Abstract 1373: Site Specific Correlation of Pericardial Fat With Cardiac Function in Metabolic Syndrome
The total volume of pericardial fat has been associated with increased coronary artery plaque calcification and reduced left ventricular (LV) function; however, it is not known whether these adverse cardiovascular observations are related to a systemic process or local anatomic effects. We hypothesized that fat volumes overlying the left and right ventricles (RV) would be differentially correlated to ventricular function in a site-specific manner.
Methods: Cardiovascular magnetic resonance imaging was performed in obese subjects (n=40, BMI range=30–47 kg/m2, age range=19–62 y, 93% female) with metabolic syndrome but no known cardiovascular disease. Pericardial fat volume was quantified using axial T1 black blood images and was normalized to body surface area. The interventricular septum was used to demarcate fat overlying the respective ventricles (LV fat and RV fat). Stroke volume (SV), cardiac output (CO), and ejection fraction (EF) were determined from short-axis cine images for each chamber. In addition, LV diastolic function was assessed from the first derivative of the volume-time curve, yielding Early filling rate (E), Late filling rate (A), and the Early/Late ratio (E/A). Fasting lipid, glucose and insulin levels were measured.
Results: In this cohort of obese subjects, total pericardial (p=0.63), LV (p=0.26), and RV (p=0.76) fat did not correlate with BMI. LV SV (r=−0.33, p=0.04), CO (r=−0.31, p=0.05) but not LVEF (p=0.26) were inversely related to LV fat but not RV fat. In contrast, RV SV and CO were not related to either LV or RV fat (all p=ns). LV E-rate(r=−0.49, p=0.01), but not A-rate or E/A ratio were inversely related to LV but not RV fat. Total pericardial (r=0.41, p=0.009), and LV(r=0.44, p=0.004) fat were related to plasma triglycerides but not to cholesterol, glucose, insulin, or HOMA levels (all p=ns).
Conclusions: Pericardial fat appears related to left and right ventricular function in a site specific manner irrespective of insulin resistance. In addition, LV diastolic filling appears correlated to pericardial fat overlying the left ventricle, perhaps explaining the observed relationship of pericardial fat to LV stroke volume. These findings suggest a local anatomic and/or paracrine mechanism that merits further examination.