Correlation of Swede score colposcopy scoring system and histopathological results in patients with high-risk HPV infection other than HPV16 and 18.

ABSTRACT

OBJECTIVE: Triage with HPV genotyping has some caveats and debates for HPV positive cases other than 16 and 18. The Swede score colposcopic scoring system has not previously been evaluated in this group of patients. OBJECTIVE: To use the Swede score colposcopic scoring system to compare scores and final histopathological results in women who have undergone colposcopy owing to infection with high risk-HPVs other than HPV16 and 18 and to establish new cut-off values to predict pre-malignant lesions in this group of patients. METHODS: This study was conducted in 613 women undergoing colposcopic evaluation because of abnormal cervical cytology together with high-risk HPV infection. All patients referred were evaluated by an expert colposcopist, given a Swede score (using the Swede score colposcopic scoring system) by using five variables (acetowhiteness, margins plus surface, vessel pattern, lesion size, and iodine staining), and had at least one biopsy procedure (either colposcopically directed or by a loop electrical excision procedure). Sensitivity, specificity, positive predictive value, negative predictive value, likelihood ratio values, and receiver operating characteristic curves for each clinico-pathological variable to detect low-grade and high-grade squamous intra-epithelial lesions, and any squamous cell abnormality (low-grade + high-grade squamous intra-epithelial lesions) were evaluated individually. RESULTS: Final histopathological results of the patients were normal in 53.2% of cases, low-grade lesions in 32.5% of cases, and high-grade lesions in 14.4% of cases. Swede score was ≥8 (median 7.97) for high-grade lesions and ≥5 (median 5.06) for low-grade lesions. The area under the curve values (95% CI) of Swede scores for low-grade and high-grade squamous intra-epithelial lesions, and low-grade + high grade lesions were 0.92, 0.98, and 0.96, respectively. A Swede score cut-off value ≥6 had a sensitivity, specificity, positive predictive value, negative predictive value, and likelihood ratios of 92%, 98%, 93%, 98%, and 50 (22.6 to 110.8), respectively, for high-grade lesions at the final pathology (P<0.001). One high-risk HPV type (except 16 and 18) was no better than another for calculating the median Swede score during colposcopy (P=0.43). CONCLUSIONS: The Swede score colposcopic scoring system appears to be a useful tool for evaluating atypical cervical cytology in women with high-risk HPV infection other than HPV types 16 and 18.