Clinical parameters associated with unsatisfactory specimens of conventional cervical smears.

Although The Bethesda System 2001 attempted to standardize the criteria for specimen adequacy, much confusion still exists, which includes the significance of unsatisfactory smears, the causes and clinical conditions related to unsatisfactory smears, and the appropriate management of unsatisfactory smears. The aim of this study is to find out the clinical factors associated with unsatisfactory cervical smears. We reviewed the medical charts of patients who received conventional Pap smears between March 2006 and August 2006 in a tertiary care center. After excluding 378 cases with incomplete demographic data, the clinical data of 7,059 cases were processed for analysis. Clinical parameters retrieved included: history of pelvic malignancy, pelvic irradiation, conization, hysterectomy, pregnancy status, within 3-months postpartum. Vaginal bleeding, abnormal vaginal discharge, intrauterine device, and cervical polyps found during pelvic examinations were also documented. The 1,397 cases with history of pelvic irradiation, pelvic malignancy, and hysterectomy were excluded. Finally, 5,662 cases were enrolled for data analysis. The relationship between clinical parameters and unsatisfactory smears were analyzed by Pearson's chi-square test with Yates' continuity correction and multivariate binary logistic regression test. The incidence of unsatisfactory smears was 4.5% (252/5,662). Clinical parameters correlated with unsatisfactory smears were postpartum status (OR = 1.92, 95% CI = 1.23-3.01, P = 0.004), vaginal bleeding (OR = 2.02, 95% CI = 1.30-3.16, P = 0.002), and endocervical polyps (OR = 2.62, 95% CI = 1.39-4.947, P = 0.003). In conclusion, if any of these parameters are noted prior to obtaining a Pap smear, optimal collecting devices, better sampling techniques, and liquid-based cytology should be considered to decrease the incidence of unsatisfactory smears.