Preload-corrected dynamic Starling mechanism in patients with heart failure with preserved ejection fraction.

10.9 ± 3.8 vs. 11.2 ± 1.3 mmHg, P = 1.00). Transfer function analysis between diastolic pulmonary artery pressure (PAD) representing dynamic changes in LVEDP vs. SV index was applied to obtain gain and coherence of the DSM. The DSM gain was significantly lower in HFpEF patients than in the controls, even at a similar level of LVEDP (0.46 ± 0.19 vs. 0.99 ± 0.39 ml·m-2·mmHg-1, P = 0.0018). Moreover, the power spectral density of PAD, the input variability, was greater in the HFpEF group than the controls (0.75 ± 0.38 vs. 0.28 ± 0.26 mmHg2, P = 0.01). Conversely, the power spectral density of SV index, the output variability, was not different between the groups ( P = 0.97). There was no difference in the coherence, which confirms the reliability of the linear transfer function between the two groups (0.71 ± 0.13 vs. 0.77 ± 0.19, P = 0.87). The DSM gain in HFpEF patients is impaired compared with age-matched controls even at a similar level of LVEDP, which may reflect intrinsic LV diastolic dysfunction and incompetence of ventricular-arterial coupling. NEW & NOTEWORTHY The beat-to-beat dynamic Starling mechanism (DSM), the dynamic modulation of stroke volume because of breath-by-breath changes in left-ventricular end-diastolic pressure (LVEDP), reflects ventricular-arterial coupling. Although the DSM gain is impaired in heart failure with preserved ejection fraction (HFpEF) patients, it is not clear whether this is because of higher LVEDP or left-ventricular diastolic dysfunction. The DSM gain in HFpEF patients is severely impaired, even at a similar level of LVEDP, which may reflect intrinsic left-ventricular diastolic dysfunction.