Histologic follow-up of atypical endocervical cells. Liquid-based,thin-layer preparation vs. conventional Pap smear

OBJECTIVE: To review the histologic findings in patients diagnosed with "atypical glandular cells of uncertain significance, endocervical cell type" (AGUS-EC) by ThinPrep Pap Test (TPPT) or conventional Pap smear (CPS) and to evaluate the clinical value of subclassifying AGUS-EC as "favor reactive" or "favor neoplastic." STUDY DESIGN: All TPPT and CPS diagnosed as AGUS-EC (favor reactive, unspecified and favor neoplastic) from January 1998 through December 1999 and all available histologic follow-up (defined as endocervical curettage, cervical biopsy, cervical conization or hysterectomy obtained within six months of the time of an AGUS-EC diagnosis) were obtained from a computerized database. RESULTS: AGUS-EC was diagnosed in 0.77% of CPS (683 of 88,825) and 0.59% of TPPT (183 of 30,968) (P = NS). There was no statistically significant difference in any of the follow-up histologic diagnoses between the CPS and TPPT groups. The majority of the follow-up biopsies demonstrated benign processes in both groups. Patients with a diagnosis of AGUS-EC "favor neoplastic" had a greater proportion of true glandular pathology as compared with AGUS-EC "unspecified" or "favor reactive" (P < .001). None of the patients with a diagnosis of AGUS-EC "favor reactive" were found to have true glandular pathology; however, a minority of them proved to have squamous pathology. CONCLUSION: In this study there was no difference in CPS and TPPT in regard to the specificity of a diagnosis of AGUS-EC for true glandular pathology. Subclassifying AGUS-EC as "favor reactive" or "favor neoplastic" may provide valuable information for directing patient follow-up.