BACKGROUND: (31)P magnetic resonance spectroscopy (MRS) allows measurement of in vivo high-energy phosphate kinetics in the myocardium. While traditionally (31)P cardiac spectroscopy is performed at 1.5T, cardiac MRS at higher field strength can theoretically increase signal to noise ratio (SNR) and spectral resolution therefore improving sensitivity and specificity of the cardiac spectra. The reproducibility and feasibility of performing cardiac spectroscopy at 3T is presented here in this study in healthy volunteers and patients with hypertrophic cardiomyopathy. METHODS: Cardiac spectroscopy was performed using a Phillips 3T Achieva scanner in 37 healthy volunteers and 26 patients with hypertrophic cardiomyopathy (HCM) to test the feasibility of the protocol. To test the reproducibility a single volunteer was scanned eight times on separate occasions. A single voxel (31)P MRS was performed using Image Selected In vivo Spectroscopy (ISIS) volume localization. RESULTS: The mean phosphocreatine/adenosine triphosphate (PCr/ATP) ratio of the eight measurements performed on one individual was 2.11+/-0.25. Bland Altman plots showed a variance of 12% in the measurement of PCr/ATP ratios. The PCr/ATP ratio was significantly reduced in HCM patients compared to controls, 1.42+/-0.51 and 2.11+/-0.57, respectively, P<0.0001. (All results are expressed as mean+/-standard deviation). CONCLUSIONS: Here we demonstrate that cardiac (31)P MRS at 3T is a reliable method of measuring in vivo high-energy phosphate kinetics in the myocardium for clinical studies and diagnostics. Based on our data an impairment of cardiac energetic state in patients with hypertrophic cardiomyopathy is indisputable.