The utility of upper urinary tract urine cytology before and after application of the Paris system

Abstract

Background The Paris System (TPS) introduced diagnostic criteria for urine cytology to improve reproducibility among pathologists. Thus far, most cytology studies have investigated application of TPS on lower urinary tract specimens. Also, it is unclear which cytologic features are most predictive of malignancy, particularly in the upper urinary tract. We evaluate concordance rates of preoperative upper urinary tract cytology specimens before and after application of TPS criteria with surgical resections and assess cytologic features associated with malignancy. Design 54 resections with high‐ and low‐grade urothelial carcinoma (HGUC, LGUC) from 2000‐2016 with available preoperative cytology (n = 61) were identified. Cytology was re‐reviewed to evaluate cytologic features and provide diagnoses before and after TPS implementation. Results The most common cytologic features associated with HGUC were N:C ratios ≥0.7 (88%), hyperchromasia (83%), coarse chromatin distribution (67%), and nuclear pleomorphism in cell clusters (65%). Application of TPS criteria resulted in 10 diagnostic downgrades and 6 diagnostic upgrades. After TPS criteria were applied, the sensitivity of a positive diagnosis decreased from 29% to 19%. The morphologic feature most consistently associated with a downgrade from positive to suspicious was a lack of marked nuclear contour irregularities in atypical urothelial cells. Conclusion Using strict TPS criteria in upper urinary tract cytology specimens may decrease the frequency of positive diagnoses with a concurrent increase in suspicious diagnoses. These findings may indicate that different morphologic features, particularly markedly irregular nuclear contours, may have different predictive values for HGUC in upper urinary tract cytology specimens compared with those from the lower urinary tract.

Should “suspicious for high‐grade urothelial carcinoma” and “positive for high‐grade urothelial carcinoma” remain separate categories?

“…A few recent studies have addressed the performance of TPS on UUT samples in comparison with LUT cytology 9,13,25,26 . Zheng et al 13 established ROMs for different categories of TPS in the UUT (0% for nondiagnostic, 6% for NHGUC, 27.3% for AUC, 0% for LGUN, 48% for SHGUC, and 95% for HGUC) and concluded that these risks were comparable to those reported in the literature.…”

Comparison of urine cytology diagnostic reports before and after the implementation of the Paris System classification system in China

“…In addition, our study further supported the notion that TPS could effectively reduce the rate of uncertain AUC diagnoses associated with patients’ anxiety, and better guide clinicians to adopt appropriate management strategies. A contrasting point of view was presented in the studies of Granados et al and Simon et al, who suggested that adopting TPS could notably increase the rate of atypical diagnoses. These authors concluded that the too strict diagnostic criteria of TPS would lead to a decrease in diagnostic accuracy.…”

“…The different viewpoints may be mainly attributed to the sample selection bias. Furthermore, the number of cytopathologists who participated in TPS diagnosis, inter‐observer variability, sample preparation methods, the type of urine sample and the location of the tumour may also affect the research results. In addition, the differences in the number of categories used in the traditional cytological analysis and TPS may also contribute to the divergent results.…”

The Paris system of Reporting Urinary Cytology: Strengths and opportunities

“…The first edition of TPS did not specifically address upper tract samples, and the application of TPS criteria to the renal pelvis and in one study ureteral washings resulted in a reduced diagnosis of HGUC and poorer concordance with surgical resection diagnosis 39 . Alternatively, other studies have shown not only the morphologic criteria of the lower urinary tract could be used for the upper urinary tract, but also that the category based outcome and PPV of both sites are similar to one another 40,41…”

“…Pathology continues to evolve and morphology is progressively aided by new technology. The next step will be to have our surgical pathology colleagues adapt the same terminology—just like Bethesda system lead the way to the Lower Anogenital Squamous Terminology (LAST) 39 . Also, there are still a number of studies that need to be performed to explore as yet unanswered questions regarding urinary cytology.…”