Urinary albumin-to-creatinine ratio and risk of recurrent worsening heart failure: insights from the spanish cardiorenal registry

Introduction

The urinary albumin-to-creatinine ratio (UACR) reflects systemic endothelial dysfunction, intrarenal inflammation and congestion. Although albuminuria predicts mortality in heart failure (HF), its utility for predicting recurrent worsening HF (WHF) remains uncertain. We evaluated the association between UACR and recurrent WHF in ambulatory chronic HF.

Methods

Two prospective cohorts of clinically stable HF patients were analyzed: a derivation cohort from the Spanish Cardiorenal Registry (n=940) and an external validation cohort from a specialized HF clinic (n=335). UACR (first-morning urine) was analyzed continuously (non-linear terms) and dichotomized at >=30 mg/g. Recurrent WHF episodes and HF hospitalizations were modelled using negative binomial regression, incorporating all-cause death as a terminal event and adjusting for demographics, comorbidities, NYHA class, LVEF, eGFR, NT-proBNP, and guideline-directed medical therapy.

Results

Derivation cohort: mean age 71.9±13.0 years, 36.7% women; mean LVEF 42.8±15.4% (64.9% <50%); median eGFR 54 (IQR 37-77) ml/min/1.73 m2 and UACR 20 (10-70) mg/g, with 57.7% having eGFR <60 and 40.3% UACR >=30 mg/g. Most patients received RASi/ARNI (76.8%) and beta-blockers (78.7%); 57.8% were on MRA and 58.7% on SGLT2i. Over a median of 1.05 years, 135 deaths and 477 WHF events (221 HF hospitalizations) occurred in 197 patients. Event rates were higher with UACR >=30 vs <30 mg/g (WHF: 89.5 vs 37.0; HF hospitalizations: 36.9 vs 13.6; death: 16.7 vs 10.4 per 100 person-years; all p<=0.008). After adjustment, UACR >=30 mg/g remained associated with total WHF (IRR 1.84, 95%CI 1.29-2.62; p=0.001) and recurrent HF hospitalizations (IRR 2.42, 95%CI 1.58-3.72; p<0.001). Continuous UACR showed a non-linear association, with a steep risk increase up to ~150-200 mg/g followed by a plateau.

Validation cohort: mean age 73.8 years, 34.3% women; about 50% had diabetes and eGFR <60 ml/min/1.73 m2; median UACR 28 mg/g and NT-proBNP 1030 (IQR 538-1983) pg/mL; ~70% received diuretics, RASi and beta-blockers, and ~50% SGLT2i. Over a median of 1.13 years, 35 deaths and 75 WHF events (54 HF hospitalizations) occurred in 46 patients. Rates were higher with UACR >=30 vs <30 mg/g (death/first WHF: 24.7 vs 7.5; WHF: 35.3 vs 6.7; HF hospitalizations: 24.8 vs 6.6 per 100 person-years; p<=0.016). Multivariable analyses confirmed the independent association of higher UACR with both endpoints and reproduced the same non-linear pattern.

Conclusions

Across two independent chronic HF cohorts, elevated UACR independently predicted recurrent WHF, including HF hospitalizations, beyond eGFR, natriuretic peptides, and contemporary HF therapies. UACR may refine ambulatory risk stratification and identify patients requiring closer monitoring.Cumulative incidence of WHF eventsFor image description, please refer to the figure legend and surrounding text.Adjusted associationFor image description, please refer to the figure legend and surrounding text.