There is no proven therapy for heart failure with preserved ejection fraction (HFpEF). Research has shown beneficial responses to cardiac rehabilitation (CR) among HF patients. To date, there are no reports comparing those responses between patients with HFpEF and those with reduced ejection fraction (HFrEF). The purpose of this study was to compare responses to CR in patients with HFpEF versus those with HFrEF. We included 78 consecutive patients (mean age 69 ± 15 years; 80% male) with HF in our CR unit who underwent cardiopulmonary exercise testing and brachial artery flow-mediated dilation (FMD) testing pre- and 5 months post-CR. Patients were judged as HFpEF (n = 40) or HFrEF (n = 38) using a left ventricular ejection fraction (LVEF) cut-off of 50%, and endothelial dysfunction was defined as FMD ≤ 5.0%. Following 155 ± 11 days and 44 ± 8 sessions, peak oxygen uptake (\( V_{{{\text{O}}_{ 2} }} \)) and plasma B-type natriuretic peptide concentrations improved significantly in both groups. The percentage change in peak \( V_{{{\text{O}}_{ 2} }} \) of HFrEF patients was significantly greater than compared with the HFpEF patients (P < 0.01). To further investigate whether a combination of LVEF and FMD values predicts the effect of CR, we divided patients into four groups according to LVEF of 50% and FMD of 50%. Post hoc analysis showed a significant difference between HFrEF patients without endothelial dysfunction and HFpEF patients with endothelial dysfunction (P = 0.01). In conclusion, although CR improves prognosis in HF patients, a larger effect can be expected in HFrEF patients than in HFpEF patients, and endothelial function may enhance the effect.