Abstract
Heart failure with preserved ejection fraction (HFpEF) is a growing clinical challenge, accounting
for over half of all heart failure cases and disproportionately affecting older individuals and
women. Despite its increasing prevalence, the pathophysiology of HFpEF is incompletely
understood. Emerging evidence implicates metabolic dysfunction, inflammation, and visceral
adiposity—particularly epicardial adipose tissue (EAT)—in its pathophysiology. EAT is a
metabolically active fat depot that lies in direct contact with the myocardium and shares its
microcirculation, making it uniquely positioned to influence cardiac function through vasocrine
and paracrine signaling. While prior imaging studies have emphasized EAT volume as a risk factor
in cardiovascular disease, recent focus has shifted to assessing EAT quality, including lipid density
and composition, as it may be a more specific marker of adipose tissue dysfunction in the context
of cardiovascular disease.
This dissertation presents the first comprehensive application of magnetic resonance
imaging (MRI) to characterize the fatty acid composition (FAC) of human EAT in vivo. The
overarching goal of this work is to develop, validate, and apply FAC MRI as a novel imaging tool
to evaluate the pro-inflammatory profile of EAT and its role in cardiovascular disease, particularly
HFpEF.
Aim 1 (Chapter 3) was to develop and validate a clinically translatable FAC MRI method
capable of noninvasively quantifying saturated (SFA), monounsaturated (MUFA), and
polyunsaturated fatty acid (PUFA) content within EAT. A rapid, ECG-gated, multi-echo gradient
echo acquisition and locally low-rank reconstruction approach was implemented to enable breath-
held imaging of EAT FAC at both 1.5T and 3T field strengths. The method was rigorously
validated in oil phantoms and demonstrated reproducibility in healthy volunteers, establishing its
technical feasibility and measurement accuracy.
Aim 2 (Chapter 4) was to apply FAC MRI in a cohort of patients with and without HFpEF
to evaluate the potential of EAT fatty acid composition—specifically the saturated fatty acid index
(SFAi)—as a novel imaging biomarker. In this study, elevated EAT SFAi was significantly
2
associated with HFpEF status and markers of diastolic dysfunction, independent of EAT volume.
These findings suggest that EAT composition may provide complementary and potentially
superior information to volumetric metrics in the evaluation of HFpEF and support the potential
role of FAC MRI in risk stratification and mechanistic phenotyping.
Aim 3 (Chapter 5) extended this investigation to a broader prospective cohort of patients
undergoing cardiac MRI. This analysis explored spatial, sex-based differences in EAT FAC and
then placed that analysis in the context of HFpEF. Notably, SFAi was higher in EAT adjacent to
the left ventricle than the right, females exhibited significantly higher SFAi than males. Notably,
females with HFpEF had significantly higher SFAi than females without, while males with HFpEF
had not significant difference in SFAi.
Together, these studies demonstrate that FAC MRI provides robust, noninvasive insight
into the metabolic state of EAT and its role in cardiac dysfunction. This novel imaging technique
has the potential to advance our understanding of adipose–cardiac interactions and inform
strategies for diagnosing and managing HFpEF.
