Background: Usefulness of low-level exercise testing as a diastolic stress test has not been investigated and data regarding change of noninvasive estimate of pulmonary vascular resistance (PVR) are not available. Methods: Supine bicycle exercise echocardiography was done in 23 patients (68짹7 years) referred for diastolic function evaluation due to cryptogenic dyspnea on exertion (DOE). All patients underwent medical evaluation to rule out significant cardiovascular and respiratory illness. Twenty three healthy volunteers (31짹3 years) were included as controls. The starting workload was 25 watts, which increased to 50 watts, the final workload. At each stage, various hemodynamic variables were measured, which included heart rate, stroke volume, mitral inflow velocity (E and A), systolic annular early diastolic velocity. Peak systolic velocity of tricuspid regurgitation (TRvmax) and time-velocity integral of pulmonary flow (RVOTtvi) were also measured and the ratio (TRvmax/RVOTtvi) was used as a noninvasive estimate of PVR. Results: Both patients and controls showed similar stroke volume at baseline and no significant change during exercise. Patients' heart rate was lower at baseline and its increase was also lower compared to controls. At baseline, patients showed significantly increased arterial pressure (p<.001), E/E��� (p<.001) and TRvmax (p=.011) with lower E (p=.001) and E��� (p<.001) compared to controls. With exercise, similar degree of increase in arterial pressure, E. E��� and E/E��� was observed in both patients and controls, whereas the increase of TRvmax was larger in patients (p=.005). TRvmax/RVOTtvi was not different at baseline and there was similar degree of increase after exercise in both patients and controls. Conclusions: Decreased chronotropic response and exaggerated increase of TRvmax during submaximal exercise is characteristic in patients with cryptogenic DOE.