Reparative changes and the false-positive/false-negative Papanicolaou test- A study from the College of American Pathologists Interlaboratory Comparison Program in Cervicovaginal Cytology

Context.-Reparative changes can be a diagnostic challenge on Papanicolaou t ests and must be distinguished from epithelial cell abnormalities. Both squ amous intraepithelial lesions (SIL) and carcinoma may be underdiagnosed as repair. This study examines laboratory and cytologist performance in the di agnosis of repair. Objectives.-To determine if laboratories and cytologists can consistently d istinguish reparative changes from SIL and carcinoma and to document how of ten SIL and carcinoma are mistaken for repair in a standardized educational slide program. Design.-Results for reparative changes, SIL, and carcinoma slides from the College of American Pathologists Interlaboratory Comparison program in Cerv icovaginal Cytology for 1998 were analyzed. Only results from validated ref erenced slide responses were used. Results.-The concordancy rate for reparative-change slides, that is, any re sponse of within normal limits or benign cellular changes, ranged from 91% to 94% for cytotechnologist, pathologist, and laboratory responses. False-n egative rates for squamous carcinoma and adenocarcinoma, high-grade SIL, an d low-grade SIL ranged from 0.47% to 5.41%; the proportion of false-negativ e diagnoses of reparative changes ranged from 24% to 62% of all discordant responses. Conclusions.-Of all benign cellular changes and within normal limits catego ries in the Interlaboratory Comparison Program in Cervicovaginal Cytology, repair most often elicits a false-positive laboratory response. Underdiagno sing epithelial abnormalities as repair is also a source of false-negative Papanicolaou test results.