In vivo chronic myocardial infarction characterization by spin locked cardiovascular magnetic resonance

Background: Late gadolinium enhanced (LGE) cardiovascular magnetic resonance (CMR) is frequently used to evaluate myocardial viability, estimate total infarct size and transmurality, but is not always straightforward is and contraindicated in patients with renal failure because of the risk of nephrogenic systemic fibrosis. T2- and T1-weighted CMR alone is however relatively insensitive to chronic myocardial infarction (MI) in the absence of a contrast agent. The objective of this manuscript is to explore T1 rho-weighted rotating frame CMR techniques for infarct characterization without contrast agents. We hypothesize that T1 rho MR accurately measures infarct size in chronic MI on account of a large change in T1 rho relaxation time between scar and myocardium. Methods: 7Yorkshire swine underwent CMR at 8 weeks post-surgical induction of apical or posterolateral myocardial infarction. Late gadolinium enhanced and T1 rho CMR were performed at high resolution to visualize MI. T1 rho-weighted imaging was performed with a B-1 = 500 Hz spin lock pulse on a 3 T clinical MR scanner. Following sacrifice, the heart was excised and infarct size was calculated by optical planimetry. Infarct size was calculated for all three methods (LGE, T1 rho and planimetry) and statistical analysis was performed. T1 rho relaxation time maps were computed from multiple T1 rho-weighted images at varying spin lock duration. Results: Mean infarct contrast-to-noise ratio (CNR) in LGE and T1 rho CMR was 2.8 +/- 0.1 and 2.7 +/- 0.1. The variation in signal intensity of tissues was found to be, in order of decreasing signal intensity, LV blood, fat and edema, infarct and healthy myocardium. Infarct size measured by T1 rho CMR (21.1% +/- 1.4%) was not significantly different from LGE CMR (22.2% +/- 1.5%) or planimetry (21.1% +/- 2.7%; p < 0.05). T1 rho relaxation times were T1 rho(infarct) = 91.7 ms in the infarct and T1 rho(remote) = 47.2 ms in the remote myocardium. Conclusions: T1 rho-weighted imaging using long spin locking pulses enables high discrimination between infarct and myocardium. T1 rho CMR may be useful to visualizing MI without the need for exogenous contrast agents for a wide range of clinical cardiac applications such as to distinguish edema and scar tissue and tissue characterization of myocarditis and ventricular fibrosis.