Effectiveness of automated cervical cytology rescreening using the AutoPap 300 QC System

Park, Yong Kyun

Abstract

OBJECTIVE
This study assessed the usefulness of the AutoPap 300 QC System in identifying of detection false negative(DFN) smears in a slide population previously screened as "within normal limits" and compared the rate of identification of DFN to that using 10% random selection. MATERIALS AND METHODS: From December 1996 to August 1998, we utilized the AutoPap 300 QC System at 30% selection rate to select cases for manual review. During the study period, 1,040(83.4%) among 1,247 smears and 7,848(84.0%) among 9,343 smears were screened as "within normal limits" and were included for the AutoPap 300 QC System and 10% random selection studies, respectively. RESULTS: 357(34.3%) among 1,040 were selected by the AutoPap 300 QC System at 30% QC selection rate and reviewed manually. Overall, 63 among 357 were determined to be abnormal including 47 ASCUS, 2 AGUS, and 14 LSIL. 785(10%) among 7,848 were selected by 10% random selection and reviewed manually. 96 among 785 were determined to be abnormal including 69 ASCUS, 3 AGUS, and 24 LSIL. These results represented an increase in pick up rate of DFN(detection false negative) of 5.1 and 4.4 fold for ASCUS/AGUS and LSIL, respectively and of 1.5 and 1.3 fold when accounting for the volume differences measured. CONCLUSIONS: The study demonstrated that the AutoPap 300 QC System is superior to 10% human random screening for the identification of DFN smears. Further studies are required using variable selection rates and a larger number of smears to fully assess the value of the device in quality control mode.