Summary
Revista Brasileira de Ginecologia e Obstetrícia. 2006;28(8):479-485
DOI 10.1590/S0100-72032006000800007
PURPOSE: to evaluate whether factors related to the adequacy of the sample, cell pattern and cytomorphological criteria are associated with false-negative (FN) results of cervical cytopathology during routine examinations. METHODS: this is a case-control study in which the study group included 100 cytopathologic smears with FN results detected during systematic internal quality control consisting of 100% rapid review. For each FN result detected, two smears with a true-positive diagnosis were identified by the same cytotechnician and these constituted the control group, making a total sample size of 300 smears. The variables were established in accordance with the criteria defined for the analysis of sample adequacy, cell pattern and cytomorphological analyzed criteria. The results were evaluated using bivariate analysis and logistic regression with stepwise variable selection criteria expressed in OR (95%). RESULTS: the number of atypical cells, the appearance of nuclear chromatin, and the distribution and presentation of atypical cells in the smear were the variables that showed the greatest risk for FN results with OR of 9.6, 4.2, 4.4, and 3.6, respectively. Inflammatory processes and the presence of blood in the smear were also identified as variables that influence the risk of FN results. CONCLUSIONS: the majority of the factors associated with FN results are dependent on the conditions and techniques of sample collection, since in the majority of cases, the lesion may not be adequately represented in the smear. Confounding factors such as blood and inflammatory processes may also impair analysis. With respect to cytomorphological alterations, thin chromatin strand was the variable that indicated the greatest risk of FN results.