#2125 Endotypes of kidney injury in patients with acute decompensation of heart failure and their impact on long-term mortality

Abstract Background and Aims Renal dysfunction with and without albuminuria is common in patients with heart failure and associated with poorer outcomes. However, the prevalence and long-term influence of increased urine albumin creatinine ratio (uACR) and/or isolated decreased glomerular filtration rate (GFR 60 ml/min/1,73 m2) have not been well studied in patients with acute decompensation of heart failure (ADHF). The aim was to investigate the prevalence of different variants of kidney injury and their impact on the mortality in patients with ADHF within 1 year of follow-up. Method This prospective, observational study included 140 patients who were discharged after ADHF. We excluded patients with terminal stage of chronic kidney disease, end-stage heart failure, malignancy. Patients were divided according to the kidney function (the maximum level of GFR) and albuminuria level taken at hospitalization into four groups: normal renal function (GFR60 ml/min/1,73 m2, uACR30 mg/gCr), isolated decreased GFR (GFR60 ml/min/1,73 m2, uACR30 mg/gCr), isolated albuminuria (GFR60 ml/min/1,73 m2, uACR30 mg/gCr), combined changes (GFR60 ml/min/1,73 m2, uACR30 mg/gCr). The observation period was 365 days. Results During hospitalization 25% (n = 35) patients had GFR60 ml/min/1,73 m2, uACR30 mg/gCr, 19.3% (n = 27) patients had isolated decreased GFR, 22.8% (n = 32) patients had isolated uACR30 mg/gCr, and 32.9% (n = 46) patients had both. The baseline and admission characteristics are presented in the table (Fig. 1). 34 (24,3%) patients died during follow-up. Mortality was remarkably higher in groups with registered albuminuria during hospitalization regardless of GFR level (25% (n = 8) patients with isolated increased uACR, 39.1% (n = 18) patients with combined changes; with non-albuminuria: (11.4% (n = 4) patients with GFR60 ml/min/1,73 m2, uACR30 mg/gCr, 14.8% (n = 4) patients with isolated decreased GFR, p = 0.019 for trend). Kaplan-Meier plots revealed worse outcomes for these groups after adjusting for ejection fraction (EF) (Log-Rank p = 0.0057) (Fig. 2). In the multivariate Cox regression, in which age, gender, reduced EF and three categories of kidney damage were included, only combination of decreased GFR and albuminuria was an independent predictor of long-term mortality (hazard ratio: 3.71 95% confidential interval 1.59 to 8.67, p = 0.002). Conclusion Albuminuria is frequently observed in patients with ADHF and, especially with combination with decreased GFR, associated with higher mortality rate during 1 year of follow-up. Albuminuria with decreased GFR during hospitalization with ADHF was an independent risk factor for mortality during one year.