APOE genotype and extent of bleeding and outcome in lobar intracerebral haemorrhage: a genetic association study

Background Carriers of APOE epsilon 2 and epsilon 4 have an increased risk of intracerebral haemorrhage (ICH) in lobar regions, presumably because of the effects of these gene variants on risk of cerebral amyloid angiopathy. We aimed to assess whether these variants also associate with severity of ICH, in terms of haematoma volume at presentation and subsequent outcome. Methods We investigated the association of APOE epsilon 2 and epsilon 4 with ICH volume and outcomes in patients with primary ICH in three phases: a discovery phase of 865 individuals of European ancestry from the Genetics of Cerebral Hemorrhage on Anticoagulation study, and replication phases of 946 Europeans (replication 1) and 214 African-Americans (replication 2) from an additional six studies. We also assessed the association of APOE variants with ICH volume and outcomes in meta-analyses of results from all three phases, and the association of APOE epsilon 4 with mortality in a further meta-analysis including data from previous reports. Admission ICH volume was quantified on CT scan. We assessed functional outcome (modified Rankin scale score 3-6) and mortality at 90 days. We used linear regression to establish the effect of genotype on haematoma volume and logistic regression to assess the effect on outcome from ICH. Findings For patients with lobar ICH, carriers of the APOE epsilon 2 allele had larger ICH volumes than did non-carriers in the discovery phase (p=2. 5x10(-5)), in both replication phases (p=0.008 in Europeans and p=0.016 in African-Americans), and in the meta-analysis (p=3.2x10(-8)). In the meta-analysis, each copy of APOE epsilon 2 increased haematoma size by a mean of 5.3 mL (95% CI 4.7-5.9; p=0.004). Carriers of APOE epsilon 2 had increased mortality (odds ratio [OR] 1.50, 95% CI 1.23-1.82; p=2.45x10(-5)) and poorer functional outcomes (modified Rankin scale score 3-6; 1-52, 1.25-1.85; p=1.74x10(-5)) compared with non-carriers after lobar ICH. APOE epsilon 4 was not associated with lobar ICH volume, functional outcome, or mortality in the discovery phase, replication phases, or meta-analysis of these three phases; in our further meta-analysis of 2194 patients, this variant did not increase risk of mortality (1.08,0.86-1.36; p=0.52). APOE allele variants were not associated with deep ICH volume, functional outcome, or mortality. Interpretation Vasculopathic changes associated with the APOE epsilon 2 allele might have a role in the severity and clinical course of lobar ICH. Screening of patients who have ICH to identify the epsilon 2 variant might allow identification of those at increased risk of mortality and poor functional outcomes.