectiveness of colposcopy in detecting cervical lesions and to grade them according to Reid score and Swede score, and compare it with histopathology results. This study was conducted on 130 patients in a tertiary care centre, who were subjected to pap smear and colposcopy. The cervical lesions were graded according to Reid score and Swede score, and a biopsy was obtained from the lesion. Histopathology results were correlated with colposcopy findings, and the scores were compared. The colposcopic findings using Reid score and Swede score correlated with histopathology results in the study population. The association between colposcopic impression and histopathology result was highly significant (p < 0.001), using both Reid score and Swede score. The sensitivity, specificity, PPV and NPV of Reid score (overall) was 86.2%, 80.20%, 55.56% and 95.3%, respectively. The diagnostic accuracy was 81.54%. At score > 5, specificity increased to 99% and diagnostic accuracy was 92.31%. The overall sensitivity, specificity, PPV, NPV and diagnostic accuracy of Swede score was 89.7%, 49.5%, 33.8%, 94.3% and 48.46%, respectively. As the cut off value increased, the sensitivity decreased. But the specificity, PPV, NPV and diagnostic accuracy increased and was statistically significant. The specificity and PPV was 100% at score > 8. As the cut off value increased, the diagnostic accuracy of both the Scores increased, and was more accurate in detecting high-grade lesions.