The dilemma of a left ventricular assist device explantation: a decision analysis.

BACKGROUND: Left ventricular assist devices (LVADs) are an important adjunct to the management of end-stage heart failure patients. Uncertainty remains regarding whether to remove an LVAD in a clinically stable, asymptomatic patient who displays signs of ventricular recovery.
OBJECTIVES: To evaluate, from a patient's perspective, the quality-adjusted life expectancy of an LVAD explantation.
METHODS: A Markov state transition model was used to assess the benefits of two strategies: remove an LVAD or continue the LVAD support. Effectiveness was measured in quality-adjusted life months. Utility and probability scores were derived from the literature and expert opinion. The base case focused on a 35-year-old man with dilated cardiomyopathy, an implantable LVAD and signs of ventricular recovery, with New York Heart Association class I heart failure symptoms.
RESULTS: In the base case, continuing LVAD support was strongly preferred and improved quality-adjusted life expectancy by nine quality-adjusted life months. In sensitivity analyses for the utility post-transplant, removal of the device would have been preferred if the utility was less than 0.7. The model was also sensitive to the probability of late complications post-LVAD implantation. As the probability of complications increased (greater than 28%), the preferred strategy was to remove the LVAD.
CONCLUSIONS: Continuing LVAD support in asymptomatic patients with signs of ventricular recovery appears to be the preferred strategy for conveying greater quality-adjusted life months compared with LVAD explantation. As the probability of complications after LVAD implantation increases, the preferred strategy is to remove the LVAD.