LVAD treatment has been used successfully for the management of patients in end-stage heart failure. In a limited number of patients, when associated with pharmacological therapy, this treatment leads to substantial clinical improvement (Yacoub, 2001). Cardiomyocytes isolated from biopsies taken at explantation of the LVAD in patients with clinical recovery showed changes in intracellular Ca2+ homeostasis (Terracciano et.al., 2003). However, whether these specific cellular changes are related to clinical recovery remained unclear. We therefore studied isolated left ventricular myocytes from patients at the time of LVAD implant (LVAD core, n=7 patients), from tissue taken at explantation in patients with clinical recovery (Recovery, n=6) and from patients who did not show clinical recovery, thus requiring transplantation (Transplanted, n=4). Ethical approval was obtained by the Brompton & Harefield NHS Trust Ethics Committee. Informed consent was obtained from each patient. Data are expressed as mean ±SD. Statistical differences between means were calculated with one-way ANOVA comparisons and Bonferroni post tests.Electrophysiology was performed using switch-clamping and high resistance (20-30 MΩ) microelectrodes. Compared with myocytes taken at implantation, both recovery and non recovery groups showed a similar reduction in cell capacitance (LVAD core:290±61 pF; 7; Recovery:148±28 pF; 5; Transplanted:167±65 pF; 4, p<0.01; mean ± SEM; n = patients). Action potentials from patients with recovery shortened significantly compared with myocytes from LVAD core. Such an effect was not observed in the non-recovery group (time -to-90% repolarisation LVAD core: 735±160 ms; 7; Recovery: 454 ±58 ms; 6, Transplanted: 801 ±125 ms; 4, p<0.01). L-type Ca current fast inactivation was faster in recovery compared with non-recovery (LVAD core: 16.4±3.3 ms; 7; Recovery:9.3±1.7 ms; 5; Transplanted:18±5.5 ms; 4; p<0.01) and SR Ca content was increased in the recovery group compared with both the implant and non-recovery groups (SR Ca content in LVAD core: 30.3±17 µM/l non mitochondrial volume (NMV); 7; in Recovery: 90.4±37 mM/l NMV; 5; in Transplanted: 35±15 mM/l NMV; 4; p<0.01).These findings further identify alterations in excitation-contraction coupling, and SR Ca2+ handling in particular, as key functional determinants in patients with heart failure. In contrast regression of cellular hypertrophy did not appear to influence clinical recovery.