RELATIONSHIP OF APOLIPOPROTEIN-E PHENOTYPES TO HYPOCHOLESTEROLEMIA

PURPOSE: Persons with total cholesterol (TC) levels less than 130 mg/d L (less than 3.26 mmol/L) make up less than 1 % of a healthy populatio n. Causes of hypocholesterolemia include a diet very low in cholestero l and saturated fat, disease, gene tic factors (including low apolipop rotein B-100 [apo B-100] and the apo E allele), and drug therapy. The purpose of this study was to determine the causes of hypocholesterolem ia in a healthy Kaiser Foundation Health Plan (KFHP) population. PATIE NTS AND METHODs: We conducted a dietary and health survey of 201 healt hy hypocholesterolemic adults (range: 2.04 to 3.88 mmol/L [79 to 150 m g/dL]) and 200 matched control subjects with TC levels in the middle q uintile of the population (range: 5.0 to 5.61 mmol/L [194 to 217 mg/dL ]) who had routine health screening from 1983 through 1985. We did apo E phenotyping studies and lipid and apo A-1 and B-100 measurements in a subgroup of 45 hypocholesterolemic subjects (mean TC level: 3.26 mm ol/L [126 mg/dL]) and in a comparison group of 49 unmatched volunteers (mean TC level: 5.04 +/- 0.75 mmol/L [ 195 +/- 29 mg/dL]).RESULTS: We found no differences in dietary intake or clinically significant medi cal illness between hyrocholesterolemic and control subjects. In the h ypocholesterolemic subgroup, we found an increased frequency of the ap o E2 allele (epsilon2) and a decreased frequency of the apo E4 allele (espilon4); the frequencies of the epsilon2, epsilon3, and epsilon4 al leles were 33.3 %, 63.3%, and 3.3%, respectively. The corresponding ap o E allele frequencies in the comparison subgroup were 8.2%, 73.5 %, a nd 18.4%, similar to those previously reported for the general populat ion and significantly different from those found in the hypocholestero lemic subgroup (p <0.0001). One hypocholesterolemic subject (a 46th pa tient) had a mutation in the apo B gene that resulted in the synthesis of a truncated species of apo B (apo B-46). CONCLUSION: Our study ind icates that hypocholesterolemia in our KFHP urban population is usuall y not caused by diet or disease. Biochemical factors, including the in creased frequency of the apo E-2 phenotype and the decreased frequency of the apo E-4 phenotype, are more important.