More accurate myocardial work index when incorporating estimated LV diastolic pressure

Abstract Introduction Current available tools to quantify stroke work, based in pressure-volume (PV) loops or pressure-strain loops, are limited by the unavailability of pressure during left ventricle (LV) filling. However, a recent study by Smiseth et al have enabled non-invasive quantification of LV diastolic pressure (1). Purpose The purpose of this study is to develop and validate a non-invasive pressure-strain loop which incorporates both LV systolic and diastolic pressures in the analysis. Methods The Pressure-Strain-loop proposed consists of merging the LV diastolic pressure trace estimated via a novel non-invasive left atrial pressure (NiLAP) methodology (1) into the LV pressure estimate used in the currently used method based in Russell et al. 2013 (2). The study was done in a population referred for left heart catheterization due to suspected coronary artery disease. LV strain was measured by speckle tracking echocardiography and LV pressure by micromanometer-tipped catheters. The area of the invasive LV pressure-strain loops were used as reference for myocardial work index (MWi). Furthermore, we additionally computed MWi according to Russell et al. (2) which assumes LV diastolic pressure is equal to zero. For further understanding of potential implications for MWi estimates at higher LV end-diastolic pressures (EDP), the patients were clustered in 3 groups (EDP 20mmHg). Results The personalized non-invasive PV-loop reflects both qualitatively and quantitatively (via MWi) the invasively generated counterpart (Figure 1). The MWi using invasive LV pressure was 1644.9 ±631.8 mmHg*% as compared to was 1696.6 ±596.4 (p=0.18) by the method Russell et al. 2013 and 1646.2 ±593.2 (p=0.97) with the method that incorporated estimated LV diastolic pressure. The only minor improvement in estimated SWi by incorporating LV diastolic pressure confirms the validity of the approach of Russel at al. in patients with low filling pressures. Nevertheless, the error in the estimation of SWi increased with moderately elevated LV filling pressures, indicating a significant error which increases when diastolic pressure is not considered (for up to 8% for EDP>20mmHg). With the novel approach suggested in this study, the error remains low (2%) at elevated LV diastolic pressure (Figure 2). Conclusions It is feasible to obtain personalized and non-invasive PV-loops over the complete cardiac cycle, including the diastolic period. Furthermore, the error in the currently most widely used method to estimate MWi is small for low/normal LV filling pressures but increases at higher filling pressures. Further studies should explore how inclusion of estimates of LV diastolic pressure can improve MWi estimates in patients with markedly elevated LV filling pressures such as patients with severe congestive heart failure or valvular heart disease.Figure 1 Figure 2