Left Ventricular Pressure-Strain-Derived Myocardial Work at Rest and during Exercise in Patients with Cardiac Amyloidosis

Background: Left ventricular pressure-strain-derived myocardial work index (LVMWI) is a novel, noninvasive method for left ventricular (LV) function evaluation in relation to LV pressure dynamics. LV global longitudinal strain (LVGLS) has proven benefit for diagnosis and risk stratification in patients with cardiac amyloidosis (CA), but LVGLS does not adjust for loading conditions. The aim of the present study was to characterize LVMWI at rest and during exercise in patients with CA. Methods: A total of 155 subjects were retrospectively included. These subjects comprised 100 patients with CA and 55 healthy control subjects. All patients had previously undergone comprehensive two-dimensional echocardiographic examinations at rest. Furthermore, a subgroup 27 patients with CA and 41 control subjects was examined using sennisu pine exercise stress echocardiography. Results: Patients with CA had significantly lower LVGLS, LVMWI, and LV myocardial work efficiency (LVMWE) than control subjects (P < .0001 for all). The reduction in LV myocardial performance was more pronounced in the basal segments, which led to significant alterations in the average apical-to-basal segmental ratios between patients with CA and control subjects (LVGLS, 2.6 [1.9 to 4.1] vs 1.3 [1.2 to 1.5]; LVMWI, 2.6 [1.7 to 3.8] vs 1.3 [1.1 to 1.5]; LVMWE, 1.1 [1.0 to 1.3] vs 1.0 [1.0 to 1.1]; P < .0001 for all). The average increase in LVMWI from rest to peak exercise was 1,974 mm Hg% (95% CI, 1,699 to 2,250 mm Hg%; P < .0001) in control subjects and 496 mm Hg% (95% CI, 156 to 835 mm Hg%; P < .01) in patients with CA. The absolute numeric LVGLS increase was 5.6% (95% CI, 3.9% to 7.3%; P < .0001) in control subjects and only 1.2% (95% CI, -0.9% to 3.3%; P = .26) in patients with CA (between groups, P < .0001) from rest to peak exercise. The LVMWI increase in patients with CA was mediated by improvement in the apical segments (P < .0001), whereas there was no significant LVMWI alterations in the midventricular or basal segments. LVMWE remained stable during exercise in control subjects (Delta -0.6%; 95% CI, -2.5% to 1.2%; P = .50) but decreased significantly in patients with CA (Delta -2.5%; 95% CI, -4.8% to -0.2%; P < .05). Conclusions: Patients with CA have significantly reduced magnitude of LVMWI compared with healthy control subjects. With exercise, the differences are even more pronounced. Even though LVMWI increased with exercise, LVMWE decreased, suggesting inefficient myocardial energy exploitation in patients with CA.