J Suh1, H Go1, C Sung1, S Baek2, H Hwang1, S Jeong1, Y Cho1. 1. Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea. 2. Department of Clinical Epidemiology and Biostatistics, University of Ulsan College of Medicine, Seoul, Korea.
Abstract
OBJECTIVE: The Paris System (TPS), which was recently introduced, emphasised the key features of malignant urine cytology: a high nuclear to cytoplasmic ratio, nuclear hyperchromasia, irregular nuclear membranes and coarse chromatin. Although other diagnostic features have been described, the diagnostic significance of such features and their application to TPS have not been fully defined for urinary tract washing specimens. METHODS: A total of 142 cases of urinary tract washing specimens with corresponding surgical pathology samples were examined for the key features of TPS and 13 previously described features. The diagnostic performance of TPS and our proposed modification of TPS was compared with that of the current system. RESULTS: In addition to the key features of TPS, in the present study, high-grade urothelial carcinoma (HGUC) frequently exhibited tumour diathesis, a ragged edge of urothelial cell groups, anisonucleosis, India ink nuclei, apoptotic bodies and pleomorphism. As anisonucleosis and India ink nuclei remained independent predictors of HGUC for the multivariate analysis, they were used to modify TPS. The reporting rate of the atypical urothelial cell (AUC) category decreased from 25.3% in the current system to 14.8% in TPS and 11.3% in our proposed modification of TPS. The sensitivity increased from 59.4% in the current system to 70.8% in TPS and 90.0% in this study. The diagnostic accuracy increased from 0.786 in the current system and 0.754 in TPS to 0.859 in this study. CONCLUSIONS: TPS is a useful diagnostic system for urinary tract washing specimens by decreasing the number of AUC cases and increasing sensitivity. In this study, anisonucleosis and India ink nuclei improved the diagnostic accuracy of HGUC.
OBJECTIVE: The Paris System (TPS), which was recently introduced, emphasised the key features of malignant urine cytology: a high nuclear to cytoplasmic ratio, nuclear hyperchromasia, irregular nuclear membranes and coarse chromatin. Although other diagnostic features have been described, the diagnostic significance of such features and their application to TPS have not been fully defined for urinary tract washing specimens. METHODS: A total of 142 cases of urinary tract washing specimens with corresponding surgical pathology samples were examined for the key features of TPS and 13 previously described features. The diagnostic performance of TPS and our proposed modification of TPS was compared with that of the current system. RESULTS: In addition to the key features of TPS, in the present study, high-grade urothelial carcinoma (HGUC) frequently exhibited tumour diathesis, a ragged edge of urothelial cell groups, anisonucleosis, India ink nuclei, apoptotic bodies and pleomorphism. As anisonucleosis and India ink nuclei remained independent predictors of HGUC for the multivariate analysis, they were used to modify TPS. The reporting rate of the atypical urothelial cell (AUC) category decreased from 25.3% in the current system to 14.8% in TPS and 11.3% in our proposed modification of TPS. The sensitivity increased from 59.4% in the current system to 70.8% in TPS and 90.0% in this study. The diagnostic accuracy increased from 0.786 in the current system and 0.754 in TPS to 0.859 in this study. CONCLUSIONS:TPS is a useful diagnostic system for urinary tract washing specimens by decreasing the number of AUC cases and increasing sensitivity. In this study, anisonucleosis and India ink nuclei improved the diagnostic accuracy of HGUC.