31P magnetic resonance spectroscopy in dilated cardiomyopathy and coronary artery disease. Altered cardiac high-energy phosphate metabolism in heart failure.

BACKGROUND: The purpose of this work was to further define the value of cardiac 31P magnetic resonance (MR) spectroscopy for patients with coronary artery disease and dilated cardiomyopathy. METHODS AND RESULTS: Blood-corrected and T1-corrected 31P MR spectra of anteroseptal myocardium were obtained at rest using image-selected in vivo spectroscopy localization, a selected volume of 85 +/- 12 cm3, and a field strength of 1.5 T. Nineteen volunteers had a creatine phosphate (CP)/ATP ratio of 1.95 +/- 0.45 (mean +/- SD) and a PDE/ATP ratio of 1.06 +/- 0.53; in four patients with left anterior descending coronary artery (LAD) stenosis, six patients with chronic anterior wall infarction, and four patients with chronic posterior wall infarction, CP/ATP and phosphodiester (PDE)/ATP ratios did not differ from those in volunteers. Twenty-five measurements of 19 patients with dilated cardiomyopathy yielded a CP/ATP of 1.78 +/- 0.51 and a PDE/ATP of 0.98 +/- 0.56 (p = NS versus volunteers). When these patients were grouped according to the severity of heart failure, however, CP/ATP was 1.94 +/- 0.43 in mild (p = NS versus volunteers) and 1.44 +/- 0.52 in severe DCM (p < 0.05), respectively. No correlation was found between CP/ATP and left ventricular ejection fraction or fractional shortening, but correlation of CP/ATP with the New York Heart Association (NYHA) class was significant (r = 0.60, p < 0.005). Six patients with dilated cardiomyopathy were studied repeatedly before and after 12 +/- 6 weeks of drug treatment leading to clinical recompensation with improvement of the NYHA status by 0.8 +/- 0.3 classes. Concomitantly, CP/ATP increased from 1.51 +/- 0.32 to 2.15 +/- 0.27 (p < 0.01), whereas PDE/ATP did not change significantly. CONCLUSIONS: Cardiac high-energy phosphate metabolism at rest is normal in LAD stenosis and chronic myocardial infarction in the absence of heart failure. The CP/ATP ratio has low specificity for the diagnosis of dilated cardiomyopathy. However, CP/ATP correlated with the clinical severity of heart failure and may improve during clinical recompensation.

[The direct measurement of the spin-grid-relaxation times of phosphorus metabolites in the human myocardium]. / Direkte Messung der Spin-Gitter-Relaxationszeiten von Phosphormetaboliten im menschlichen Myokard.

The T1 relaxation times of the phosphorus metabolites in human heart muscle measurable by 31P-MR spectra were determined in 12 individuals using a 1.5 Tesla system. Several spectra were recorded consecutively with a pulse repetition time of 1.6 s to 24 s. The T1 times of creatine phosphate (CP), of gamma-, alpha-, beta-adenosintriphosphate (ATP), 2,3-diphosphoglycerate (2,3-DPG) together with anorganic phosphate) and phosphodiester (PDE) showed mean measurements of 6.1 +/- 0.5, 5.4 +/- 0.5, 5.0 +/- 0.5, 5.8 +/- 1.0, 7.6 +/- 1.0, and 5.0 +/- 1.0 s (M +/- SE). The accuracy of the ISIS technique was tested with a special phantom. T1 times were also measured in standard solutions (20 mM CP, 10 mM ATP); CP was 8.7 +/- 0.2 s and gamma-ATP was 9.9 +/- 0.7 s. Corrections for partially saturated 31P-MR spectra--at least for CP/ATP ratios--are relatively small.

Direct measurement of spin-lattice relaxation times of phosphorus metabolites in human myocardium.

T1 values of phosphorus metabolites visible in human cardiac 31P-MR spectra were determined in 12 volunteers at 1.5 T. Consecutive spectra were acquired with varying pulse repetition time (TR) from 1.6 to 24 s; volume selection was achieved with ISIS. T1's of creatine phosphate (CP), [gamma-P], [alpha-P], and [beta-P]ATP, 2-3 diphosphoglycerate, and phosphodiesters were 6.1 +/- 0.5, 5.4 +/- 0.5, 5.5 +/- 0.5, 5.8 +/- 1.0, 7.6 +/- 1.0, and 5.0 +/- 1.0 s, respectively. CP/ATP ratios showed little change with varying TR; linear regression of CP/ATP vs TR was of borderline significance (r = 0.28, P = 0.06). T1's for CP and ATP were also determined in standard solution (20 mM CP, 10 mM ATP) yielding T1CP of 8.7 +/- 0.2 and T1[gamma-P]-ATP of 9.9 +/- 0.7 s. Thus, T1's for CP and ATP were similar at 1.5 T in both human heart and standard solution. In human cardiac 31P-MR spectra, CP/ATP ratios may need little correction for partial saturation.