Risk factors for the presence and development of Familial Dysbetalipoproteinemia in subjects from the general population with an APOE2E2 genotype

Abstract Background Subjects who carry one or two ɛ2 alleles of the APOE gene are protected from cardiovascular disease due to low LDL-cholesterol levels. However, 10–18% of ɛ2 homozygotes (ɛ2ɛ2) develop Familial Dysbetalipoproteinemia (FD), which is characterized by extensive remnant lipoprotein accumulation, making it a good model to study the effect of remnants on atherosclerosis and cardiovascular disease. Important causal factors for FD, or an “FD like phenotype” in ɛ2 heterozygotes (ɛ2ɛ3), are thought to be adiposity and insulin resistance. However, to date this relation was only evaluated in cross-sectional studies. Purpose To evaluate the cross-sectional and longitudinal association of adiposity and insulin resistance on the presence and development of FD in ɛ2ɛ2 or an “FD-like” phenotype in ɛ2ɛ3 subjects. Methods For this study we included 18042 subjects with an APOE measurement from two Dutch population-based cohorts; the PREVEND cohort (follow-up 4.1 (interquartile range (IQR) 4.0–4.4) years) and the Rotterdam Study (follow-up 10.1 (IQR 5.6–10.8) years). Subjects with an ɛ3ɛ3 genotype (n=10391) and ɛ4 carriers were excluded (n=5265). FD and “FD like” phenotype were defined as triglyceride levels >3 mmol/l or use of lipid lowering medication. Logistic regression models were used to evaluate the effect of age, sex, BMI, waist circumference, type 2 diabetes mellitus (T2DM) and metabolic syndrome (MetS) on FD lipid phenotype. Changes in adiposity measures and insulin resistance between baseline and follow-up were compared in subjects with and without development of FD at follow-up. Results In total, 2386 subjects were included of whom 118 participants had ɛ2ɛ2 and 2268 had ɛ2ɛ3 of whom 68% completed a follow-up visit. Subjects mean age was 59±14 years and 44% were male. In ɛ2ɛ2 subjects, 19% (n=23) had FD at baseline and 16% (n=11) developed FD during follow-up. In ɛ2ɛ3 subjects 13% (n=305) had an “FD like” phenotype at baseline and 11% (n=146) at follow-up. Cross-sectional determinants for the presence of an FD or “FD like” phenotype at baseline were male sex, BMI, waist circumference and non-lipid MetS criteria. In ɛ2ɛ2 subjects who developed FD during follow-up, markers of adiposity and insulin resistance did not change compared to baseline. However, these FD patients more often had adiposity (BMI, weight and waist circumference) and T2DM at baseline, compared to those and who did not developed FD. Conclusions These results show for the first tim