Liquid-based cytology (LBC) has recently become the preferred method for urine cytology analysis, but differences with conventional cytology (CC) have been observed. The purpose of this study is to analyze these differences and the clinical relevance of non-atypical urothelial cell groups (UCG) in voided urine specimens. Reporting terminology is discussed. Initially, diagnostic categories from 619 LBC and 474 CC samples, reviewed by five different pathologists, were compared (phase 1). Five years after LBC was implemented and applying strict cytologic criteria for UCG diagnosis, 760 samples were analyzed (phase 2) and compared to previous LBC specimens. Diagnostic differences, interobserver variability and clinicopathological correlation with a 6-month follow-up, were analyzed. UCG increased from 6.5% with CC to 20.7% (218%, 3.2 fold, P < 0.0001) with LBC. This difference was not related to interobserver variability. Five years later, the rate of UCG had decreased to 13 2%. While 6% of cases with a negative cytology had urothelial carcinoma (UC) within 6 months of diagnosis, this percentage increased to 15.7% with UCG. The sensitivity of the UCG category for UC was low (30.4%), but the specificity and the negative predictive value (NPV) were high (87.1% and 94%, respectively). LBC increases UCG when compared to CC. This can be corrected with observeŕs experience and using set cytological criteria. Due to its association with carcinoma, the presence of UCG in voided urine should be framed in a diagnostic category other than "negative for malignancy." Diagn. Cytopathol. 2016;44:582-590. © 2016 Wiley Periodicals, Inc.
Keywords: Liquid-based cytology; cell groups; increased risk; malignancy; urothelial; voided urine.
© 2016 Wiley Periodicals, Inc.