Intraventricular hemodynamic forces do not differentiate between healthy controls and heart failure patients with preserved ejection fraction

Abstract

Abstract Background Hemodynamic force analysis has been proposed as a noninvasive marker of cardiac function. In a recent study, longitudinal (apical-to-basal) hemodynamic forces were derived from anatomical MRI images and found decreased in heart failure with preserved ejection fraction (HFpEF) patients compared to controls, indicating a potential use for prognostication and testing of therapeutic response. This issue has not been investigated using the reference method of measurement. Purpose To investigate whether intraventricular hemodynamic forces computed using gold-standard cardiac magnetic resonance flow maps can reproducibly differentiate between healthy controls and HFpEF patients. Methods 4D flow data were acquired in 59 subjects through cardiac magnetic resonance imaging using a 1.5T scanner (Siemens Healthcare, Erlangen, Germany). Hemodynamic forces within the LV were computed across the cardiac cycle using the Navier-Stokes equation to find the global pressure gradient, which was then integrated over the LV volume to produce the instantaneous hemodynamic force (unit: Newton) and subsequently normalized to ventricular volume, resulting in a force-volume index (N/l). Average longitudinal forces (root mean square, FRMS) were quantified over the entire cardiac cycle, with and without volume normalization. Results We studied 33 healthy subjects, 14 patients with HFpEF, 6 patients with HFmEF and 6 patients with HFrEF. Groups were similar with regards to sex, cardiac output, heart rate, systolic and diastolic blood pressure, and body surface area. Volume-normalized FRMS did not differ between controls and HFpEF (0.86±0.19 vs. 0.75±0.19 N/l, p=0.08) while lower values were found in HFmEF (0.60±0.19 N/l, p=0.004) and HFrEF (0.38±0.15 N/l, p<0.0001) compared to controls (Figure 1A). There was a significant positive correlation between EF and FRMS, both for the entire population (R2 = 0.54, Figure 1B) and for patients (R2 = 0.67, p<0.0001 for both). Importantly, non-normalized FRMS did not differ between controls (Figure 1C, 0.10±0.03 N) and HFpEF (0.09±0.03 N, p=0.25), HFmEF (0.11±0.02 N, p=0.18) or HFrEF (0.09±0.02 N, p=0.67). Moreover, no correlation was seen between non-normalized FRMS and EF (Figure 1D). Conclusions Hemodynamic forces computed from reference standard 4D flow CMR data do not differentiate between healthy controls and HFpEF patients regardless of whether volume normalization is used or not. Our findings do not support a role for hemodynamic forces in HFpEF assessment. Figure 1. (A) Volume-normalized hemodynamic forces over the entire cardiac cycle (lines: average values, shaded area: ±1SD for HFpEF), and (B) variation of volume-normalized average force, FRMS, with left ventricular ejection fraction (LVEF). (C), (D): When indexing to LV volume was not performed, the differences between groups was attenuated, and no correlation was seen between EF and FRMS. Funding Acknowledgement Type of funding source: Public grant(s) – National budget only. Main funding source(s): Swedish Heart and Lung Foundation, Region of Scania