Cytohistological correlation of urine cytology in a tertiary centre with application of the Paris system.

OBJECTIVES: Urine cytology has a high sensitivity in morphological recognition of high-grade urothelial carcinoma (HGUC) as compared to low-grade urothelial neoplasms. The Paris system (TPS) was designed to standardise the reporting of HGUC on urine cytology. We evaluated the diagnostic efficacy of urine cytology in routine practice with the cytohistological correlation and application of TPS.

METHODS: A retrospective review of urine cytology reported from 2014 to 2015 was carried out with cytohistological correlation, wherever available. The cases were also recategorised according to TPS with the calculation of malignancy risk of all categories.

RESULTS: More than 4000 specimens from 1396 patients were evaluated. Histopathology was available in 244 cases. Urine cytology had a sensitivity of 70.5%, specificity of 78.4%, with an overall diagnostic accuracy of 71.7%. The urine cytology when categorised by TPS had 0.9% unsatisfactory cases, 75.9% negative for HGUC, 8.5% atypical urothelial cells, 0.2% suspicious for HGUC, 14.1% HGUC and 0.4% cases of other malignancies. The risk of HGUC with each category of negative for HGUC, atypical, suspicious for HGUC, HGUC and other malignancy was 11.6%, 12.3%, 33.3%, 58.8% and 80%, respectively.

CONCLUSION: Urine cytology continues to show high diagnostic accuracy for the detection of urothelial carcinoma with a relatively high sensitivity for HGUC. TPS for reporting urine cytology defines specific criteria for diagnosing atypical and HGUC but increases the cases in the atypical category.