Validation of non-invasive assessment of myocardial work in aortic stenosis: improvements by modifying the method for estimating the left ventricular pressure waveform

Abstract

Abstract Introduction Non-invasive myocardial work (MW) index incorporates strain by speckle-tracking echocardiography (STE) and individually estimated left ventricular pressure (LVP) curves to calculate the area of the pressure strain loop without the need for invasive LVP measurements. The method was validated in patients without aortic stenosis (AS) where a reference pressure curve is adjusted for individually measured aortic and mitral valve events and the peak LVP is defined by the brachial artery cuff pressure. Before applying this method in patients with AS, potential limitations which can influence the area of the pressure strain loop, such as the LVP curve profile, correct scaling of peak LVP and correct assessment of aortic events must be addressed. Purpose The present study aimed to assess the impact of the potential limitations specific to patients with AS and thereby the validity of non-invasive MW index in patients with AS. Methods In 20 patients with severe AS we obtained simultaneous LVP, by a micromanometer-tipped catheter, and strain by STE. For each patient, LVP curve estimations were done using three different models: 1. The established LVP reference model based on patients without AS. 2. Enhancement of the established LVP reference model by defining aortic valve opening with diastolic cuff pressure. 3. A new AS specific LVP reference model based on our current invasive measurements. Valvular events were determined by 2D and Doppler echocardiography, and peak LVP estimated as a sum of mean trans-aortic gradient and systolic cuff pressure. Estimated LVP curve tracings were thereafter directly compared with simultaneous invasive measurements (Figure 1). Furthermore, area of the pressure-strain loops using the different estimations of LVP curve were calculated to assess MW and compared to simultaneous invasive measurements for direct comparison. Results All three methods had excellent average correlation coefficient between estimated and invasively measured LVP traces. However, estimations with the AS specific reference curve and those enhanced with incorporation of diastolic pressure for aortic valve opening had a higher correlation coefficient (r=0.99, p<0.001) and a more physiological profile during early systole compared to that of the previously validated reference curve (r=0.96, p<0.001) (Figure 1). Furthermore, there was an excellent correlation (r=0.98, p<0.001) and good agreement between MW calculated with all three non-invasive estimation methods and invasive LVP (Figure 2). Conclusions The present study is the first to confirm the validity of non-invasive MW in patients with AS. Furthermore, a AS specific reference curve and the enhanced reference curve incorporating diastolic cuff pressure to define aortic valve opening both increased the accuracy of the estimated LVP curve and hence estimation of MW. This could be pivotal when assessing AS patients with marked regional differences such as LBBB or regional ischaemia. Funding Acknowledgement Type of funding sources: Public hospital(s). Main funding source(s): Oslo University Hospital Rikshospitalet