关键词: IgA肾病, 预后模型, 外部验证

Abstract: Objective To externally validate the international IgA nephropathy prediction tool in a modern Chinese cohort, so as to provide more evidence for clinical application. Methods This study included IgA nephropathy patients diagnosed by kidney puncture in Tongji Hospital Affiliated to Huazhong University of Science and Technology from January 2012 to December 2016. The predicted risk of each patient was calculated according to the international IgA nephropathy prediction tool, with the 50% decline of estimated glomerular filtration rate (eGFR) or end-stage renal failure as the endpoint. This study evaluated the performance of two models (Model 1: without racial factor; Model 2: with racial factor) in the international IgA nephropathy prediction tools in discrimination, calibration, reclassification and clinical utility. Results The median follow-up time in the modern Chinese cohort in this study was 4 years, and 48(10%) subjects reached the endpoint. The C indices of the two models were 0.86, and the survival curves were well separated, showing excellent discrimination ability. The performance of the two models in calibration is basically the same, both slightly underestimate the risk of 5 years. The reclassification indicates that there is no significant difference between the two models in predicting 5-year risks. The decision curves analysis showed that the two models were clinically practical in predicting five-year risks, and there was no significant difference between two models. Conclusion The international IgA nephropathy prediction tools showed significant discrimination, acceptable calibration and satisfactory clinical utility, and there is no significant difference between the two models.

Key words: IgA nephropathy, prognostic model, external validation