Digoxin is Associated with Decreased Survival Free from Hemocompatibility-Related Adverse Events in LVAD Patients - A Propensity Score Matched Analysis

Abstract

Purpose Hemocompatibility-related adverse events (HRAEs) (stroke, suspected pump thrombosis (sPT), gastrointestinal bleeding (GIB)) reduce survival and quality of life in LVAD pts. The association between pharmacotherapy and both HRAEs and survival is poorly characterized. We aimed to assess this association in HeartMate II (HMII) pts. Methods We retrospectively studied pts implanted between Jan 2011 and Aug 2016. The primary endpoint was survival free from HRAEs. Cox proportional hazard models assessed time to primary endpoint by medication status at the time of the first post-implant outpt visit. Medication status was defined as receiving vs. not receiving PDE5 inhibitors, Digoxin, Aldosterone Receptor Antagonist (ARAs), ACE inhibitors (ACEi) or Angiotensin Receptor Blockers (ARBs), and β Blockers (BBs). A 1:1 propensity score (PS) match was performed based on 16 covariables. Results Of 209 HMII pts (age 57±14, 19% F), 96 (46%) met the primary endpoint - 16 deaths, 10 strokes, 30 sPT, 40 GIBs - over a mean follow-up of 1.68±1.31 years. In univariable analysis for the primary endpoint, ACE/ARBs were associated with reduced event rates (HR: 0.62, 0.41-0.96, p=0.03), while Digoxin had a trend toward increased event rates (HR: 1.49, 0.98-2.25, p=0.06). In the PS-matched analysis, the association between the primary endpoint and Digoxin reached significance (HR: 1.56, 1.02-2.39, p=0.04), while the one with ACE/ARBs did not (HR: 0.74, 0.45-1.20, p=0.22). Other drugs did not show any significant association (Fig.). HRs for Digoxin were generally consistent for individual components of the primary endpoint (death: 1.72, 0.75-3.94; stroke: 1.74, 0.55-5.50; sPT: 1.50, 0.71-3.17; GIB: 1.34, 0.65-2.74). Conclusion Our results suggest that the outpt use of Digoxin is associated with decreased survival free from HRAEs. Prospective randomized studies are warranted to assess the impact of Digoxin as well as other medications on outcomes among LVAD pts. Hemocompatibility-related adverse events (HRAEs) (stroke, suspected pump thrombosis (sPT), gastrointestinal bleeding (GIB)) reduce survival and quality of life in LVAD pts. The association between pharmacotherapy and both HRAEs and survival is poorly characterized. We aimed to assess this association in HeartMate II (HMII) pts. We retrospectively studied pts implanted between Jan 2011 and Aug 2016. The primary endpoint was survival free from HRAEs. Cox proportional hazard models assessed time to primary endpoint by medication status at the time of the first post-implant outpt visit. Medication status was defined as receiving vs. not receiving PDE5 inhibitors, Digoxin, Aldosterone Receptor Antagonist (ARAs), ACE inhibitors (ACEi) or Angiotensin Receptor Blockers (ARBs), and β Blockers (BBs). A 1:1 propensity score (PS) match was performed based on 16 covariables. Of 209 HMII pts (age 57±14, 19% F), 96 (46%) met the primary endpoint - 16 deaths, 10 strokes, 30 sPT, 40 GIBs - over a mean follow-up of 1.68±1.31 years. In univariable analysis for the primary endpoint, ACE/ARBs were associated with reduced event rates (HR: 0.62, 0.41-0.96, p=0.03), while Digoxin had a trend toward increased event rates (HR: 1.49, 0.98-2.25, p=0.06). In the PS-matched analysis, the association between the primary endpoint and Digoxin reached significance (HR: 1.56, 1.02-2.39, p=0.04), while the one with ACE/ARBs did not (HR: 0.74, 0.45-1.20, p=0.22). Other drugs did not show any significant association (Fig.). HRs for Digoxin were generally consistent for individual components of the primary endpoint (death: 1.72, 0.75-3.94; stroke: 1.74, 0.55-5.50; sPT: 1.50, 0.71-3.17; GIB: 1.34, 0.65-2.74). Our results suggest that the outpt use of Digoxin is associated with decreased survival free from HRAEs. Prospective randomized studies are warranted to assess the impact of Digoxin as well as other medications on outcomes among LVAD pts.