Endocervical Curettage in Evaluating Abnormal Cervical Cytology

Abstract

Between 1980 and 1995, 2126 patients were referred to the Colposcopy Clinic of Baylor College of Medicine (Houston, TX) for evaluation of an abnormal cervical smear. All patients were studied with a repeated cervical smear, colposcopy, and endocervical curettage (ECC). They were treated for cervical abnormality with cryosurgery (N = 1279), laser vaporization (N = 358), or the loop electrosurgical excision procedure (LEEP) (N = 489). The medical charts of these patients were reviewed to evaluate the role of ECC in the management of abnormal cervical cytology. The mean age of patients was 27 years (range = 13–77 years). The likelihood of atypia being found on ECC increased with the age of the patient, especially for high-grade lesions. Less than 5 percent of the women under 20 years of age (17 of 345) had human papillomavirus (HPV)/cervical intraepithelial neoplasia 1 (CIN-1), compared with 14 percent of women aged 41 years or more (P = .0001 for trend). Only 2 percent of the younger women (8 of 345) had CIN-2 or CIN-3, compared with 22 percent of the older women (27 of 125) (P = .0000 for trend). When the colposcopic examination was satisfactory (1854 patients), ECC results were abnormal 10 percent of the time (186 of 1854); but when the colposcopic examination was not satisfactory (272 patients), atypical epithelium was seen in 34 percent of the ECC specimens (91 of 2720) (P < .001). Biopsy samples of suspicious lesions seen on colposcopy were obtained. Of 128 women whose biopsy was read as normal, 27 women (21 percent) had abnormal ECC results. Of the 54 women in whom no abnormality was seen on colposcopy and who had no biopsy performed, ECC results were abnormal in 23 patients (43 percent) (P = .001). Ten percent of the 1045 women with biopsy results showing HPV or CIN-1 had abnormal ECC results (2.5 percent were high-grade lesions); 14 percent of the 898 women whose biopsy showed CIN-2 or CIN-3 had an abnormal ECC specimen (4 percent were low-grade lesions). One patient had cervical carcinoma, which was identified by both biopsy and ECC. Of the patients treated with LEEP, 202 had an abnormal ECC specimen. Ectocervical disease was found in 175 of the 202 women (86 percent), and endocervical disease was also seen in 24 of these 175 women (12 percent). In the 120 women who had CIN-2 or CIN-3 on ECC specimens, CIN-2 or CIN-3 was also present in the ectocervical epithelium (LEEP specimens) in 83 cases (69 percent), and invasive cancer was found in 6 other women. In addition, invasive cancer was found in one woman whose ECC specimen was read as HPV/CIN and in one woman with a normal ECC specimen. There were three invasive cancers found in the endocervical epithelium, all in women with an abnormal ECC specimen. Endocervical disease was found in 27 of the 287 LEEP specimens (9 percent) from women with a normal ECC. None of these women had invasive cancer. A cervical smear was obtained and ECC was performed in 1054 patients at the 1-year follow-up examination. Of the 40 women with low-grade disease on ECC, the Pap smear was read as normal in 34, as low-grade disease in 4, and as high-grade disease in 2. Of the 48 women with high-grade lesions on ECC, Pap smear results were read as normal in 31, as low-grade disease in 4, and as high-grade disease in 13. Invasive carcinoma was found in the ECC specimens of two patients; cervical smears indicated one low-grade and one high-grade squamous intraepithelial lesion. In all, 65 of the 986 women with normal Pap smears (6.6 percent) had abnormal ECC specimens; 31 (3.2 percent) were subsequently found to have CIN-2, CIN-3, or carcinoma in situ. J Reprod Med 2000;45:285–292