Objective: To validate the diagnostic performance of the Paris system for reporting urinary cytology (TPS). Methods: A total of 7 046 urine cytology samples from 3 402 patients collected in the Department of Pathology, Beijing Hospital, China from January 2020 to January 2022 were analyzed. 488 patients had a biopsy or resection specimen during the follow-up period of 6 months. The sensitivity, specificity, risk of malignancy (ROM) and risk of high-grade malignancy (ROHM) of the TPS were evaluated using histological diagnosis as the golden standard. Results: Among the 7 046 samples, high-grade urothelial carcinoma (HGUC) accounted for 5.7% (399/7 046), suspicious for high-grade urothelial carcinoma (SHGUC) for 3.2% (227/7 046), atypical urothelial cells (AUC) for 8.4% (593/7 046), and negative for high-grade urothelial carcinoma (NHGUC) for 72.9% (5 139/7 046) including low-grade urothelial neoplasm (LGUN) for 0.8% (59/7 046) and insufficient samples for 9.8% (688/7 046). 488 patients had a bladder biopsy or resection in the follow-up of six months, including 314 males and 174 females, aged 27 to 92 years (average, 66 years). The ROHM of TPS was 94.7% in HGUC, 83.3% in SHGUC, 41.3% in AUC and 18.8% in NHGUC. The sensitivity and specificity of urine cytology were 70.1% (169/241) and 97.0% (162/167), respectively. The negative predictive value of NHGUC was 69.2% (162/234). Conclusions: The study has shown that TPS classification has high sensitivity and specificity, high ROHM for HGUC and SHGUC, and high negative predictive value for NHGUC. The application of TPS reporting system can better interpret the clinical significance of cytology samples, improve the accuracy of urine cytopathology and ensure continuous diagnostic consistency.
目的： 评价应用尿细胞学巴黎报告系统（TPS）的有效性。 方法： 回顾北京医院病理科2020年1月至2022年1月来自3 402例患者送检尿细胞学样本共计7 046例，其中488例患者在随访6个月内行组织学活检或切除，以组织学诊断为金标准，评价尿细胞学TPS的灵敏度、特异度、诊断分类的恶性风险度及高级别恶性风险度。 结果： 7 046例尿细胞学样本中，高级别尿路上皮癌占5.7%（399/7 046），可疑高级别尿路上皮癌占3.2%（227/7 046），不典型尿路上皮细胞占8.4%（593/7 046），高级别尿路上皮癌阴性占72.9%（5 139/7 046），其中低级别尿路上皮肿瘤占0.8%（59/7 046），样本不足占9.8%（688/7 046）。在随访6个月内行组织学活检或切除488例患者中男性314例，女性174例，年龄27~92岁（平均66岁）。TPS分类高级别恶性风险度分别为高级别尿路上皮癌94.7%、可疑高级别尿路上皮癌83.3%、不典型尿路上皮细胞41.3%及高级别尿路上皮癌阴性18.8%。TPS的灵敏度和特异度分别为70.1%（169/241）和97.0%（162/167），高级别尿路上皮癌阴性的阴性预测值为69.2%（162/234）。 结论： 本研究显示TPS诊断的灵敏度和特异度高、恶性诊断的高级别恶性风险度高，以及高级别尿路上皮癌阴性的阴性预测值高。应用TPS诊断系统可以更好诠释诊断的临床意义，深入了解本实验室尿细胞学诊断的准确度，从而不断提高诊断水平。.