Post-mortem cardiac magnetic resonance parameters in normal and diseased conditions

Background Post-mortem cardiac magnetic resonance (CMR) is a non-invasive alternative to conventional autopsy. At present, diagnostic guidelines for cardiovascular conditions such as hypertrophic cardiomyopathy have not been established. We correlated post-mortem CMR images to definite conventional autopsy findings and hypothesed that elevated T2-weighted signal intensity and RV to LV area ratios can identify myocardial infarction and pulmonary emboli respectively. Methods For this unblinded pilot sub-study, we selected cases from the original blinded study that compared post-mortem imaging to conventional autopsy in patients referred for coronial investigation between October 2014 to November 2016. Three groups of scans were selected based on the cause of death identified by conventional autopsy: non-cardiovascular causes of death with no structural cardiac abnormality i.e., control cases, acute/subacute myocardial infarction and pulmonary emboli. Left ventricular (LV) wall thickness, LV myocardial signal intensity and ventricular cavity areas were measured. Results Fifty-six scans were selected [39 (69.6%) males]: 37 (66.1%) controls, eight (14.3%) acute/subacute myocardial infarction and eleven (19.6%) pulmonary emboli. The median age was 61 years [Interquartile range (IQR) 50-73] and the median time from death to imaging and autopsy was 2 days (IQR 2-3) and 3 days (IQR 3-4). The septal and lateral walls were thicker {15 mm [13-17] and 15 mm [14-18]} on post-mortem CMR than published ante-mortem measurements. Areas of acute/subacute myocardial infarction had significantly higher T2-weighted signal intensity (normalised to skeletal muscle) compared to normal myocardium in those who died from other causes {2.5 [2.3-3.0.] vs. 1.9 [1.8-2.3]; P<0.001}. In cases with pulmonary emboli, there was definite RV enlargement with a larger indexed RV to LV area ratio compared to those who died from other causes {2.9 [2.5-3.0] vs. 1.8 [1.5-2.0]; P<0.001}. Conclusions We present potential post-mortem CMR parameters to identify important cardiovascular abnormalities that may be beneficial when conventional autopsy cannot be performed. In patients without cardiovascular disease, LV wall thickness was found to be unreliable in diagnosing hypertrophic cardiomyopathy without histological and/or genetic testing. Elevated T2 signal intensity and RV to LV area ratios may be useful markers for acute/subacute myocardial infarction and pulmonary emboli. Larger studies will be necessary to define cut-offs.