Optimal Mechanical Unloading in Left Ventricular Assist Device Recipients Relates to Progressive Up-Titration in Pump Speed

Abstract

Left ventricular (LV) assist devices (LVADs) are known to elicit reverse remodeling by mechanically unloading the left ventricle. Current guidelines target a reduction in LV end-diastolic diameter (LVEDD) of15% compared with pre-LVAD dimensions; however, there is significant heterogeneity in the degree of unloading achieved. We sought to investigate factors associated with mechanical unloading at 6months of LVAD support. Data were retrospectively collected for 75 LVAD recipients at five time points: pre-LVAD, within 14days post-LVAD, and at 1, 3, and 6months post-LVAD. The percentage change in LVEDD between the pre-LVAD and 6months post-LVAD time points was termed ΔLVEDD. Optimal LV unloading was defined as ΔLVEDD of ≥15% at 6months. Patients who achieved optimal unloading (group A, n=30) were compared with patients who did not (group B, n=45). At 6months, optimally unloaded patients (group A) demonstrated higher fractional shortening (15%±10% vs 10%±7%, P=.007), lower rates of moderate or severe mitral regurgitation (10% vs 33%, P=.02), and lower pulmonary capillary wedge pressure (9±4 vs 16±7mm Hg, P=.02). Right ventricular dysfunction was more prevalent at 6months in poorly unloaded (group B) patients (73% vs 43%, P=.008). Between hospital discharge and 6months, the percentage increase in pump speed (Δ revolutions per minute) washigher in group A patients (4.4%±3.7% vs 0.1%±2.6%, P<.001). In a multivariate analysis, Δ revolutions per minute and tricuspid annular systolic velocity (S') at 6months were independently associated with 6-month ΔLVEDD. Recipients of LVADs who undergo progressive pump speed up-titration during outpatient follow-up are more likely to sustain optimal LV unloading. Progressive LVAD-related right ventricular failure is prevalent in suboptimally unloaded patients.