Abstract 10776: Market-Level Variation in Utilization of Percutaneous Ventricular Assist Devices Utilization in the United States

Abstract

Introduction: Percutaneous ventricular assist devices (PVAD) have begun to replace intra-aortic balloon pumps (IABP) for hemodynamic support during high-risk percutaneous coronary intervention (PCI) procedures even though data on effectiveness of PVAD compared to IABP for hard clinical endpoints remains limited. We examined variation in utilization of PVADs during PCI across U.S. markets and identify factors associated with PVAD utilization. Methods: Using 2013-2017 MEDPAR data, we identified all patients aged 65 years or older who underwent PCI with placement of either a PVAD or an IABP on the same date as the PCI procedure. The primary outcome was 30-day mortality. We used hospital referral region (HRR) to define regional healthcare markets with geographic contiguity. We examined variation in utilization of PVADs across HRRs and examined patient- and market-level characteristics across HRR quartiles of PVAD utilization. Using multilevel models, we examined the association of market-level factors with utilization of PVADs and whether higher utilization of PVADs was associated with lower risk of 30-day mortality at a market-level. Results: A total of 52,755 patients underwent inpatient PCI with either an IABP (36,260 [68.7%]) or PVAD (16,495 [31.3%]). There was a significant increase in the proportion of PCI procedures with PVAD (17% in 2013 to 46% in 2017), with a corresponding decrease in IABP (83% to 54%; P<.0001 for both). There was large variation in PVAD utilization across markets (0% to 84%), which persisted even after adjustment for differences in patient characteristics (adjusted median OR of 2.19; 95% CI: 2.03-2.35). Markets in the highest quartile of PVAD utilization were larger with a higher number of hospitals and had more market-level competition compared to markets in the lowest quartile of PVAD utilization. Greater utilization of PVAD at the market-level was not associated with lower risk-adjusted 30-day mortality. Conclusions: Utilization of PVADs for PCI in the U.S. continues to increase with marked variation in PVAD use across U.S. healthcare markets. Although patient- and market-level factors were associated with use of PVADs at the market-level, greater market utilization of PVADs was not associated with lower 30-day mortality.