Cardiovascular MRI Predicts 5-Year Adverse Clinical Outcome in Heart Transplant Recipients

Imaging recommendations for the follow-up of heart transplant recipients (HTRs) lack evidence justifying their prognostic value. Cardiovascular magnetic resonance imaging (CMRI) can characterize heart structure and function and has prognostic value in many myocardial diseases. We hypothesized that CMRI evaluation of cardiac allografts would predict adverse events. We performed CMRI on 60 HTRs evaluating biventricular size, function and myocardial scar. We performed survival analysis to identify independent predictors of cardiovascular (CV) death or hospitalization. Participants had a mean age of 51 ± 14 years, mean graft age of 3.5 years (±4) and 75% are male. Median follow-up time was 4.9 years with 22 CV hospitalizations and 7 CV deaths. A multivariable survival analysis of imaging and clinical variables identified myocardial scar (hazard ratio HR of 10.7, p = 0.005), right ventricular end- diastolic volume index (RVEDVI; 1.1/mL/m2, p = 0.001), graft age (HR = 1.2/year, p = 0.004) and previous allograft rejection (HR = 4.4, p = 0.006) as predictive of time to CV death or hospitalization. CMRI-derived myocardial scar and RVEDVI are independently associated with CV outcomes in HTRs. Imaging recommendations for the follow-up of heart transplant recipients (HTRs) lack evidence justifying their prognostic value. Cardiovascular magnetic resonance imaging (CMRI) can characterize heart structure and function and has prognostic value in many myocardial diseases. We hypothesized that CMRI evaluation of cardiac allografts would predict adverse events. We performed CMRI on 60 HTRs evaluating biventricular size, function and myocardial scar. We performed survival analysis to identify independent predictors of cardiovascular (CV) death or hospitalization. Participants had a mean age of 51 ± 14 years, mean graft age of 3.5 years (±4) and 75% are male. Median follow-up time was 4.9 years with 22 CV hospitalizations and 7 CV deaths. A multivariable survival analysis of imaging and clinical variables identified myocardial scar (hazard ratio HR of 10.7, p = 0.005), right ventricular end- diastolic volume index (RVEDVI; 1.1/mL/m2, p = 0.001), graft age (HR = 1.2/year, p = 0.004) and previous allograft rejection (HR = 4.4, p = 0.006) as predictive of time to CV death or hospitalization. CMRI-derived myocardial scar and RVEDVI are independently associated with CV outcomes in HTRs.