Carcinoma of the uterine cervix is a theoretically preventable disease because its precursor lesions can be detected by cervical Papanicolau smears and appropriately treated, Although cervical cytology screening programmes have resulted in the redution of cervical cancer incidence and mortality, Pap smear have been subjected to intense scrutiny and criticism in recent years. The focus of criticism has been the false-negative Pap smear, and the false-negative Pap smear is the major quality issue currently facing the physicians. To reduce the false-negative rate of Pap smear, it is essential to improve the accuracy of Pap smear. But false-negative rate of Pap smear has been reported variously. OBJECTIVE: This study was undertaken to evaluate accuracy of Pap smear by study false-negative and false-positive rate of Pap smear and to determine whether false-negative and false-positive rate had any correlations with clinical factors.

The study population was comprised of 346 women, who were undertaken gynecologic operation at the Department of Obstetrics & Gynecology at Hanyang University hospital between March, 1997 and April, 1998. All patients were taken Pap smear before operation. In 93 women of these, preoperative diagnosis were cervical intraepithelial neoplasia and carcinoma in situ of uterine cervix, and in 253 women of these, preoperative diagnosis were benign disease as uterine myoma or adenomyosis, etc. All of their surgical specimen were examined. Pap smear, pathology, medical charts of all patients were reviewed retrospectively, and false-negative rate and false-positive rate were calculated. Clinical factors that associated with false-negative and false-positive rate were evaluated. Fishers exact test and Pearson chi-square test were used of statistical analysis.

False-negative rate of Pap smear was 7.2%, false-positive rate was 4.6%, corresponding rate with histology was 88.2%. Sensitivity and specificity of PAP smear were 87.0% and 97.0% respctively. According to gross finding of uterine cervix, erosion was 46.6% in cervical intraepithelial neoplasia, 67.8% in carcinoma in situ, 66.6% in microinvasive carcinoma of uterine cervix and 55.3% of 103 erosion findings was cervical intraepithelial neoplasia, carcinoma in situ or microinvasive carcinoma. 23.1% of cervical lesion were normal gross finding. Menopause was associated with false-negative rate and previous vaginal infection history, previous cervical minor operation, delivery mode, contraception method, pelvic inflammatory disease history, vaginal bleeding at Pap smear and gross finding of cerbix were not associated. There were no clinical factors that were associated with false-positive rate.

Compared with other reports, false-negative rate(7.2%) and false-positive rate(4.6%) of Pap smear was lower and corresponding rate(88.2%) was higher in Hanyand university hospital. Because of higher false-negative rate in menopausal women, it need more careful to take and interpretate Pap smear in these group.