Cutoff point separating affected and unaffected familial hypercholesterolemic patients validated by LDL-receptor gene mutants
Hiroshi Mabuchi, Toshinori Higashikata, Atushi Nohara, Hong Lu, Wen Xin Yu, Tsuyoshi Nozue, Yoshihiro Noji, Shoji Katsuda, Masa-Aki Kawashiri, Akihiro Inazu, Junji Kobayashi, Junji Koizumi, Hiroshi Mabuchi, Toshinori Higashikata, Atushi Nohara, Hong Lu, Wen Xin Yu, Tsuyoshi Nozue, Yoshihiro Noji, Shoji Katsuda, Masa-Aki Kawashiri, Akihiro Inazu, Junji Kobayashi, Junji Koizumi
Abstract
Familial hypercholesterolemia (FH) results from low-density lipoprotein (LDL) receptor gene mutations. Heterozygotes have twice normal LDL-cholesterol concentrations in early childhood, and experience early myocardial infarction. We demonstrated bimodal cholesterol frequency distributions, independently confirming existence of an identifiable hypercholesterolemic subpopulation. We assayed blood lipids in 181 FH patients genetically diagnosed and 100 unaffected relatives. Receiver operating characteristics curves were constructed. Total cholesterol and LDL-cholesterol concentrations showed bimodality. A total cholesterol cutoff of 225 mg/dl produced results agreeing with DNA testing (specificity, 98.5%; sensitivity, 99.4%). An LDL-cholesterol cutoff of 161-163 mg/dl produced 98.5% specificity and 98.3% sensitivity. Areas under curves were 0.9826 +/- 0.0058 for total cholesterol, and 0.9852 +/- 0.0043 for LDL-cholesterol. In conclusion, we define total cholesterol and LDL-cholesterol levels of 225 and 160 mg/dl, respectively, as cutoff points of normal subjects and FH patients.
Source: PubMed