#963 Discrepancy between actual and estimated urine creatinine excretion predicts mortality in chronic kidney disease: from the KNOW-CKD study

Abstract Background and Aims Urine creatinine excretion is a surrogate marker of muscle mass and low urine creatinine excretion is a predictor of poor clinical outcome. There are several methods to estimate urine creatinine excretion. However, little is known about the clinical implication of discrepancy between actual and estimated urine creatinine excretion in chronic kidney disease (CKD). The aim of this study is to investigate the association between actual and estimated urine creatinine excretion discrepancy and mortality in CKD. Method This study analysed 1805 participants from the KNOW-CKD cohort. They underwent 24-hr urine collection and measured urine creatinine excretion at baseline. Their estimated urine creatinine excretion was calculated by Joachim's method. Discrepancy between actual and estimated urine creatinine excretion (dUCr) was defined as actual urine creatinine excretion minus estimated urine creatinine excretion. Patients were divided into 3 groups according to their tertile of dUCr. The study endpoint was all-cause death. Results During a follow-up of 13 062 person-years (median 8.0 years), 151 patients (8.4%) died, with a corresponding death rate of 11.6 (95% CI 9.9-13.6) per 1000 patients-years. 86 (14.4%), 41 (6.8%) and 24 (3.4%) patients from each 1st to 3rd tertile of dUCr died (P < 0.001). In multivariate Cox proportional hazard analysis, there was a graded association of dUCr with all-cause mortality. The adjusted hazard ratios (95% CI) of 2nd and 3rd tertile were 0.58 (0.39-0.86) and 0.36 (0.22-0.58) compared with the 1st tertile. Conclusion Higher discrepancy between actual and estimated urine creatinine excretion is associated with lower risk of mortality in predialysis CKD patients. This association was independent of various conventional and CKD-related risk factors.