Background:
Although hemodynamic-driven Left Ventricular Assist Device (LVAD) speed changes have been shown to improve patient outcomes, there is no standardized algorithm for how to best adjust LVAD speed for optimization of cardiac output (CO) and decompression of left sided filling pressures as measured by pulmonary capillary wedge (PCW).
Methods:
We performed a retrospective study of LVAD patients at Penn State Hershey Medical Center from 2015 to present, to identify those that ramp studies during right heart catheterizations. 470 patients were identified, of which 60 had ramp studies. Only studies with at least 4 of 5 standard ramp speeds were included. Any study that did not exhibit at least a 20% decrease in PCW, suggesting improper LVAD function or recovered heart function, was also excluded. 32 studies were included in final analysis: 11 HVAD, 12 HMIII, and 9 HMII. CO measured by thermodilution was reported. Standard ramp speeds, in rpm, were as follows: HVAD (2100, 2400, 2700, 3000, 3300), HMIII (4500, 5000, 5500, 6000, 6500), and HMII (8000, 9000, 10,000, 11,000, 12,000).
Results:
Combined data showed an average CO improvement of 22.3% and average PCW decrease of 89.7%. HMII and HMIII had similar improvements in CO, 44.4% and 43.8%, respectively; much greater than HVAD (17.2%). HMIII had the largest magnitude of decompression 141.5% versus 106.4% for HMII and 56.1% for HVAD. Suction events occurred at an overall rate of 9.4%, and were greatest in HVADs.
Conclusion:
We report the ramp study hemodynamics for three separate LVADs. Overall, our results suggest the HMIII is most responsive to changes in speed in terms of improvement of CO and decreased PCW. Addition of samples to this analysis may allow for algorithm development to provide clinical guidance regarding LVAD speed settings.
In-silico assessment of the effects of right ventricular assist device on pulmonary arterial hypertension using an image based biventricular modeling framework
