Clinical Factors That Predict Discrepancy between Cervical Cytology and Colposcopically Directed Biopsies

23  Abstract— Objective: The aim of this study was to identify clinical variables that might account for the discrepant results between cervical cytology (CC) and corresponding colposcopically directed cervical biopsy (CB). Methods: The pathology database was reviewed for consecutive cases of cytology and their corresponding biopsies between 2010 and 2012. 114 discordant cases and 190 concordant cases (control group) were identified. Discrepant cases were classified as (1) major and (2) minor. A major discrepancy is a case with high-grade cytology (ASC-H or HSIL) and histologic benign or CIN 1. A minor discrepancy is a case with low-grade cytology (ASC-US or LSIL) and histologic CIN-2 or -3. Uni- and multivariate analyses were performed to identify clinical variables associated with CC/CB discrepancy. Results: The mean age for patients was 41 (range 14-73). Of the total study population, 57% were Hispanic, 24% were white and 19% identified with another racial group. Major discrepancies were 4 times more likely to occur among CC collected by mid-level practitioners compared to CC collected by physicians (p=0.037). Major discrepancies were also found to be associated with benign or CIN 1 on clinical impression at colposcopy (p=0.036). Furthermore, an immune compromised status showed a trend towards significance for major discrepancies (p=0.062). High-risk HPV positivity was associated with a 3.5 fold increase in observing a minor discrepancy between CC and CB (p= 0.008). Minor discrepancy was also associated with younger age of patient (p = 0.001). The training level of the colposcopist did not significantly impact the rate of CC/CB discrepancy. Conclusions: This study identifies several variables associated with CC/CB discrepancies that are important for a rigorous multi-disciplinary quality assurance effort.