Evolution and prognostic impact of left ventricular myocardial work indices after transcatheter aortic valve replacement in patients with severe aortic stenosis

Abstract Introduction Left ventricular myocardial work (LVMW) is an echocardiographic-based technique to evaluate LV function using pressure-strain loops and therefore correcting for the left ventricular (LV) afterload. LVMW indices have been shown to be of prognostic importance in patients with aortic stenosis (AS) before transcatheter aortic valve replacement (TAVR), but no data is available on the evolution and the prognostic value of these parameters after TAVR. Purpose To measure LVMW indices before and after TAVR in patients with severe AS and explore the association with outcome. Methods As previously validated, LV global longitudinal strain (GLS) and LV systolic pressure (as the sum of the brachial arterial blood pressure and mean aortic valve gradient before TAVR and as brachial arterial blood pressure after TAVR) were incorporated to build pressure-strain loops and to calculate the following LVMW indices: global work index (LV GWI), global constructive work (LV GCW), global wasted work (LV GWW) and global work efficiency (LV GWE). The endpoint was all-cause mortality. Results A total of 255 patients were included with a median age of 82 years (IQR 77-85); 51% were men and 67% presented with NYHA class III/IV. After TAVR, LVEF and LV GLS remained in average unchanged (from 56 to 55%, p=0.470 and from 13.6 to 13.2%, p=0.068). While LV GWW remained unchanged after TAVR (from 247 to 258mmHg%, p=0.080), LV GWI, LV GCW and LV GWE significantly decreased (from 1882 to 1291mmHg%, p <0.001, from 2248 to 1671mmHg%, p <0.001 and from 89 to 85%, p <0.001 respectively). During a median follow-up of 59 months (IQR 40-72), 129 patients died. Univariable Cox regression analysis identified an association between all-cause mortality and male sex, diabetes mellitus, chronic obstructive pulmonary disease (COPD), renal function, atrial fibrillation, Charlson comorbidity index, pacemaker implantation, and post-TAVR LV GLS, LV GWI and LV GCW. These clinical parameters were used as a basal multivariable Cox regression model of which on top either post-TAVR LV GLS, LV GWI or LV GCW were added. All these parameters of post-TAVR LV systolic function showed an independent association with all-cause mortality, but LV GWI demonstrated the highest increase in model predictivity (Figure 1). Given the prognostic relevance of post-TAVR LV GWI, the population was divided into LV GWI tertiles. The group with the lowest LV GWI tertile showed a significantly higher proportion of male patients, higher prevalence of atrial fibrillation and also worse pre-TAVR LVEF, LV GLS, LV GWI, LV GCW and LV GWE. Kaplan Meier survival analysis confirmed the worse survival in the group with the lowest LV GWI compared to the highest tertile (log-rank ꭓ2: 8,224, p 0.016, Figure 2). Conclusion LVMW indices change after TAVR and post-TAVR LV GLS, LV GWI, LV GCW are independently associated with all-cause mortality. Among these parameters, LV GWI yielded the highest prognostic value.