Abstract
BACKGROUND AND AIMS: Heart failure with preserved ejection fraction (HFpEF) is increasingly recognized as a syndrome of reserve dysfunction. However, integrated assessment of biventricular (LV/RV) volumetric reserve under physiological stress remains underexplored. We aimed to investigate whether exercise cardiac magnetic resonance (Ex-CMR) can reveal distinct volumetric reserve profiles across the HFpEF spectrum.
METHODS: In this retrospective analysis of a prospective observational, multicentre study, supine ergometer Ex-CMR was performed in HFpEF patients across early to advanced stages (stage B, exercise-induced, stage C), along with healthy controls and a non-cardiac dyspnoea (NCD) group. Percentage changes in LV/RV end-diastolic (ΔEDV%) and end-systolic volumes (ΔESV%) from rest to stress defined EDV reserve and ESV reserve, respectively. Ventricular efficiency index (EI) was defined as ΔEDV%-ΔESV%; biventricular EI as LVEI + RVEI. Group comparisons were performed using ANOVA and post hoc testing. Multivariable general linear model analyses adjusted for age, sex, BMI, and exercise response. A composite phenotyping assessment incorporating all four key reserve parameters was explored.
RESULTS: Among 140 participants (40 healthy, 27 NCD, and 73 HFpEF), all HFpEF subgroups showed impaired LVEDV reserve and reduced LVEI (P < .0001). LVESV reserve was impaired only in stage C (P < .0001). Exercise-induced RV dysfunction was a hallmark of HFpEF with pulmonary hypertension (P < .0001). Biventricular EI declined progressively with advancing HFpEF stage (P < .0001) and was significantly lower in NYHA > II (P = .0006). Six distinct reserve phenotypes emerged.
CONCLUSION: Ex-CMR-based assessment of LV/RV volumetric reserve reveals progressive biventricular dysfunction across HFpEF stages and supports biventricular volumetric reserve-based phenotyping for characterizing HFpEF pathophysiology.
KEY QUESTION: Can the integration of left and right ventricular end-diastolic and end-systolic volume reserve under physiological stress reveal distinct profiles across the HFpEF spectrum and enhance our understanding of its haemodynamic heterogeneity?
KEY FINDINGS: Non-invasive assessment of biventricular volumetric reserve, along with their intra- and interventricular interactions using exercise CMR, revealed a significant, stepwise deterioration across HFpEF subgroups, worsening with NYHA class > II. Exercise CMR enabled composite volumetric reserve-based phenotyping and identified six distinct reserve phenotypes.
TAKE HOME MESSAGE: Biventricular volumetric reserve assessment is feasible through exercise CMR and may support future precision therapy strategies.