Myocardial steatosis impairs left ventricular diastolic-systolic coupling in healthy humans

Abstract

Mounting evidence suggests that myocardial steatosis contributes to left ventricular diastolic dysfunction, but definitive evidence in humans is lacking due to confounding comorbidities. As such, we utilized a 48-h food restriction model to acutely increase myocardial triglyceride (mTG) content - measured by ¹ H magnetic resonance spectroscopy - in 27 young healthy volunteers (13 men/14 women). Forty-eight hours of fasting caused a more than 3-fold increase in mTG content (P < 0.001). Diastolic function - defined as early diastolic circumferential strain rate (CSRd) - was unchanged following the 48-h fasting intervention, but systolic circumferential strain rate was elevated (P < 0.001), indicative of systolic-diastolic uncoupling. Indeed, in a separate control experiment in 10 individuals, administration of low-dose dobutamine (2 μg/kg/min) caused a similar change in systolic circumferential strain rate as was found during 48 h of food restriction, along with a proportionate increase in CSRd, such that the two metrics remained coupled. Taken together, these data indicate that myocardial steatosis contributes to diastolic dysfunction by impairing diastolic-systolic coupling in healthy adults, and suggest that steatosis may contribute to the progression of heart disease. KEY POINTS: Preclinical evidence strongly suggests that myocardial lipid accumulation (termed steatosis) is an important mechanism driving heart disease. Definitive evidence in humans is limited due to the confounding influence of multiple underlying comorbidities. Using a 48-h food restriction model to acutely increase myocardial triglyceride content in young healthy volunteers, we demonstrate an association between myocardial steatosis and left ventricular diastolic dysfunction. These data advance the hypothesis that myocardial steatosis may contribute to diastolic dysfunction and suggest myocardial steatosis as a putative therapeutic target.

Keywords: coupling; diastole; lipotoxicity; relaxation; steatosis.

Figure 1.. Myocardial Triglyceride Content.

(A) Representative cardiac ¹H magnetic resonance spectra, collected from a single volume of interest positioned over the interventricular septum using two end-systolic cardiac cine images along the long and short axis. Representative changes in myocardial triglyceride content (mTG) are also illustrated from a single individual transitioning from baseline to 48 hours fasting to recovery. RA, right atrium; LA, left atrium; and RV, right ventricle; LV, left ventricle. (B) Changes in myocardial triglyceride content (mTG) after a 48-hour fast. Individual data superimposed on Mean ± SD. A linear mixed model for repeated measures was used to make statistical comparisons. n = 27. Main effect p<0.001, f=1.02.

Figure 2.. Impact of low dose (2 μg/kg/min) intravenous dobutamine on left ventricular systolic and diastolic coupling.

(A) A representative tissue tagging image taken from the middle of the left ventricle along the short axis is shown to reflect the imaging approach used to acquire these data. Dobutamine infusion significantly increased (B) left ventricular systolic circumferential strain rate (Cohen’s d = −2.58) and (C) left ventricular early diastolic circumferential strain rate (Cohen’s d = 1.54), such that (D) the ratio between the two components remained unchanged. Individual data shown alongside Mean ± SD. Statistical comparisons were made using dependent samples t-tests. n = 10.

Figure 3.. Myocardial steatosis impairs left ventricular diastolic-systolic coupling.

(A) A representative tissue tagging image taken from the middle of the left ventricle along the short axis is shown to reflect the imaging approach used to acquire these data. (B) Left ventricular early diastolic circumferential strain rate (CSRd) normalized for the preceding systolic circumferential strain rate (CSRs). Individual data superimposed on Mean ± SD. A linear mixed model for repeated measures was used to make statistical comparisons. n = 27. Main effect p<0.001, f=0.76.

Figure 4.. Diastolic function is impaired after 48-hours of fasting but does not appear to be sexually dimorphic.

(A) A representative tissue tagging image taken from the middle of the left ventricle along the short axis is shown. (B) Left ventricular early diastolic circumferential strain rate (CSRd) normalized for the preceding systolic circumferential strain rate (CSRs). Individual data superimposed on Mean ± SD. A linear mixed model for repeated measures was used to make statistical comparisons (n = 13M/14W).