Quantitative myocardial perfusion in mice based on the signal intensity of flow sensitized CMR

Background: In the conventional approach to arterial spin labeling in the rodent heart, the relative difference in the apparent T ₁ relaxation times corresponding to selective and non-selective inversion is related to perfusion via a two compartment model of tissue. But accurate determination of T ₁ in small animal hearts is difficult and prone to errors due to long scan times and high heart rates. In this study we introduce the theoretical frame work for an alternative method (SI-method) based purely on the signal intensity of slice-select and non-select inversion recovery images at a single inversion time at short repetition time. Methods. A modified Bloch equation was solved to derive perfusion as a function of signal intensity of flow sensitized segmented gradient echo acquisitions. A two compartment fast exchanging model of tissue was assumed. To test the new technique first it was implemented on a flow phantom and then it was compared with the conventional T ₁ method in an in vivo study of healthy C57BL/6 mice (n=12). Finally the SI-method was used in comparison to a Late Gadolinium Enhanced (LGE) method to qualitatively and quantitatively assess perfusion deficits in an ischemia-reperfusion mouse model (n=4). Results: The myocardial perfusion of healthy mice obtained by the SI-method, 5.6 ± 0.5 ml/g/min, (mean ± standard deviation) was similar (p=0.38) to that obtained by the conventional method, 5.6 ± 0.3 ml/g/min. The variance in perfusion within the left ventricle was less for the SI-method than that for the conventional method (p<0.0001). The mean percentage standard deviation among repeated measures was 3.6%. The LGE regions of the ischemia reperfusion model were matched with regions of hypo-perfusion in the perfusion map. The average perfusion in the hypo perfused region among all four IR mice was 1.2 ± 0.9 ml/g/min and that of the remote region was 4.4 ± 1.2 ml/g/min. Conclusions: The proposed signal intensity based ASL method with a segmented acquisition scheme allows accurate high resolution perfusion mapping in small animals. It's short scan time, high reproducibility and ease of post process makes it a robust alternative to the conventional ASL technique that relies on T₁ measurements.