Correcting human heart ³¹P NMR spectra for partial saturation. Evidence that saturation factors for PCr/ATP are homogeneous in normal and disease states

Heart PCr/ATP ratios measured from spatially localized ³¹P NMR spectra can be corrected for partial saturation effects using saturation factors derived from unlocalized chest surface-coil spectra acquired at the heart rate and approximate Ernst angle for phosphor creatine (PCr) and again under fully relaxed conditions during each ³¹P exam. To validate this approach in studies of normal and disease states where the possibility of heterogeneity in metabolite T₁ values between both chest muscle and heart and normal and disease states exists, the properties of saturation factors for metabolite ratios were investigated theoretically under conditions applicable in typical cardiac spectroscopy exams and empirically using data from 82 cardiac ³¹P exams in six study groups comprising normal controls (n = 19) and patients with dilated (n = 20) and hypertrophic (n = 5) cardiomyopathy, coronary artery disease (n = 16), heart transplants (n = 19), and valvular heart disease (n = 3). When T_R ≪ T₁,(PCr), with T₁(PCr) ≥ T₁(ATP), the saturation factor for PCr/ATP lies in the range 1.5 ± 0.5, regardless of the T₁ values. The precise value depends on the ratio of metabolite T₁ values rather than their absolute values and is insensitive to modest changes in T_R. Published data suggest that the metabolite T₁ ratio is the same in heart and muscle. Our empirical data reveal that the saturation factors do not vary significantly with disease state, nor with the relative fractions of muscle and heart contributing to the chest surface-coil spectra. Also, the corrected myocardial PCr/ATP ratios in each normal or disease state bear no correlation with the corresponding saturation factors nor the fraction of muscle in the unlocalized chest spectra. However, application of the saturation correction (mean value, 1.36 ± 0.03 SE) significantly reduced scatter in myocardial PCr/ATP data by 14 ± 11% (SD) (p ≤ 0.05). The findings suggest that the relative T₁ values of PCr and ATP are substantially the same in chest and heart muscle and are unchanged in these disease states. The mean value for T₁ (PCr)/T₁(ATP) is 2.16 ± 0.29 SE.