Clin Genet 1999: 55: 69–70 Printed in Ireland. All rights reser6ed Letter to the Editor Impact of ApoE4 allele on total cholesterol levels of children in northern Spain To the Editor: The e4 allele of apolipoprotein E (ApoE) has been associated with an increased risk of coronary heart disease, although its impact seems to vary accord- ing to factors such as gender, ethnic origin and lifestyle (1, 2). In some populations with classic cardioprotective diets and a relatively low inci- dence of cardiovascular disease, the ApoE4 iso- form seems to have only a minor influence on plasma lipid levels (3), while in others with non- cardioprotective diets, the influence, in terms of average excess, is significant (1, 4, 5). However, despite the recognition that in childhood signifi- cant developmental changes occur in lipid metabolism and that genetic factors, such as ApoE polymorphism, are related to the initial stages of atherosclerosis, most studies have been carried out on adults. Determinations for total cholesterol (TC), low density lipoprotein (LDL)-cholesterol, high density lipoprotein (HDL)-cholesterol, triglycerides (TG), ApoAI, ApoB and lipoprotein(a) (Lp(a)) plasma levels, and ApoE genotype tests were performed on 191 healthy children (110 boys and 81 girls), aged 8 – 15 years, from northern Spain. They were included in a Spanish longitudinal study, Cardiovascular Risk Factors in Childhood and Adolescence (known as RICARDIN), carried out from 1992 to 1996. The most common genotype was e3/e3 (0.76; n = 145), followed by e3/e4 (0.15; n = 29), e2/e3 (0.07; n = 13), e2/e4 (0.01; n = 2) and e4/e4 (0.01; n = 2). No e2/e2 genotype was observed. No sig- nificant differences between genders were found in the distribution of the ApoE allele (p = 0.42). The relative frequencies of the e2, e3 and e4 alleles were 0.04, 0.87 and 0.09, respectively. The sample re- sults did not differ significantly from Hardy – Weinberg expectations (p = 0.42) (6). Statistical analysis showed that mean levels of TC, LDL-cholesterol and ApoB were higher in children with the genotype e3/e4 and lower in those with e2/e3, while carriers of the genotype e3/e3 presented intermediate values (p B 0.004; Table 1). Differences among the ApoE genotypes in the levels of TC, LDL-cholesterol and ApoB were only significant in girls (p B 0.004). In the complete sample, the average excess of the e2 allele showed lower concentrations of TC ( - 16.16 mg/dl), LDL-cholesterol ( - 15.07 mg/dl) and ApoB ( - 9.90 mg/dl), whereas the average excess of the e4 allele presented high concentra- tions of TC ( + 10.90 mg/dl), LDL-cholesterol ( + 6.72 mg/dl) and ApoB ( + 4.73 mg/dl). Estimation of the average excess of the e3 allele was minimal in all cases. Our data on ApoE allelic distribution agree with other European studies that indicated the existence of a gradient, falling north–south, for the fre- quency of the e4 allele, ranging from 0.22 in Fin- land to 0.09 in our population. In the Mediterranean area, the allelic distribution of ApoE is similar to that found in Japan and China, whereas in northern European populations, it is closer to that of North America (2). Worldwide ApoE allele frequency distributions are concordant with the type of diet, from the fat-rich diets of northern Europe to the more varied diets along the Mediterranean strip. According to other studies, a consistent relation- ship exists between ApoE genotype and lipid levels from childhood (7, 8), which could be defined as the impact of each allele on the mean lipid level or average excess (4). For the e3 allele, the average effect is nearly zero due to its high frequency, and because the mean lipid levels of e3/e3 individuals are close to the average. The average excess of the e2 allele shows lower concentrations of lipid car- diovascular risk values (TC, LDL-cholesterol and ApoB), whereas the e4 allele increases these lipid levels. In spite of having a lower prevalence of cardio- vascular illness (9) and e4 allele frequency than northern European populations, in our population, the average excess of the e4 allele in TC levels is similar to that of northern Europeans, and higher than that in groups with a low incidence of cardio- vascular illness and variable e4 allele frequency, such as in Japanese, Icelandic or Sudanese popula- tions (5). This could be explained by the fact that in Spain, dietetic habits are shifting to a more atherogenic diet; the TC levels of Spanish children are currently at the upper limit of cardiovascular risk. 69