Residual congestion at discharge assessed by bioelectrical impedance and its prognostic value in acute heart failure: a multi-centre, prospective registry study

Residual congestion at discharge is an important risk factor for early readmission and mortality in acute heart failure (AHF); however, it is often underestimated. This study aimed to evaluate the clinical value of oedema index (EI) and phase angle (PA) from bioelectrical impedance analysis (BIA) in assessing residual congestion and predicting short-term outcomes in patients with AHF.

A multi-centre registry database from the steady movement with innovating leadership for heart failure (HF) between 2019 and 2023 was analysed. EI and PA were measured using direct segmental multi-frequency BIA device. The worsening heart failure events (WHF) were defined as a composite of HF readmission and all-cause mortality at 90- and 180-days. Receiver operating characteristic analysis determined the EI and PA cut-off values. Logistic regression and Cox proportional hazard analyses were used to assess the associations between BIA parameters and clinical outcomes.

A total of 600 patients were included, of whom 360 (60%) were male, with a median age of 72.1 years. Within 180 days, 63 patients (11%) experienced WHFs. EI (odds ratio [OR] 1.83; 95% confidence interval [CI] 1.43-2.35 at 90 days, OR 1.76; CI 1.44-2.14 at 180 days, all P < .01) and PA (OR 0.93; CI 0.90-0.96 at 90 days, OR 0.94; CI 0.91-0.96 at 180 days, all P < .01) showed correlation in univariate analysis. EI ≥ 0.4 and PA < 4.5° were identified as prognostic thresholds for unfavourable outcomes, with predictive values of 75% (P < .01, sensitivity 83%, and specificity 56%), and 71% (P < .01, sensitivity 81%, and specificity 54%), respectively. Increased EI (adjusted hazard ratio [aHR] 1.71; CI 1.26-2.31 at 90 days, aHR 1.64; CI 1.30-2.07 at 180 days, all P < .01) and decreased PA (aHR 0.91; CI 0.86-0.96 at 90 days, aHR 0.92; CI 0.89-0.97 at 180 days, all P < .01) were independently associated with WHFs. Based on the estimated cut-offs, when EI and PA were considered together to classify the high-risk BIA group (EI ≥ 0.4 and PA < 4.5°), the risk was apparently increased (aHR 6.79; CI 1.71-26.93 at 90 days, aHR 3.43; CI 1.41-8.32 at 180 days, all P < .05) compared with normal BIA (EI < 0.4 and PA ≥ 4.5°) group.

The EI and PA measured at discharge independently predicted short-term WHFs in patients with AHF. The suggested thresholds for identifying patients at high-risk of residual congestion were EI ≥ 0.4 and PA < 4.5°. As a non-invasive and accessible modality, BIA holds promise as a complementary tool to established biomarkers for evaluating congestion at discharge.