Sensitivity and specificity of endocervical curettage and the endocervical brush for the evaluation of the endocervical canal

Abstract

A series of 87 consecutive conization specimens was studied to evaluate the accuracy of endocervical curettage for the detection of dysplasia in the endocervical canal and to investigate the role of the endocervical brush for the outpatient management of patients with atypical Papanicolaou smears. For patients having cervical intraepithelial neoplasia within the endocervical canal in conization specimens, the false-negative rate observed for endocervical curettage was 45% and the false-positive rate for the detection of endocervical involvement was 25%. The false-negative rate for the endocervical brush was only 8.4%, but the false-positive rate for endocervical involvement was 62.5%. The false-negative rate of endocervical curettage could be reduced to 16.7% if an abundant volume of endocervical material (as determined by point counting) was required. The utility of endocervical curettage to detect cervical intraepithelial neoplasia in the endocervical canal appears to be dependent on the adequacy of the specimen. The determination of adequacy is a critical factor for the proper interpretation of specimens obtained by endocervical curettage.