Exercise cardiac magnetic resonance imaging reveals persistent impairment of dynamic left ventricular systolic function despite sympathetic recovery after heart transplantation

Background Cardiac denervation is a key clinical feature in patients after heart transplantation (HTx). Impaired heart rate (HR) response may contribute to an exercise intolerance. Evaluating dynamic cardiac function may provide further insight into the pathophysiology in HTx recipients. Purpose To investigate the relationship between sympathetic nerve activity (SNA) and dynamic cardiac function in HTx recipients using exercise cardiac magnetic resonance imaging (exCMR). Methods ExCMR was performed in 8 consecutive males (44 ± 12 years-old) who underwent HTx between April 2024 and October 2024 at 6 and 12 months after HTx (HTx6m and HTx12m), and in 8 healthy males (36 ± 5 years-old) as Controls. Cardiopulmonary exercise test (CPX) and 123I-MIBG scintigraphy were performed within two weeks of exCMR. Short-axis cine views were scanned at rest during exercise at 40% MWR while subjects pedaling in a supine position. Images using cvi42 provided left ventricular end-diastolic volume index (LVEDVI) and end-systolic volume index (LVESVI). Left ventricular stroke volume index (LVSVI) was calculated as LVEDVI − LVESVI, and cardiac index (CI) was calculated as HR × LVSVI. Data are presented as mean ± SD. Statistical analysis was performed using two-way repeated measures ANOVA with post-hoc Tukey’s multiple comparison or paired t-test. Results Compared with HTx6m, HTx12m showed improvement in %peak VO2 (63 ± 7 vs. 55 ± 10%, p=0.026) and ΔHR/ΔWR (0.40 ± 0.13 vs. 0.27 ± 0.09 bpm/watt, p<0.001), indicating recovery of exercise tolerance and HR response. 123I-MIBG scintigraphy showed increased heart-to-mediastinum (H/M) ratios (early: 1.82 ± 0.21 vs. 1.55 ± 0.20, p=0.006, late: 1.47 ± 0.14 vs. 1.36 ± 0.18, p=0.039), suggesting recovery of SNA at HTx12m compared with HTx6m. ExCMR showed higher HR at rest and poor HR response during exercise in HTx compared to Controls (p<0.0001, Figure A). Exercise did not change LVEDVI in Controls but increased in both HTx6m and HTx12m (p=0.0129, Figure B). Exercise decreased LVESVI in Controls but remained unchanged in both HTx6m and HTx12m (p<0.0001, Figure C). Although LVSVI was lower in HTx at rest, there was no difference in LVSVI augmentation during exercise (p=0.1412, Figure D). No difference was observed in resting LVEF and CI, but significant reduction of LVEF and CI response to exercise was seen in both HTx6m and HTx12m (p=0.0002, Figure E, p=0.0005, Figure F). Conclusion Poor SNA, HR response and exercise tolerance together with impaired dynamic LV systolic function was observed in HTx6m compared to Controls. Although a certain degree of recovery in SNA, HR response and exercise tolerance was observed in HTx12m compared to HTx6m, exCMR showed no improvement of dynamic LV systolic function. These findings suggest that factors beyond SNA may contributed to impaired dynamic dysfunction in HTx recipients. Further investigation with longer follow-up period is required to confirm the results.