Atypical glandular cells of undetermined significance. Subtyping as a predictor of outcome.

Objective: To determine whether subtyping of atypical glandular cells of undetermined significance (AGUS) cervical smears into reactive process favored (AGUS-RPF) and not otherwise specified (AGUS-NOS) correlates with rates of underlying pathology.

Study Design: We performed a retrospective chart review of 129,676 routine Pap smears at Kaiser Permanente, San Diego, from June 1998 to November 1999. One hundred fifty of the 129,676 were evaluated as AGUS (0.12%). Subjects with concomitant AGUS/SIL Pap smears, a prior history surgery for dysplasia, prior AGUS, a history of cancer or prior hysterectomy were excluded from the study. The remaining AGUS smears were then subtyped into AGUS-NOS and AGUS-RPF based on the criteria of the 2nd Workshop of the Bethesda System. Sixty-eight subjects with an initial AGUS smear underwent evaluation with colposcopy, endocervical curettage, endometrial biopsy and directed cervical biopsies. Significant pathology was determined to be any tissue diagnosis that required further treatment more than a follow-up Pap smear. AGUS subclassifications and underlying pathology were then compared using the chi 2 test with Fisher's Exact Test.

Results: Twenty-seven patients (40%) had AGUS-NOS, and 41 (60%) had AGUS-RPF. There were no significant differences between the groups in regard to age, race, parity, menopause status, HRT use or tobacco use.

Conclusion: Subtyping AGUS cervical smears correlates with underlying rates of pathology. However, AGUS-RPF smears were still associated with an approximately 10% incidence of significant underlying pathology. Patients with AGUS on cervical smears need thorough evaluation, regardless of subtyping.