Frequency-Based Analysis of Diastolic Function: The Early Rapid Filling Phase Generates Negative Intraventricular Wave Reflections

Abstract

ACS ∗,† To assess global diastolic function (DF), both invasive and noninvasive methods have been utilized. Except for the end-diastolic pressure–volume relationship, currently all proposed parameters for diastolic function are derived purely from pressure or flow. To characterize the physiology of diastole in the context of atrioventricular pressure gradient generated transmitral flow, and in analogy to frequency-based characterization of ventricular–vascular coupling, we subjected the simultaneously recorded transmitral flow (E-wave) and micromanometric intraventricular pressure (LVP) waveforms to Fourier analysis in 20 subjects. This permitted computation of input impedance, characteristic impedance, the phase angle φ relating pressure to flow, and the complex reflection coefficient R ∗ during the E-wave. We found that the magnitudes of input impedance were 32 ± 12, 13.9 ± 4.4, 37 ± 13, and 53 ± 23 mmHg s/m for DC, 1st, 2nd, and 3rd harmonics, respectively. The characteristic impedance was 30 ± 15 mmHg s/m. The magnitude and phase angle of complex reflection coefficient R ∗ were 0.43 ± 0.11 and 3.58 ± 0.52 rad, respectively. The magnitude of the input impedance carrying most oscillatory power (1st harmonic) is lower than the characteristic impedance, verifying our finding that the real portion of R ∗ was negative. We also found that E waves with prolonged deceleration time (DT > 180 ms—“delayed relaxation” pattern) manifest increased phase differences between pressure and flow, voiding an optimal pressure–flow relationship. These findings elucidate the frequency-based (amplitude/phase) mechanisms