³¹P MRS detects signals from endogenous phosphorus-containing metabolites (e.g. adenosine triphosphate, phosphocreatine), which noninvasively reflect the in vivo energy metabolism without using ionizing radiation. It has been applied in clinical studies of cardiac energetics under different pathophysiological conditions, e.g. heart failure, ischemic heart disease and cardiomyopathies[1]. To date, few studies have focused on valvular heart diseases[2]. We evaluated myocardial energetics in patients with degenerative mitral regurgitation (DMR) using ³¹P MRS.Thirty patients with moderate or severe DMR due to flail/prolapse and ten healthy controls were enrolled. ³¹P MRS was performed on a 3T whole-body MR scanner (Magnetom Trio Tim, Siemens AG, Germany) using a ³¹P/¹H dual tuned surface coil (RAPID Biomedical GmbH, Germany). The participants lay supine, with the center of the coil over the subject’s heart and at the isocenter of the magnet. Localization images were obtained to check the position of the coil and heart. Three-dimensional chemical shift imaging (3D CSI) protocol was applied with the following parameters: TR = 800 ms, TE = 2.3 ms, average = 10, flip angle=90˚, field of view = 200*200*200 mm, matrix size = 11*11*11 (interpolated to 16*16*16), bandwidth = 3000 Hz, vector size = 1024, with the use of nuclear Overhauser enhancement. The voxels at the center of the anterior-posterior axis of the interventricular septum were selected for analysis. The spectra were first analyzed using Siemens spectroscopy package. Before Fourier transformation, the FIDs were filtered exponentially with a width of 50 ms and zero filled to 4096 data points. The resulting spectra were manually corrected for phase and baseline distortions, and referenced to phosphocreatine (PCr) at 0 ppm. The spectra were fitted for the signals of adenosine triphosphate (ATP) (including α-ATP, β-ATP, γ-ATP), PCr, inorganic phosphate (Pi), glycerophosphocholine (GPC), phosphoethanolamine (PE) and glycerophosphoethanolamine (GPE). The spectra were also fitted by Java-based version of the magnetic resonance user interface (jMRUI)[3,4], using the AMARES (advanced method for accurate, robust and efficient spectral fitting) algorithm[5] for fitting of ATP, PCr, Pi, phosphomonoester (PME), and phosphodiester (PDE), along with prior knowledge on the resonating frequencies and the J-couplings. The integral of each metabolite was normalized to the summed integrals of all phosphorus-containing metabolites (TP), which represented the relative concentration of each metabolite, e.g. PCr/TP represents the relative concentration of PCr. The ratio of the integrals from different metabolites were calculated, e.g. PCr to γ-ATP, PCr to Pi. Myocardial pH was calculated based on the chemical shift difference between PCr and Pi[6]. Data are expressed as mean ± standard deviation (SD). Two tailed Student’s t-test was used for the statistical comparisons.Figure 1 showed the location of a representative voxel selected in the centre of the anterior-posterior axis of the ventricular septum from a patient. The spectrum corresponding to the selected voxel was shown in Siemens spectroscopy package along with the fitted spectrum (Figure 2). Table 1 summarized the results of PCr/γ-ATP, PCr/Pi, relative concentration of all observed metabolites (i.e. PCr, ATP, PE, GPC, GPE, Pi), and pH in patients and control subjects, analyzed by Siemens spectroscopy package. None of the parameters showed significant difference between patients and controls. Similar results were also obtained from jMRUI analysis.We found no significant differences in the myocardial PCr/γ-ATP ratio, PCr/Pi ratio, pH and relative concentrations of phosphorus-containing metabolites between patients and controls, indicating that cardiac energetics was not significantly altered in our patients with DMR. These results were similar whether the Siemens spectroscopy package or jMRUI was used.