Genetic Link Found in Recurrent Lobar Intracerebral Hemorrhage

abstract & commentary

Source: O’Donnell HC, et al. Apolipoprotein E genotype and the risk of recurrent lobar intracerebral hemorrhage. N Engl J Med 2000;342:240-245.

Cerebral amyloid angiopathy (caa) is characterized by abnormal deposition of amyloid protein in cortical and leptomeningeal blood vessels. CAA is frequently associated with lobar intracerebral hemorrhage. Previous studies have shown a link between the occurrence of CAA and possession of either the e2 or e4 allele of apolipoprotein E (APOE). The results of a new prospective cohort study by O’Donnell and colleagues from the Massachusetts General Hospital and Spaulding Rehabilitation Hospital in Boston indicate that the APOE genotype may be useful for identifying patients at increased risk for recurrence of intracerebral bleeding after an episode of lobar hemorrhage attributable to CAA.

The finding is based on the study of 71 patients with a mean age of 75 years who had suffered at least one lobar intraparenchymal bleed and consented to undergo genetic testing and longitudinal follow-up for two years. Ten members of the cohort subsequently died and underwent a postmortem examination that confirmed the presence of definite CAA. An additional 39 subjects were thought to have probable CAA on the basis of radiographic evidence of multiple small lobar hemorrhages. The remaining 22 were found to have had only one hemorrhagic event and were accordingly deemed to have possible CAA. During the two-year follow-up period, a total of 19 patients in this cohort had recurrent symptomatic hemorrhages, eight of whom died.

As in previous studies of patients, both the APOE e2 and e4 alleles were found to be over-represented among the study participants with CAA. Among several variables examined (including age, gender, the presence of hypertension, diabetes, and dementia), the only significant predictors of recurrence were a prior history of hemorrhagic stroke and APOE genotype. The two-year recurrence rate among carriers of APOE e2 and e4 was 28%, compared to 10% among those with APOE e3. Time to recurrence was least in those with the e2/e4 genotype, with four of eight patients in this group having recurrence of lobar hemorrhage within the first six months of follow-up. Previous hemorrhagic stroke was associated with more than sixfold increased risk of recurrence (risk ratio = 6.4, 95% CI = 2.2-18.5). The risk ratios for recurrence with possession of e2 and e4 were 4.7 (95% CI = 1.4-15.9) and 3.7 (95% CI = 1.1-11.7), respectively.

Using a Cox proportionate hazards model, it was inferred that APOE e2 and e4 exerted independent effects on recurrence. When the previous occurrence of hemorrhagic stroke was added to the model, the effect of APOE e2 was reduced below cutoffs for statistical significance, while the effects of e4 remained significant. Although no clear relationship was found between the number of copies of e2 or e4 and risk of recurrence, the number of subjects studied may have been too small to discern gene dose effects. O’Donnell et al suggest that determination of APOE genotype might be clinically useful for assessing prognosis of future hemorrhage among patients who have already had one or more such events.


Lobar intracerebral hemorrhage accounts for as much as one-third of all nontraumatic hemorrhage strokes and has its highest incidence among the elderly. Recurrence results in significant morbidity and mortality. No proven treatment or preventative strategy for CAA has yet to be found and management is empiric, involving control of hypertension and reduction of other stroke risk factors. The implication of a genetic factor in recurrence of cerebral hemorrhage could contribute to better understanding the underlying disease and may accelerate the process of discovery of new treatment.

The current circumstances, however, are in many respects reminiscent of those that exist for Alzheimer’s disease (AD). As many Neurology Alert readers know, APOE e4 is a proven risk factor for AD; however, APOE genotyping is not recommended for the purposes of predicting AD largely because no curative treatment or prevention is yet available. O’Donnell et al acknowledge that APOE genotype alone is an imperfect predictor of recurrent lobar intracerebral bleeding, and that multiple genetic and environmental factors are likely to be involved. Extrapolating from the precedent of AD, it seems unlikely that APOE genotyping will enjoy immediate and widespread use as a means of predicting future recurrence of lobar hemorrhage, at least until a treatment becomes available. —nrr

Risk of recurrent lobar intracerebral hemorrhage from cerebral amyloid angiopathy is significantly:

a. increased by a history of diabetes and hypertension.

b. decreased by possession of APOE e2 or e4.

c. unchanged by a past history of lobar hemorrhage.

d. affected by APOE e2 and e4 independently.

e. greater in females than males.