Performance of two colposcopic indices for predicting premalignant cervical lesions

Background: Colposcopy is an essential tool for diagnosing premalignant cervical lesions in women. Colposcopic scoring systems, such as the Reid’s colposcopic index (RCI) and Swede score, aim to improve diagnostic accuracy and reduce interobserver variability. This study compared the diagnostic performances of these two indices in predicting high-grade cervical intraepithelial neoplasia (CIN2+). Methods: A cross-sectional study of 300 women aged ≥18 years with abnormal cervical screening results was performed at a tertiary care centre in Dahak, Bangladesh. All patients underwent colposcopic examination using both RCI and Swede scores, followed by biopsy, irrespective of colposcopic findings. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated considering histopathology as the gold standard. The agreement between the two scores was also examined. Results: At a cutoff of 5, RCI showed a sensitivity of 37.0% and specificity of 94.5% (PPV, 40.1%; and NPV, 93.8%). For the Swede score, a cutoff of 5 yielded a sensitivity of 74.1% and specificity of 45.0% (PPV, 11.8%; and NPV, 94.6%), whereas a cutoff of 8 reduced sensitivity (11.1%) but increased specificity (92.3%). The RCI and Swede scores had a moderate agreement (κ=0.4). Conclusion: Although RCI offers high specificity, its low sensitivity limits its screening utility. The Swede score is a flexible tool for screening at cutoff 5 and for “see and treat” management at cutoff 8.