Background: LVAD speed adjustment according to a functioning aortic valve has hypothetic advantages but could lead to submaximal support. The consequences of an open aortic valve policy on exercise capacity and hemodynamics have not yet been investigated systematically. Methods: Ambulatory patients under LVAD support (INCOR r, Berlin Heart, mean support time 465 +/- 7 days, average flow 4.0 +/- 0.3 L/ min) adjusted to maintain a near normal aortic valve function underwent maximal cardiopulmonary exercise testing (CPET) and right heart catheterization (RHC) at rest and during constant work rate exercise (20 Watt). Results: Although patients (n = 8, mean age 45 +/- 13 years) were in NYHA class 2, maximum work-load and peak oxygen uptake on CPET were markedly reduced with 69 +/- 13 Watts (35% predicted) and 12 +/- 2 mL/ min/ kg (38% predicted), respectively. All patients showed a typical cardiac limitation pattern and severe ventilatory inefficiency with a slope of ventilation to carbon dioxide output of 42 +/- 12. On RHC, patients showed an exercise-induced increase of mean pulmonary artery pressure (from 16 +/- 2.4 to 27 +/- 2.8 mmHg, p < 0.001), pulmonary artery wedge pressure (from 9 3.3 to 17 5.3 mmHg, p = 0.01), and cardiac output (from 4.7 +/- 0.5 to 6.2 +/- 1.0 L/ min, p = 0.008) with a corresponding slight increase of pulmonary vascular resistance (from 117 35.4 to 125 +/- 35.1 dyn* sec* cm-5, p = 0.58) and a decrease of mixed venous oxygen saturation (from 58 +/- 6 to 32 +/- 9%, p < 0.001). Conclusion: An open aortic valve strategy leads to impaired exercise capacity and hemodynamics, which is not reflected by NYHA-class. Unknown compensatory mechanisms can be suspected. Further studies comparing higher vs. lower support are needed for optimization of LVAD adjustment strategies.