The paradox of transient worsening renal function in patients with acute heart failure: the role of B-type natriuretic peptide and diuretic response

J Cardiovasc Med (Hagerstown). 2017 Nov;18(11):851-858. doi: 10.2459/JCM.0000000000000546.

Abstract

Aims: Worsening renal function (WRF) occurs in one-third of patients hospitalized for acute decompensated heart failure. Recently, WRF was categorized in two subtypes: persistent and transient WRF. Thus, we sought to investigate the different prognostic impact of persistent vs. transient WRF; we also evaluate the relation of two WRF phenotypes with congestion, B-type natriuretic peptide (BNP) changes, and diuretic response at discharge.

Methods: The prospective was a single centre study including patients screened for interventional Diur-heart failure Trial (NCT01441245). Patients were eligible if they were admitted with a primary diagnosis of acute heart failure with evidence of volume overload. Persistent WRF was defined as a sustained creatinine increase by at least 0.3 mg/dl throughout the hospitalisation; transient WRF was defined as creatinine increase by at least 0.3 mg/dl within 72 h and a return to baseline levels at discharge. Patients were followed for 6 months after discharge.

Results: Our population included 192 acute decompensated heart failure patients. In total, 61 patients developed persistent WRF and 29 developed transient WRF. Patients with persistent WRF showed a lower mean urine output with respect to the transient WRF group and patients with preserved renal function (1618 ± 374 vs. 2132 ± 392 vs. 2075 ± 442 ml; P < 0.001). Similarly, patients with transient WRF demonstrated a higher rate of BNP decrease more than 30% than seen in patients with stable creatinine levels and in the persistent WRF group (95 vs. 76 vs. 54%; P = 0.001). Univariate Cox regression analysis demonstrated that BNP decrease less than 30% [HR 2.15 (1.40-3.40); P < 0.001] and persistent WRF [HR 1.70 (1.11-2.61); P = 0.01] were related to poor outcome; conversely, transient WRF should be considered as a protective factor [HR 0.42 (0.19-0.93); P = 0.03]. In the multivariable model, only persistent WRF appeared to be related to poor prognosis [HR 1.61 (1.02-2.57); P = 0.04].

Conclusion: WRF occurring during hospitalization has a different significance: transient deterioration appears to be associated with a favourable clinical course; conversely, persistent WRF is related to poor outcome.

MeSH terms

  • Acute Disease
  • Aged
  • Aged, 80 and over
  • Biomarkers / blood
  • Creatinine / blood
  • Female
  • Glomerular Filtration Rate
  • Heart Failure / complications*
  • Hospitalization
  • Humans
  • Male
  • Natriuretic Peptide, Brain / blood*
  • Prognosis
  • Proportional Hazards Models
  • ROC Curve
  • Renal Insufficiency / blood*
  • Renal Insufficiency / epidemiology*
  • Retrospective Studies

Substances

  • Biomarkers
  • Natriuretic Peptide, Brain
  • Creatinine