Altered composition of triglyceride-rich lipoproteins and coronary artery disease in a large case-control study

Paul N. Hopkins, M. Nazeem Nanjee, Lily L. Wu, Michael G. McGinty, Eliot A. Brinton, Steven Hunt, Jeffrey L. Anderson

Research output: Contribution to journalArticle

16 Citations (Scopus)


Background: Traditional beta-quantification of plasma lipoproteins by ultracentrifugation separates triglyceride-rich lipoproteins (TGRL) from higher density lipoproteins. The cholesterol in the TGRL fraction is referred to as measured very low-density lipoprotein cholesterol (VLDL-C) recognizing that other TGRL may be present. The measured VLDL-C to total plasma triglyceride (VLDL-C/TG) has long been considered an index of average TGRL composition with abnormally high VLDL-C/TG ratios (≥0.30 with TG > 150 mg/dL) indicative of atherogenic remnant accumulation (type III hyperlipidemia). However, virtually no reports are available which examine potential associations between CAD and VLDL-C/TG at the lower end of the spectrum. Methods and results: We performed ultracentrifugation in 1170 cases with premature-onset, familial CAD and 1759 population-based controls and examined the VLDL-C/TG ratio as an index of TGRL composition. As expected, we found very high CAD risk associated with severe type III hyperlipidemia (OR 10.5, p = 0.02). Unexpectedly, however, we found a robust, graded, and independent association between CAD risk and lower than average VLDL-C/TG ratios (p < 0.0001 as ordered categories or as a continuous variable). Among those in the lowest VLDL-C/TG category (a ratio <0.12), CAD risk was clearly increased (OR 4.5, 95% CI 2.9-6.9) and remained significantly elevated in various subgroups including those with triglycerides below 200 mg/dl, in males and females separately, as well as among those with no traditional CAD risk factors (OR 5.8, 95% CI 1.5-22). Significant compositional differences by case status were confirmed in a subset whose samples were re-spun with measurement of lipids and apolipoprotein B (apo B) in each subfraction. Conclusions: We found a strong, graded, independent, and robust association between CAD and lower VLDL-C/TG ratios. We consider this a novel, hypothesis-generating observation which will hopefully generate additional future studies to provide confirmation and further insight into potential mechanisms.

Original languageEnglish
Pages (from-to)559-566
Number of pages8
Issue number2
Publication statusPublished - Dec 2009
Externally publishedYes



  • Coronary artery disease
  • Dysbetalipoproteinemia
  • HDL cholesterol
  • Lipids
  • Plasma triglycerides
  • Type III hyperlipidemia
  • Ultracentrifugation
  • VLDL

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this