The frequencies of apo E alleles in subjects suffering from ischemic cerebrovascular disease have been reported in two articles in Stroke. 1 -2 As discussed in a previous letter, 3 in the two studies E3 allele frequency in such subjects was decreased in comparison with aged control subjects. On the basis of the data from Reference 2 (indicating increased E4 allele frequency in noncardioembolic stroke patients), Dr Venarucci hypothesized such an increase in E4 allele frequency in 35 carotid

more »

... othrombotic plaques from subjects probably suffering from ischemic stroke. He found that 34 plaques contained E3 DNA. The probability of randomly selecting a group of 40 subjects (5 femoral plaques were also analyzed) from the general population with an incidence of E3/E3 phenotype of 98% is extremely low (^-2 =18; P<10~4). Because E3 allele frequency in the population tends to increase with age, this probability increases when the subjects selected have an average age of more than 70 years, but the increase is not enough to become insignificant. There are at least three explanations for these data: (1) they have been obtained by chance, in which case the confidence range would be unproved by increasing the number of patients, (2) the amount and the quality of the DNA extracted from the plaques were not suitable for apo E genotyping, which may be verified by genotyping DNA extracted from peripheral blood leukocytes or by using isoelectric focusing on blood plasma proteins, 4 and (3) the subjects in this study, and the cause of their cerebral infarctions, were atypical and may constitute a special subgroup of patients with cerebral infarction. However, none of the groups described in previous studies had such a high apo E3 allele frequency; for example, in our study 77% of the patients with carotid atherosclerotic stenosis had E3/E3 alleles. 1 Response We appreciate the interest shown by Dr De Smet in our patient who had bilateral infarction in the ACA vascular territory, and welcome his attempt to add some historical perspective to this unusual case. However, we can only partially agree with the issues raised. The described anomaly did not involve the circle of Willis as such but actually involved an absence of the postcommunicating part of the right ACA; in addition, the postcommunicating part of the left ACA had an increased diameter and split into two arteries by guest on July 22, 2018