Atypical glandular cells: improvement in cytohistologic correlation by the 2001 Bethesda system

The objective of this article was to evaluate clinical significance of glandular atypia on Papanicolaou smear, to compare the Bethesda system (TBS) 2001 with the 1991 revision, and to determine whether there is any improvement in the cytohistologic correlation by the new system. Cytology files of 18,955 patients were reviewed for diagnosis of atypical glandular cells of undetermined significance (AGUS), and histopathology files were searched. Cervical smears of these patients were reclassified according to TBS 2001. Of the 18,955 specimens, 89 (0.46%) were diagnosed as AGUS. Of these 89 women, 76 (85.3%) accepted the follow-up protocol of our hospital. In reevaluation according to TBS 2001, 31 specimens were reevaluated as atypical glandular cells (AGC) and 3 were reevaluated as adenocarcinoma in situ, 8 as AGC with concomitant squamous cell abnormalities, 1 as atypical squamous cells that cannot exclude high-grade squamous intraepithelial lesions, and 33 as negative. Thirty-one (93.9%) of these 33 negative cases were in the AGUS-reactive category in the initial examination. The difference between the rates of the malignant pathologies in the AGUS (25%, 19/76) and in the AGC (42.8%, 18/42) categories was significant (chi(2) = 4.0, P = 0.04). The new terminology of AGC is more likely to suggest a clinically significant lesion than TBS 1991. Repeated cytologic testing during follow-up seems to be unacceptable.