CLUSTERING OF DYSLIPIDEMIA, HYPERURICEMIA, DIABETES, AND HYPERTENSIONAND ITS ASSOCIATION WITH FASTING INSULIN AND CENTRAL AND OVERALL OBESITY IN A GENERAL-POPULATION

Citation
Mi. Schmidt et al., CLUSTERING OF DYSLIPIDEMIA, HYPERURICEMIA, DIABETES, AND HYPERTENSIONAND ITS ASSOCIATION WITH FASTING INSULIN AND CENTRAL AND OVERALL OBESITY IN A GENERAL-POPULATION, Metabolism, clinical and experimental, 45(6), 1996, pp. 699-706
Citations number
64
Categorie Soggetti
Endocrynology & Metabolism
ISSN journal
00260495
Volume
45
Issue
6
Year of publication
1996
Pages
699 - 706
Database
ISI
SICI code
0026-0495(1996)45:6<699:CODHDA>2.0.ZU;2-V
Abstract
Clustering of elevated triglycerides, decreased high-density lipoprote in cholesterol (HDL-C), hyperuricemia, diabetes, and hypertension has been related to insulin resistance/high insulin levels and central and /or overall obesity. The extent to which these abnormalities cluster a nd whether hyperinsulinemia, central adiposity, and overall obesity ea ch independently associate with this clustering were evaluated in 14,4 81 US whites and African-Americans 45 to 64 years of age. With the exc eption of hypertension, abnormalities rarely existed in isolated form. Clustering greatly exceeded chance association (P <.001). Although th is clustering was greater in relative terms (ratio of observed to expe cted cluster frequency) in the lean and less centrally obese, it was g reater in absolute terms (observed minus expected cluster frequency as a percent of total population) in the more centrally and more general ly obese. The greatest excesses were found for clusters that included both hypertriglyceridemia and low HDL-C. Multiple logistic regression models showed strong and independent graded relationships of clusters with quintiles of fasting insulin (fifth quintile odds ratio, 10 to 54 , P <.001) and to a lesser degree with quintiles of the waist to hip r atio (2.2 to 5.4, P <.001 for most) and of body mass index (1.6 to 4.5 , P <.05 for most). In conclusion, all abnormalities cluster in excess of that predicted by chance, with clusters showing remarkable and gra ded independent associations with fasting hyperinsulinemia and to a le sser extent with central and overall obesity. Thus, a metabolic syndro me occurs in both lean and obese middle-aged US adults. Copyright (C) 1996 by W.B. Saunders Company.