Abstract WMP54: Cerebrovascular Accidents in Patients With a Left Ventricular Assist Device

    loading  Checking for direct PDF access through Ovid

Abstract

Background: Cerebrovascular accidents (CVAs), including ischemic stroke, transient ischemic attack, and intracranial hemorrhage, are major and devastating complications of left ventricular assist devices (LVADs). However, little is known about their incidence and risk factors.

Materials and methods: We retrospectively reviewed the medical records of 75 consecutive cases supported with a LVAD (mean age of 39 years, 60 men, paracorporeal pulsatile-flow LVAD in 31, implantable centrifugal-flow in 18, and implantable axial-flow in 26) in our institution between December 2009 and June 2014. The incidence rate of CVAs was estimated as events per patient-year, and baseline characteristics (age, sex, etiologies of heart failure, comorbid diseases, and types of LVAD) were assessed to determine their contribution to the incidences of total, ischemic and hemorrhagic CVAs.

Results: During 113.7 patient-years of observation, a total of 59 CVA events (26 hemorrhagic and 33 ischemic) were identified in 31 cases, yielding an incidence rate of 0.52 events per patient-year. Univariate negative binomial regression (NBR) analysis demonstrated that patients with an axial-flow LVAD had significantly lower incidence of total CVAs [incidence rate ratio (IRR) 0.29, 95% confidence interval (CI) 0.10 - 0.88, p=0.03] and subsignificantly of both of ischemic (0.30, 0.05 - 1.69, p=0.17) and hemorrhagic CVAs (0.29, 0.08 - 1.02, p=0.05) than those with a paracorporeal LVAD (reference), whereas those with a centrifugal-flow LVAD did not. Patients with diabetes had a significantly higher incidence of total CVAs (IRR 2.39, 95% CI 1.03 - 5.57, p=0.04) and ischemic CVAs (4.01, 1.31 - 12.2, p=0.02) than those without diabetes but not of hemorrhagic CVAs (1.41, 0.54 - 3.66, p=0.48). Results of multivariate NBR analysis were similar with those of univariate analysis.

Conclusion: Among patients on LVAD support, CVA occurred at a rate of 0.52 events per patient-year. Axial-flow LVADs reduced the incidence of CVAs, regardless of ischemic or hemorrhagic subtype, compared with paracorporeal LVADs. Diabetes was associated with an increased risk of CVAs, but the influence was different between ischemic and hemorrhagic CVAs.

Related Topics

    loading  Loading Related Articles