Trends in utilization, mortality, and resource use after implantation of left ventricular assist devices in the United States

Objectives: Adoption of implantable left ventricular assist devices has dramatically improved survival and quality of life in suitable patients with end-stage heart failure. In the era of value-based healthcare delivery, assessment of clinical outcomes and resource use associated with left ventricular assist devices is warranted. Methods: Adult patients undergoing left ventricular assist device implantation from 2008 to 2016 were identified using the National Inpatient Sample. Hospitals were designated as low-volume, medium-volume, or high-volume institutions based on annual institutional left ventricular assist device case volume. Multivariable logistic regression was used to evaluate adjusted odds of mortality across left ventricular assist device volume tertiles. Results: Over the study period, an estimated 23,972 patients underwent left ven-tricular assist device implantation with an approximately 3-fold increase in the num-ber of annual left ventricular assist device implantations performed (P for trend<.001). In-hospital mortality in patients with left ventricular assist devices decreased from 19.6% in 2008 to 8.1% in 2016 (P for trend<.001) and was higher at low-volume institutions compared with high-volume institutions (12.0% vs 9.2%, P < .001). Although the overall adjusted mortality was higher at low-volume compared with high-volume institutions (adjusted odds ratio, 1.66; 95% confidence interval, 1.28-2.15), this discrepancy was only significant for 2008 and 2009 (low-volume 2008 adjusted odds ratio, 5.5; 95% confidence interval, 1.9-15.8; low-volume 2009 adjusted odds ratio, 2.3; 95% confidence interval, 1.4-3.8). Conclusions: Left ventricular assist device use has rapidly increased in the United States with a concomitant reduction in mortality and morbidity. With maturation of left ventricular assist device technology and increasing experience, volume-related variation in mortality and resource use has diminished. Whether the apparent uni-formity in outcomes is related to patient selection or hospital quality deserves further investigation.