Systemic perfusion at peak incremental exercise in left ventricular assist device recipients: Partitioning pump and native left ventricle relative contribution

Background In continuous-flow left ventricular assist device (LVAD) recipients, little is known about the relative pump- and left ventricle-generated blood flow (PBF and LVBF, respectively) contribution to peak systemic perfusion during incremental exercise and about how PBF/LVBF interplay and exercise capacity may be affected by pump speed increase. Methods Twenty-two LVAD recipients underwent ramp cardiopulmonary exercise tests at fixed and increasing pump speed (+ 1.5% of baseline speed/10 W workload increase), echocardiography and NT-proBNP dosage. Peak systemic perfusion was peak VO2/estimated peak arterio-venous O2 difference, and LVBF was systemic perfusion minus PBF provided by LVAD controller. A change of peak percentage of predicted VO2max (Δpeak%VO2) ≥ 3 in increasing- vs. fixed-speed test was considered significant. Results Tricuspid annular plane systolic excursion (TAPSE) and NT-proBNP were significantly lower and higher, respectively, in Δpeak%VO2 < 3 than ≥ 3. A LVBF contribution to systemic perfusion significantly larger than that of PBF was observed in Δpeak%VO2 ≥ 3 vs. < 3 in fixed-speed test, which was further amplified in increasing-speed test (2.4 ± 1.7 l/min vs. 2.0 ± 1.5 l/min and 0.8 ± 2.2 l/min vs. 1.3 ± 2.3 l/min, respectively, p for trend < 0.0005). Among several clinical-instrumental parameters, logistic regression selected only TAPSE > 13 mm as a predictor of Δpeak%VO2 ≥ 3. Conclusions A significant LVBF contribution to peak systemic perfusion and pump speed increase-induced peak VO2 improvement was detectable only in patients with a more preserved right ventricular systolic function and stable hemodynamic picture. These findings should be taken into consideration when designing LVAD controllers aiming to increase pump speed according to increasing exercise demands.