Background: Cardiogenic shock (CS) due to heart failure (HF) represents a substantial proportion of all CS cases. In this CS subtype, the prognostic impact of in-hospital left ventricular ejection fraction (LVEF) recovery remains unclear.
Purpose: The objective of this study is to evaluate the influence of in-hospital LVEF recovery on outcomes in HF-CS. Additionally, this study aims to investigate the role of patient characteristics and the use of different treatment strategies on LVEF recovery within this cohort.
Methods: In this international observational study, patients with HF-CS (e.g. caused by severe de-novo or acute-on-chronic HF; but not by acute myocardial infarction) from 16 tertiary-care centers in five countries were enrolled between 2010 and 2021. To investigate differences in patient characteristics and LVEF recovery, adjusted multivariable mixed effects linear and logistic regression models were fitted. Adjusted Cox regression models were used to evaluate the association between LVEF recovery and 30-day mortality (plus interaction term between MCS use and LVEF recovery).
Results: A total of 423 CS patients were analyzed: median age 59 (IQR 47.5-70.0) years, 296 (70.0%) male, median lactate level upon admission 4.6 (IQR 2.7-8.5) mmol/l, baseline pH 7.3 (IQR 7.2-7.4), 128 (30.5%) with prior cardiac arrest. The median baseline LVEF was 20% (IQR 15-30%), increasing to 30% (IQR 20-42.5%) at discharge. A total of 234 (55.3%) patients presented with in-hospital LVEF recovery. In comparison to patients without LVEF recovery, those with LVEF recovery were younger (57 vs. 64 years), had lower rates of known history of HF (47.0% vs. 68.8%) and hypertension (58.0% vs. 40.2%). Upon adjustment, age has a negative effect on LVEF recovery, with older age associated with lower LVEF recovery (OR 0.97, 95% CI 0.95 – 0.98, p<0.001; Beta -0.02, 95% CI -0.02 – -0.01, p<0.001). Other patient characteristics like cardiovascular comorbidities, existing HF medication, or shock severity during the index event were not different between groups. Patients with an inotropic score of 34 or higher had 41% lower odds of experiencing LVEF improvement (OR 0.59, 95% CI 0.35 – 1.00, p=0.049). Importantly, patients with improved in-hospital LVEF had a significantly lower mortality rate compared to those without LV recovery (2.7% for improved LVEF, 36.5% for patients without change in LVEF, 50.5% for worsened LVEF), with an adjusted hazard ratio for improved vs. intermediate/worsened LVEF of 0.05, 95% CI 0.03-0.15, p<0.001, Figure 1. In interaction analysis, LVEF recovery was not positively impacted by mechanical circulatory support (MCS).
Conclusions: In patients with HF-CS, in-hospital LVEF recovery was frequent and associated with lower 30-day mortality. However, it was not positively impacted by use of MCS, indicating that LVEF recovery could serve as a potential therapeutic target for outcome improvement, but not as a surrogate endpoint for efficacy in MCS trials.
