Invasive validation of pressure-volume loops derived from cardiovascular magnetic resonance imaging and brachial blood pressure in heart failure patients

Abstract Introduction Left ventricular (LV) pressure-volume (PV) loops provide gold-standard physiological information but require invasive measurements of ventricular intracavity pressure, limiting clinical and research applications. Recent development has seen the introduction of non-invasively computed PV loops from cardiovascular magnetic resonance (CMR) volumetry and a brachial blood pressure measurement. The approach combines LV volumes with a time-varying elastance function to compute time-resolved LV pressures and was validated on invasive pressure data from a porcine model. The method is readily implemented using standard CMR sequences and provides measures of hemodynamic parameters including stroke work, myocardial efficiency, and contractile state. However, the method remains to be validated in patients using invasive left ventricular pressure recordings. Purpose To validate for the first time in human patients the performance of non-invasively computed PV loops against invasive measures. Methods Four heart failure patients underwent two subsequent sessions of CMR cine imaging and simultaneous brachial blood pressure measurement, with intravenous administration of two different vasoactive drugs, resulting in two different haemodynamic states for each patient. LV catheterization was then conducted with repeat administration of the same infusions. Pressure-volume loops were computed from CMR volumes combined with 1) a time-varying elastance function scaled to brachial blood pressure and temporally stretched to match volume data, and 2) invasive pressures averaged from multiple sampled beats. Method comparison was conducted using linear regression and Bland-Altman analysis. Results Figure 1 shows non-invasively derived PV loop parameters compared to invasive data. The non-invasive method demonstrated strong correlations and low bias for stroke work (R2=0.97, bias 4.6%, p<0.0001), potential energy (R2=0.83, bias 1.5%, p=0.001), end-systolic pressure-volume relationship (R2=0.90, bias 5.4%, p=0.0003), energy per ejected volume (R2=0.93, bias 3.5%, p=0.0001), ventricular efficiency (R2=0.99, bias 1.1%, p<0.0001), arterial elastance (R2=0.87, bias −7.8%, p=0.0006), and mean external power (R2=0.89, bias 4.6%, p=0.0005). Conclusions Pressure-volume loops can be precisely and accurately computed from cardiovascular magnetic resonance imaging and brachial cuff blood pressure in humans, and is ready for use in research applications. Funding Acknowledgement Type of funding sources: Foundation. Main funding source(s): Swedish Heart Lung Foundation