Abstract
AIMS: Cardiac allograft vasculopathy (CAV) is a significant complication that contributes to both morbidity and mortality after heart transplantation. The aim of this study was to (i) assess if quantitative cardiovascular magnetic resonance (CMR) myocardial perfusion could detect different stages of CAV and (ii) establish a myocardial perfusion reserve (MPR) cut-off for significant CAV. METHODS AND RESULTS: Patients with a heart transplant who had performed a clinical CMR scan and invasive angiography at two centres in Sweden were included in the study (n = 110). Quantitative short-axis perfusion maps were acquired using single-bolus gadolinium contrast, dual-sequence perfusion imaging at rest and during stress. Global myocardial perfusion (MP) was averaged across all segments at rest and stress and MPR was defined as the ratio between stress and rest MP. All invasive angiographies were reported according to the International Heart and Lung Transplantation CAV classification. Patients were classified as follows: 53% (58/110) as CAV0, 38% (42/110) as CAV1, and 9% (10/110) as CAV2-3. There was a gradual decrease of stress MP and MPR with increased CAV grade. The MPR could discriminate CAV2-3 with an area under the curve-receiver operating characteristic of 0.88, 95% confidence interval 0.78-0.98, and using a cut-off of 2.2, the sensitivity was 100%, specificity was 68%, and positive and negative predictive values were 21 and 100%. CONCLUSION: In this multi-centre retrospective study, MPR assessed by CMR could discriminate CAV2-3 with both high sensitivity and negative predictive value and a cut-off of MPR 2.2 is suggested.