OBJECTIVE: To define the risk of high-grade cervical intraepithelial neoplasia (CIN 2,3) among women with unsatisfactory colposcopy results who underwent a loop electrosurgical excision procedure. METHODS: Loop electrosurgical excision procedures were performed for clinical indications by residents supervised by attending obstetrician-gynecologists at an urban public hospital referral clinic. Specimens obtained between July 1, 1996, and April 30, 2002, were retrieved retrospectively after grading and recording in an institutional database. The endpoint of interest was high-grade cervical disease, a composite of CIN 2, CIN 3, and cancer, in excision specimens. RESULTS: Of 169 evaluable patients, five (3%) had cancer. High-grade disease was found in 6 of 21 patients (29%) without a colposcopic lesion, in 13 of 33 patients (36%) with only koilocytosis on colposcopic biopsy, in 15 of 55 patients (27%) with CIN 1, in 13 of 25 patients (54%) with CIN 2, and in 26 of 35 patients (74%) with CIN 3 (p < 0.001). High-grade disease was associated with the grade of referral cytologic results, cytologic analysis repeated at colposcopy, and colposcopic biopsy (p < 0.001 for all). Limiting excision to women with cytologic results at the time of colposcopy read as atypical squamous cells of undetermined significance or worse yielded a high-grade disease prevalence of 12%, with a sensitivity of 92%, specificity 46%, negative predictive value 88%, and positive predictive value 56%. Referral cytologic results, colposcopic biopsy, age, and endocervical curettage results did not seem to identify women at low risk for high-grade disease. CONCLUSIONS: Women with negative cytologic results at the time of colposcopy have a low risk for high-grade disease and may avoid a loop electrosurgical excision procedure despite unsatisfactory colposcopy.
OBJECTIVE: To define the risk of high-grade cervical intraepithelial neoplasia (CIN 2,3) among women with unsatisfactory colposcopy results who underwent a loop electrosurgical excision procedure. METHODS: Loop electrosurgical excision procedures were performed for clinical indications by residents supervised by attending obstetrician-gynecologists at an urban public hospital referral clinic. Specimens obtained between July 1, 1996, and April 30, 2002, were retrieved retrospectively after grading and recording in an institutional database. The endpoint of interest was high-grade cervical disease, a composite of CIN 2, CIN 3, and cancer, in excision specimens. RESULTS: Of 169 evaluable patients, five (3%) had cancer. High-grade disease was found in 6 of 21 patients (29%) without a colposcopic lesion, in 13 of 33 patients (36%) with only koilocytosis on colposcopic biopsy, in 15 of 55 patients (27%) with CIN 1, in 13 of 25 patients (54%) with CIN 2, and in 26 of 35 patients (74%) with CIN 3 (p < 0.001). High-grade disease was associated with the grade of referral cytologic results, cytologic analysis repeated at colposcopy, and colposcopic biopsy (p < 0.001 for all). Limiting excision to women with cytologic results at the time of colposcopy read as atypical squamous cells of undetermined significance or worse yielded a high-grade disease prevalence of 12%, with a sensitivity of 92%, specificity 46%, negative predictive value 88%, and positive predictive value 56%. Referral cytologic results, colposcopic biopsy, age, and endocervical curettage results did not seem to identify women at low risk for high-grade disease. CONCLUSIONS:Women with negative cytologic results at the time of colposcopy have a low risk for high-grade disease and may avoid a loop electrosurgical excision procedure despite unsatisfactory colposcopy.