Urinary Angiotensinogen as a Novel Mortality Predictor in Diabetic Kidney Disease: A Propensity-Matched Analysis from the KNOW-CKD

Background Patients with diabetic kidney disease (DKD) face a substantially higher risk of premature death compared to non-diabetic counterparts, yet accurate biomarkers for stratifying this mortality risk beyond conventional kidney function measures are currently lacking. We investigated whether urinary angiotensinogen (UAGT), reflecting intrarenal renin-angiotensin-aldosterone system activity, predicts all-cause mortality differently among patients with versus without diabetes in chronic kidney disease (CKD). Methods This prospective cohort study included 824 propensity score-matched patients (412 with diabetes, 412 without diabetes) from the Korean Cohort Study for Outcome in Patients with Chronic Kidney Disease. Participants were categorized into quintiles by baseline urinary angiotensinogen-to-creatinine ratio (UAGT/Cr). Cox proportional hazards models estimated adjusted hazard ratios (HRs) for all-cause mortality. Results During median follow-up of 8.9 years, 149 deaths occurred. In patients with diabetes, the highest UAGT/Cr quintile was independently associated with increased mortality compared to the reference quintile (adjusted HR 3.61, 95% confidence interval 1.67-7.81, P<0.001) after adjustment for kidney function, proteinuria, and comorbidities. Penalized spline analysis demonstrated a monotonic relationship between UAGT/Cr and mortality in patients with diabetes. In contrast, no significant association was observed in patients without diabetes (P for interaction <0.001). Findings remained robust in sensitivity analyses. Conclusion UAGT/Cr independently predicts all-cause mortality specifically in CKD patients with diabetes but not in those without diabetes, suggesting its potential utility as an etiology-specific prognostic biomarker for risk stratification in diabetic kidney disease.