Association of UACR and UPCR with kidney failure: a pooled analysis of clinical trials

Abstract Background and Hypothesis Urinary albumin-to-creatinine ratio (UACR) and urinary protein-to-creatinine ratio (UPCR) are widely used markers for chronic kidney disease (CKD) risk stratification. While albuminuria is considered a sensitive predictor of kidney outcomes, proteinuria remains in use due to cost considerations or traditions. To assess their relative utility in prediction, we compared associations of UACR and UPCR with kidney failure in CKD clinical trials. Methods We conducted a pooled analysis of individual participant-level data from four landmark CKD trials: DAPA-CKD, EMPA-KIDNEY, RENAAL, and IDNT. Overall, these trials included 6 219 participants with CKD, with and without diabetes, with available UACR and UPCR measurements derived from spot or 24-hour urine samples. Associations with kidney failure—defined as dialysis, transplantation, or sustained estimated glomerular filtration rate (eGFR) decline < 15 or <10 mL/min/1.73 m2—were assessed using Cox proportional hazards models adjusted for demographic and clinical covariates. Hazard ratios (HRs) per standard deviation increase in log-transformed UACR and UPCR were compared using relative ratios (RRs) and pooled across trials using random-effects models. Results Across trials, both UACR and UPCR were strongly associated with kidney failure. Pooled HRs per standard deviation increment in log-transformed UACR or UPCR were 3.29 (95% CI: 2.91, 3.72) and 2.89 (95% CI: 2.52, 3.32), respectively, yielding an overall RR of 1.10 (95% CI: 1.03, 1.18), favouring UACR. This difference was clearest in the subgroup with type 2 diabetes and CKD (RR: 1.12; 95% CI: 1.04, 1.20), with less clear evidence for a difference among participants with glomerular disease (RR: 0.97; 95% CI: 0.83, 1.13). Conclusions Both UACR and UPCR are robust predictors of kidney failure, but UACR demonstrated stronger associations, particularly in diabetic CKD. With a standardized urinary albumin assay becoming clinically available in 2027, these results support UACR as the preferred urinary protein measure for risk stratification.