Quantitative cardiovascular magnetic resonance perfusion imaging: inter-study reproducibility

Abstract

To evaluate the inter-study reproducibility of quantitative cardiovascular magnetic resonance (CMR) myocardial perfusion imaging and the influence of diurnal variation on perfusion. Data on these are limited, despite being crucially important for performing serial examinations both in clinical practice and in trials. Sixteen healthy volunteers underwent high-resolution 3 T perfusion imaging three times during a single day to evaluate inter-study reproducibility and the effects of diurnal variation. Absolute perfusion was determined in each coronary artery territory and globally by Fermi constrained deconvolution of myocardial signal intensity curves. Left ventricular (LV) volumes and function were also calculated. Eleven full data sets were suitable for quantitative perfusion analysis. Global rest and stress perfusion and myocardial perfusion reserve (MPR) were 0.6 ± 0.1 and 2.5 ± 0.5 mL/min/g and 4.3 ± 0.9, respectively, for the first scan and were 0.5 ± 0.2 and 2.1 ± 0.5 mL/min/g and 4.2 ± 1.2 for the second (P= 0.1, 0.19, and 0.37, respectively). Inter-study reproducibility was moderate. The coefficient of variation (CV) was 16.0, 26.8, and 23.9% for global rest and stress perfusion and MPR, respectively. The corresponding territorial CVs were 27.5, 35.2, and 33.5%. The reproducibility of LV volumes and function was excellent (CV 4, 7.7, and 4.6% for end-diastolic volume, end-systolic volume, and ejection fraction, respectively). There were no significant detectable diurnal variations in perfusion or LV volumes and function (P≥ 0.05 for all). The inter-study reproducibility of quantitative myocardial perfusion is reasonable and best for global rest perfusion. No significant diurnal variation in perfusion was observed.