Study population

A total of 40 healthy volunteers (22 males, 18 females; mean age ± standard deviation (SD) 31.9 ± 7.9 years for males vs. 31.6 ± 11.0 years for females) were included in this study and underwent cardiac MRI and ¹H-MRS.

Cardiac imaging and spectroscopy

Double-triggered high-field cardiac MRI and ¹H-MRS were performed using a 3 T magnet (MAGNETOM Trio^®, Siemens AG Healthcare Sector, Erlangen, Germany) equipped with a 12-channel phased array coil. Subjects were scanned in supine position. All participants underwent a fasting period of at least 2 hours prior to imaging.

Conventional dynamic Steady-state free precession (SSFP) cine images in the short-axis and four-chamber views were acquired prior to cardiac ¹H-MRS to allow precise voxel placement in the desired cardiac cycle.

Double-triggered cardiac single voxel ¹H-MR spectra (voxel size 6 ml) were obtained from the interventricular septum (Fig. 1), thus avoiding contamination by epicardial fat.  Single voxel ¹H-MRS was performed at end-systole and end-expiration, using a point-resolved spectroscopy sequence (PRESS).

Spectral analysis

For each calculation two spectra (with and without water suppression) were obtained from identical voxel positions. Determination of resonance signals was supported by prior knowledge. The water signal was set to a ∂-value of 4.7 ppm. Protons in methylene groups (CH₂) appeard at 1.3 ppm, protons in methyl groups (CH₃) appeard at 0.9 ppm. The areas under the curve (AUC) of both triglyceride peaks were summed for calculation of total myocardial TG content. Relative myocardial TG was calculated by expressing the myocardial metabolite signal as a percentage of the unsuppressed water signal (myocardial TG / water resonance ratio).