The essential of urine cytology for the diagnosis and the follow-up of urothelial neoplasia has been widely recognized. However, there are some cases in which a definitive diagnosis cannot be made due to difficulty in discriminating between benign and malignant. This study evaluated the practicality of nucleolar/nuclear volume ratio (%) for the discrimination.Methods:
Using Papanicolaou-stained slides, 253 benign urothelial cells and 282 malignant urothelial cells were selected and divided into a benign urothelial cell and an urothelial carcinoma (UC) cell groups. Three suspicious cases and four cases in which discrimination between benign and malignant was difficult were prepared for verification test. Subject cells were decolorized and stained with 4′,6-diamidino-2-phenylindole for detection of the nuclei and the nucleoli. Z-stack method was performed to analyze.Results:
When the cutoff point of 1.514% discriminating benign urothelial cells and UC cells from nucleolar/nuclear volume ratio (%) was utilized, the sensitivity was 56.0%, the specificity was 88.5%, the positive predictive value was 84.5%, and the negative predictive value was 64.4%. Nuclear and nucleolar volume, number of the nucleoli, and nucleolar/nuclear volume ratio (%) were significantly higher in the UC cell group than in the benign urothelial cell group (P <0.001). In the verification test using the nucleolar/nuclear ratio (%), four of the seven cases were concordant with the final diagnosis.Conclusion:
This study analyzed the nuclear and nucleolar volume to establish an index for discrimination of benign and malignant urothelial cells, providing possible additional information in urine cytology. Diagn. Cytopathol. 2016;44:483–491. © 2016 Wiley Periodicals, Inc.