In heart failure, left ventricular assist device (LVAD) implantation is performed to ensure sufficient cardiac output. Whereas some patients are subsequently weaned from LVAD support, other patients still need heart transplantation. To elucidate underlying mechanisms, we assessed the arrhythmogenic SR-Ca(2+) leak at the time of LVAD implantation (HF-Im) and heart transplantation (HF-Tx) and evaluated the effects of CaMKII-inhibition. Human left-ventricular cardiomyocytes were isolated, paced at 1 Hz for 10 beats to ensure SR-Ca(2+) loading and scanned for diastolic Ca(2+) sparks (confocal microscopy). In HF-Im, the high diastolic spark frequency (CaSpF) of 0.76 ± 0.12 × 100 μm(-1) × s(-1) could be reduced to 0.48 ± 0.10 × 100 μm(-1) × s(-1) by CaMKII inhibition (AIP, 1 μM). The amplitude of Ca(2+) sparks, width, and length was not significantly altered. In sum, CaMKII inhibition yielded a clear tendency toward a reduction of the SR-Ca(2+) leak (n cells/patients = 76/6 vs. 108/6, P = 0.08). In HF-Tx, we detected an even higher CaSpF of 1.00 ± 0.10 100 μm(-1) × s(-1) and a higher SR-Ca(2+) leak compared with HF-Im (increase by 81 ± 33%, n cells/patients = 156/7 vs. 130/7, P < 0.05), which fits to the further decreased LV function. Here, CaMKII inhibition likewise reduced CaSpF (0.35 ± 0.09 100 μm(-1) × s(-1,) P = 0.06) and significantly reduced spark duration (n sparks/patients = 58/3 vs. 159/3, P < 0.05). Conclusively, the SR-Ca(2+) leak was reduced by 69 ± 12% in HF-Tx upon CaMKII inhibition (n cells/patients = 53/3 vs. 91/3, P < 0.05). These data show that the SR-Ca(2+) leak correlates with the development of LV function after LVAD implantation and may represent an important pathomechanism. The fact that CaMKII inhibition reduces the SR-Ca(2+) leak in HF-Tx suggests that CaMKII inhibition may be a promising option to beneficially influence clinical course after LVAD implantation.