Evaluation of Abnormal Papanicolaou Smears: Role of Cytology and Colposcopy

Abstract

C ervical cytology as a screening method for cervical abnormalities varies in sensitivity, specificity, and positive and negative predictive value [1]. Atypical squamous cells of undetermined significance yield a large percentage of cervical intraepithelial neoplasia grades 2 and 3 (CIN2 and CIN3) lesions [2] and, contrarily, CIN1 is confirmed histologically in not more than 24% of cases [3]. Concordance between colposcopic prediction and histology is 66% ( ϭ 0.40) [3][4][5][6]. This study compares Papanicolaou (Pap) smear results that initiated [5] referral Pap smears taken at the first colposcopic evaluation with histological workup; age at which satisfactory and unsatisfactory colposcopy results were obtained; positive predictive value of acetowhite epithelium; and punctation and mosaicism as compared to histological yield. AbstractObjective. Our goal was to compare results of referral and control cytology and associate colposcopic images with histological biopsy and endocervical curettage (ECC).Material and Methods. A retrospective review of 1,144 cases of colposcopy was undertaken from December 1991 to July 1997. Kappa statistics were applied to referral and control Papanicolaou (Pap) smears during the first colposcopy with histological evaluation, and the chi-square test was used to analyze statistically for age, acetowhite epithelium, and punctation or mosaicism.Results. Referral Pap smears agreed poorly with those of controls and with histological findings. The age of patients at satisfactory colposcopy (defined as visualized squamocolumnar junction; 32 years (range, 14-84) differed from that of patients at unsatisfactory colposcopy (44 years; range, 18-81) (Pearson coefficient ϭ 203.4; p Ͻ .0001). Patients in whom cervical intraepithelial neoplasia grade 1 (CIN1) and CIN2/CIN3 were diagnosed were younger, at 30 years (range, 14-58) and 31 years (range, 18-66), respectively, as compared to patients in whom findings were negative (34 years; range, 16-66) or in whom metaplasia was found (33 years; range, 16-63) (Pearson coefficient ϭ 30.26; p ϭ .003). Dense acetowhite epithelium (grade 2) predicted CIN in 159 of 200 cases (79.5%), as compared to light acetowhite epithelium (grade 1) in 102 of 154 cases (66.2%) (Pearson coefficient ϭ 17.96; p ϭ .0001). If punctation or mosaicism did not predict CIN as opposed to negative histology ( p ϭ .82), when CIN1 and CIN2/CIN3 histological findings were divided into two groups, prediction of CIN from punctation or grade 2 mosaicism was made in 32 of 47 cases (68.1%), as compared to 41 of 62 cases (66.1%) from grade 1 findings (Pearson coefficient ϭ 10.67; p ϭ .005). Performed ECCs were found to be positive for CIN in 59 of 641 patients (9.2%), including 37 (5.8%) CIN1 and 22 (3.4%) CIN2 and CIN3 cases.Conclusions. Control Pap smears do not concur with referral smears; they must be repeated. Satisfactory colposcopy is observed in those younger patients with CIN. Acetowhite epithelium grade 2 significantly predicts CIN, and punctation or grade 2 mosaicism predicts CIN2/C...

Endocervical Curettage in Cervical Evaluation

“…Analysis of variance was done for age and parity, and Pearson's chi‐square test for other variables, such as cytology, acetowhite epithelium, microbiopsy, and satisfactory–unsatisfactory colposcopy. Acetowhite epithelium was classified as grade I and II according to the level of whiteness [7]. The same pathologist reviewed all the microbiopsies and endocervical curettages.…”