ATYPICAL SQUAMOUS CELLS OF UNDETERMINED SIGNIFICANCE - A COMPARATIVE REVIEW OF ORIGINAL AND AUTOMATED RESCREEN DIAGNOSIS OF CERVICOVAGINAL SMEARS WITH LONG-TERM FOLLOW-UP

BACKGROUND. There is an increasing number of articles regarding the lo ng term follow-up of Papanicolaou (Pap) smears with the diagnosis of a typical squamous cells of undetermined significance (ASCUS). Much cont roversy exists regarding the management of patients with this diagnosi s. In a prior study in 1992, the authors performed automated rescreeni ng of 101 ASCUS cases and 91 negative (control) cases. They found that through PAPNET-directed rescreening, 35 of 101 ASCUS cases (35%) coul d be reclassified as a squamous intraepithelial lesion (SIL). METHODS. These 192 women were followed since 1992 through manual look backs of subsequent Pap smears and surgical biopsies over a 4-year period. The population studied was comprised of predominantly black women between the ages of 14 and 85 years. The majority were considered a high risk population because many had a history of several sexual partners and multiple pregnancies. RESULTS. Eighteen of 74 patients (24.3%) with an original diagnosis of ASCUS were found on subsequent Pap smears to ha ve an SIL. Only 4 of 64 patients (6%) who originally had a negative Pa p smear subsequently were found to have a low grade squamous intraepit helial lesion (LGSIL) within 4 years. Through ordinal logistic regress ion analysis, it was found that patients with an ASCUS diagnosis had a risk of developing SIL that was 2.6 times greater than the risk for p atients with a negative smear diagnosis. Comparing the surgical biopsi es in the control and ASCUS groups, there was no statistically signifi cant difference in the risk of developing SIL. This may be because the number of follow-up biopsies were small. CONCLUSIONS. A statistically significant difference of the risk of developing SIL exists between p atients with a negative smear versus those with an ASCUS smear. Long t erm follow-up is essential in the management of the patients with an A SCUS smear because there is clearly an increased risk of developing SI L. (C) 1997 American Cancer Society.