Value of colposcopy screening independently for cervical lesions in basic hospitals

Objective To evaluate the value of colposcopy screening independently for cervical lesions in basic hospitals. Methods A retrospective cohort study of 574 patients who both underwent colposcopy and had pathologic results in our department. The consistency, authenticity and predictability of colposcopy and pathological results, the relevant factors affecting the accuracy of colposcopy diagnosis, and the analysis of unsatisfactory colposcopy results were analyzed. Results Agreement between colposcopic diagnosis and cervical pathology was 50.57%, and Kappa value of consistency was 0.358 (P<0.01). Agreement within one grade was 91.38 % (Kappa value was 0.871, P<0.01). The sensitivity of colposcopy in diagnosing of low-grade squamous intraepithelial lesion (LSIL) and above was high (96.91%), while the specificity was 52.60% (81/154); False positive rate was 47.40%, and false negative rate was 3.09%. The sensitivity of colposcopy in diagnosing of high-grade squamous intraepithelial lesion (HSIL) and above was 81.82% (153/187), and the specificity in diagnosing of LSIL and below was 90.68% (146/161); False positive rate was 9.3%, and false negative rate was 18.2%. The stage of cervical cancer and skills of colposcopist would be factors that influence the accuracy of colposcopy screening in HSIL and above (P<0.01). Average age in unsatisfactory colposcopy was (52±9)y, which was significantly different from satisfactory colposcopy cases (P<0.01). The proportion of early cervical lesions in patients with unsatisfactory colposcopy HSIL or above was high. Conclusions The complete coincidence rate between colposcopy and pathological diagnosis is limited, but the consistency within one grade is good; the stage of cervical lesions and the experience of examiners are the factors related to the accuracy of colposcopy in diagnosing HSIL and above lesions; Unsatisfactory colposcopy cases can be further combined with cytology and HPV examination for shunt and treatment. Key words: Colposcopy; Uterine cervical diseases; Sensitivity and specificity