Association between colposcopic findings and histology in cervical lesions: the significance of the size of the lesion.

The relation between eight specific colposcopic diagnostic findings and the histologic grade of a cervical lesion in 896 women was evaluated. The size of the transformation zone (TZ), the size of the lesion, the intensity of the color tone, distinct margins, the pathology of the vessels, and the presence of micropapillae as single findings were highly statistically correlated to the histologic grade (P < 0.0001). By logistic regression analysis the risk for a higher histologic grade when assessed by colposcopy was greatest in women with variation of the acetowhite color (odds ratio (OR) = 16.0; 95% CI, 10.0-26.0) followed by coarse vessels (OR = 10.0; CI, 3.2-34.0). Lesion-size larger than 50% of the visible cervix had an OR of 3.6 (CI, 2.1-6.3). Extention beyond TZ had an OR of 0.4 (CI, 0.2-0.4) and larger TZ had an OR of 0.5 (CI, 0.3-0.9). In conclusion we found that the size of the cervical lesion had some independent predictive value and should be considered in future trials.

Diagnostic accuracy of cytology and colposcopy in cervical squamous intraepithelial lesions.

OBJECTIVE: To evaluate the diagnostic accuracy of cytology and colposcopy in women with an abnormal cervical smear using histology as the 'gold standard'. DESIGN: Survey of consecutively referred women with abnormal smear. SETTING: The out-patient colposcopical clinic of Herning Hospital, Denmark. PATIENTS: 813 women with a median age of 29.0 years (range 15-71 years) with their first abnormal smear. RESULTS: For detecting cervical high-grade lesions (HGL) the sensitivity of cytology was 41% (36-47%), of colposcopy 67% (62-72%) and in combination 75% (70-80%), so at least 25% of HGL were underestimated. Colposcopy underestimated more CIN-2 than CIN-3 lesions and more small lesions and lesions in smaller transformation zones. Cytology underestimated more CIN-2 lesions but equal numbers of small and large lesions and transformation zones. CONCLUSIONS: Colposcopy was a better tool for diagnosing HGL than cytology, but even in combination too many HGL were missed. All women with abnormal cytology should therefore have colposcopical and histological investigation and prospective studies of the natural history of cervical squamous lesions should include histological evidence.