Colposcopic scoring system for biopsy decisions in different patient groups.

Colposcopic scoring system have been used for distinguishing low-grade from high-grade cervical lesions. However, none of the previous studies have reported colposcopic scoring systems for biopsy decisions in different patient groups. The purpose of our study was to evaluate the safety of biopsy decisions using the colposcopic score elaborated by Stellato and Paavonen (IL) in 21 nonpregnant HIV infected patients (NP+) and 36 uninfected patients (NP-), 12 HIV infected pregnant patients (P+) and 20 uninfected pregnant patients (P-) in the diagnosis of cervical intraepithelial neoplasias (CIN) and HPV infection. The receiving operator curve was used for the establishment of a cut-off point in the scoring system graduation. The chi-square test was used for the statistical analysis. We obtained a safety cut-off value in the colposcopic scoring system for each patient group: 4.5 for NP+; 4.0 for NP- and 3.5 for P+ and P- patients. The sensitivity and specificity of the colposcopic score for the detection of high-grade lesions for each group were respectively: 87.5 and 92.3% for NP+ patients; 90.9 and 92% for NP- patients; 100 and 87.5% for P+ patients and 100 and 91.7% for P- patients. Our results suggest that the colposcopic scoring system is a practical tool for a colposcopy-guided punch biopsy decision and detection of high-grade cervical lesions in different patient groups. Further studies are needed to prove its clinical utility.