Abstract Body (Do not enter title and authors here): BACKGROUND:
Diffuse myocardial fibrosis contributes to diastolic dysfunction and may impair exercise capacity in patients with heart failure and preserved ejection fraction (HFpEF). However, its impact on exercise intolerance in obese versus non-obese HFpEF phenotypes remains unclear.

METHODS:
We retrospectively identified 38 HFpEF patients (26 non-obese individuals [mean age 70 years, 13 female]; 12 obese patients [mean age 69 years, 8 female]) undergoing 3T cardiovascular magnetic resonance (CMR), including triple-slice T₁ mapping by modified look-locker inversion recovery sequence, and cardiopulmonary exercise testing (CPET) between April 2013 and March 2025. The HFpEF diagnosis followed current consensus statements at various points during the study. Extracellular volume (ECV) was quantified, excluding infarcted segments. In a subgroup of patients, left ventricular (LV) chamber stiffness was also measured invasively by evaluating the LV end-diastolic pressure-volume relationship during preload manipulation.

RESULTS:
The median interval between CMR and CPET was 22 days. There were no significant differences in CMR-ECV or exercise capacity between non-obese and obese patients (ECV: 32.0±3.2 vs. 32.8±3.9%, p=0.56; peak oxygen consumption: 19.3±5.7 vs. 17.3±5.7 ml/min/kg, p=0.32). Among non-obese individuals, ECV demonstrated a significant inverse correlation with peak oxygen consumption (r=-0.52, p=0.007), and this relationship remained significant after adjusting for age; however, it was not observed in obese patients (r=-0.06, p=0.84). In a subset of 11 non-obese patients, ECV showed a positive correlation with LV stiffness, as assessed by invasive pressure-volume curve analysis (r=0.67, p=0.048). Notably, non-obese patients with lower ECV (below the cohort median of 31.3%) exhibited relatively preserved exercise capacity, in contrast to non-obese individuals with higher ECV and obese patients, who demonstrated reduced functional capacity (peak oxygen consumption: 21.9 ± 4.8 vs. 16.3 ± 5.2, 17.5 ± 7.3, and 17.1 ± 4.2 ml/min/kg, respectively; ANOVA p = 0.06).

CONCLUSIONS:
This analysis revealed distinct pathophysiological mechanisms underlying exercise intolerance in non-obese versus obese HFpEF patients, underscoring the need for HFpEF phenotype-specific management strategies. In the non-obese group, ECV was a key factor of exercise intolerance, suggesting that fibrosis-targeting therapies may be a viable and potentially beneficial option for this subgroup.