ABSTRACT
OBJECTIVE: The goal of this study was to determine the clinical factors independently associated with inaccuracy of colposcopically directed punch biopsy in defining extent and severity of epithelial lesions of the cervix. METHODS: The study was conducted in a colposcopy clinic devoted to the management of patients with abnormal Pap smear and/or histology diagnosis of cervical intraepithelial neoplasia (CIN) on punch biopsy or endocervical curettage. Seven hundred and thirty-nine patients (median age, 33 years; range, 18-83 years) referred with a punch biopsy diagnosis of CIN2-3 and treated with electrosurgical conization were evaluated. Cone histology diagnosis was assumed to supply the correct diagnosis. Two types of deviation of biopsy diagnosis were considered: (1). unconfirmed high-grade CIN diagnosis, defined as a cone diagnosis of CIN1 and reactive/reparative changes, and (2). nondiagnosis of carcinoma, defined as a cone diagnosis of microinvasive and invasive squamous carcinoma. Multinomial logistic regression analysis was used to assess the determinants of the probability of each type of deviation. RESULTS: The probability of unconfirmed high-grade CIN diagnosis (n = 190, 25.7%) was inversely related to number of quadrants involved, severity of referral Pap smear, and grade of biopsy. The probability of nondiagnosis of carcinoma (n = 43, 5.8%) was positively related to patient age, invisibility of the squamocolumnar junction, number of quadrants involved, and cone width. Endo-ectocervical location, cone depth, and time period had no effect whatsoever. CONCLUSIONS: The study confirmed previous observations regarding the positive association of patient age and invisibility of squamocolumnar junction with the probability of nondiagnosis of carcinoma. The concomitant positive effects of number of quadrants involved and cone width were suggested to mirror the effect of circumferential development and, respectively, surface area of CIN. Severity of referral Pap smear and grade of biopsy were shown to be inversely related to the probability of a cone diagnosis of CIN1 and benign changes.
