Clinical profiles and prognostic impact of residual intravascular and tissue congestion in acute heart failure

Aims Residual congestion (RC) is common at discharge after acute decompensated heart failure (ADHF) and is associated with early mortality and rehospitalization. The prognostic value of distinct RC phenotypes (i.e. intravascular and tissue congestion) remains unclear. This analysis investigated RC phenotypes and their outcomes. Methods and results Patients with congestion at admission from two large ADHF trials, PROTECT (rolofylline; index) and RELAX‐AHF‐2 (serelaxin; replication), were classified based on clinical signs at day 7/discharge as intravascular (jugular venous pressure) or tissue (pulmonary rales/peripheral oedema) congestion, each alone, combined or neither. Cox regression assessed 180‐day mortality after adjusting for risk factors. Overall, 1557 patients with predominantly combined (i.e. tissue and intravascular) congestion at admission were included, with a median age of 72 years. By day 7 or discharge, 580 (37%) patients had RC. In these patients, intravascular congestion ( n = 260; 45%) was most common, followed by combined ( n = 185; 32%) and tissue ( n = 135; 23%) congestion. During hospitalization, patients with solely intravascular RC had greater diuretic responses, shorter hospital stays and received lower doses of intravenous loop diuretics than those with tissue or combined congestion (all p < 0.05). Residual intravascular and tissue congestion were independently associated with increased 180‐day mortality (hazard ratio [HR] 1.69, 95% confidence interval [CI] 1.15–2.49, and HR 2.07, 95% CI 1.25–3.41, respectively) compared to decongested patients. In the RELAX‐AHF‐2 substudy ( n = 476), similar findings were observed. Conclusions Patients with intravascular RC had better diuretic responses and shorter hospital stays than those with tissue/combined RC, but worse outcomes than decongested patients. This study highlights the importance of RC assessment to identify at‐risk patients. Future studies should evaluate phenotype‐guided treatments.