- To determine the value of apolipoprotein E4 (APOE*E4) allele in predicting discharge impairment and disability in ischemic stroke patients after acute rehabilitation. Prospective study comparing results of rehabilitation in patients with different APOE genotypes. Acute neurologic rehabilitation department in Israel. One hundred one consecutive patients 75 years old or less with a first ischemic stroke. Not applicable. Main Outcome Measure: Impairment, as measured by the National Institutes of Health Stroke Scale (NIHSS), and disability, as assessed with the FIM trade mark instrument. On admission, there was no significant difference in the FIM or NIHSS measurements between the apo E4 group and other patients, but the prevalence of aphasia was 2.07 times more frequent in those with the APOE*E4 genotype (95% confidence interval, 0.98-4.4). A logistic regression model demonstrated that score measurements on admission were highly predictive of the NIHSS score at discharge (receiver operator curve=96.1%), whereas the presence of the APOE*E4 genotype did not add significantly to the model in predicting poorer rehabilitation treatment outcome as measured by the FIM or the NIHSS. The presence of the apo E4 allele did not predict a poorer outcome of rehabilitation treatment after ischemic stroke, but it was associated with an increased prevalence of aphasia. Further studies are warranted to confirm this association.