Ischemic and hemorrhagic brain injury during venoarterial-extracorporeal membrane oxygenation.

BACKGROUND: Structural neurological complications (ischemic stroke and intracranial bleeding) and their risk factors in patients receiving venoarterial-extracorporeal membrane oxygenation (VA-ECMO) are poorly described. Our objective was to describe frequencies, outcomes and risk factors for neurological complications (ischemic stroke and intracranial bleeding) in patients receiving VA-ECMO.

METHODS: Retrospective observational study conducted, from 2006 to 2014, in a tertiary referral center on patients who developed a neurological complication(s) on VA-ECMO.

RESULTS: Among 878 VA-ECMO-treated patients, 65 (7.4%) developed an ECMO-related brain injury: 42 (5.3%) ischemic strokes and 20 (2.8%) intracranial bleeding, occurring after a median [25th;75th percentile] of 11 [6;18] and 5 [2;9] days of support, respectively. Intracranial bleeding but not ischemic stroke was associated with higher mortality. Multivariable analysis retained only platelet level > 350 giga/L as being associated with ischemic stroke. Female sex, central VA-ECMO and platelets < 100 giga/L at ECMO start were independently associated with intracranial bleeding with respective odds ratios [95% CI] of 2.9 [1.1-7.5], 3.8 [1.1-10.2] and 3.7 [1.4-9.7]. In a nested case-control study, rapid CO2 -level change from before-to-after ECMO start also seemed to be associated with intracranial bleeding.

CONCLUSIONS: Neurological events are frequent in VA-ECMO-treated patients. Ischemic stroke is the most frequent, occurs after 1 week on ECMO support, has no specific risk factor and is not associated with higher mortality. Intracranial bleeding occurs earlier and is associated with female sex, central VA-ECMO, low platelet count and rapid CO2 change at ECMO start, and high mortality.

LEVEL OF EVIDENCE: This study provides Class IV evidence that central VA-ECMO, low platelet count and rapid CO2 change at ECMO start are associated with intracranial bleeding and high mortality.