Pressure-strain-volume loop-derived myocardial work is a predictor of adverse outcomes in transcatheter aortic valve replacement patients

Abstract Background Accurately assessing left ventricular (LV) function is a key step in cardiac disease management, however, it can prove especially challenging in pressure overload conditions such as aortic stenosis (AS). Since conventional echocardiographic measures like ejection fraction (EF) and global longitudinal strain (GLS) are significantly influenced by loading conditions, their reliability in quantifying true myocardial performance is quite limited. Myocardial work indices, derived from non-invasively measured pressure-strain loops are load-adjusted parameters but still overlook preload and LV geometry. Incorporating instantaneous LV volume into myocardial work calculations may provide a more physiologically accurate technique. Purpose Accordingly, our study aimed to incorporate instantaneous LV volume into myocardial work analysis and to test the prognostic value of our novel pressure-volume-strain loop-derived myocardial work index parameter in patients undergoing transcatheter aortic valve replacement (TAVR). Methods We enrolled 316 TAVR patients (age: 79±6 years, 41% female). Examinations were performed one day prior to TAVR. Using dedicated semiautomatic tracking software, EF and LV volumes throughout the entire cardiac cycle were measured and GLS values were determined via speckle-tracking analysis. Individual pressure curves were generated using noninvasive blood pressure and the Doppler-derived mean transaortic gradient. Strain rate and LV pressure curves were utilized to calculate global constructive work (GCW). Then the pressure-strain loops were further adjusted to instantaneous LV volume, creating pressure-strain-volume loops from which we quantified volume-adjusted GCW (GCWVi). 68 patients reached our primary outcome of all-cause death during a median follow-up of 25 months. Results Using univariate Cox regression analysis, EF was not a predictor, however, GLS (HR: 1.083 [1.017-1.154], p=0.01), GCW per 100 unit change (HR: 0.951 [0.907-0.997], p=0.04) and GCWVi per 10 unit change (HR: 0.861 [0.766-0.967], p=0.01) were significantly associated with the primary outcome. In a multivariable Cox regression model, GCWVi emerged as a significant independent predictor (HR: 0.876 [0.773-0.992], p=0.037) along with age, sex, atrial fibrillation, hemoglobin and creatinine while other LV functional parameters did not. Conclusions While conventional measures and myocardial work analysis failed to independently predict outcome in a multivariate model with robust clinical markers of mortality in our cohort, our novel proposed method of GCWVi proved to be a significant independent preoperative marker of adverse outcome. Our new method may be a more sensitive, load-independent marker of LV function, thus providing added prognostic value in TAVR patients.