Low High-Density Lipoprotein Cholesterol to Monitor Long-Term Average Increased Triglycerides

Abstract

Context: Increased triglyceride-rich remnants represent a causal risk factor for ischemic cardiovascular disease.

Objective: We tested the hypothesis that low high-density lipoprotein (HDL) cholesterol can be used to monitor long-term high triglycerides/remnant cholesterol, just as high hemoglobin A1c (HbA1c) can be used to monitor long-term high glucose levels.

Design, setting, participants, and interventions: We studied cross-sectionally 108 731 individuals, dynamically 1313 individuals with lipid measurement at 10 repeated visits, short-term 305 individuals during a fat load, and long-term 10 479 individuals with 2 lipid measurements 10 years apart.

Main outcome measures: Levels of HDL cholesterol and triglycerides.

Results: Cross-sectionally, HDL cholesterol was inversely associated with triglycerides (R2 = 0.26) and remnant cholesterol (R2 = 0.26). Dynamically, major changes in triglyceride levels from measurement to measurement were mimicked by corresponding modest changes in HDL cholesterol. In the short-term after a fat load, median triglycerides increased 96% while HDL cholesterol decreased only 1%. Long-term, in individuals with measurements 10 years apart, those who initially had the highest triglycerides and corresponding lowest HDL cholesterol, still had highest triglycerides and lowest HDL cholesterol 10 years later. Prospectively, individuals with increased triglycerides/remnant cholesterol had increased risk of myocardial infarction; however, when the HDL cholesterol monitoring was removed, increased triglycerides/remnant cholesterol were largely no longer associated with increased risk of myocardial infarction.

Conclusions: Low HDL cholesterol is a stable marker of average high triglycerides/remnant cholesterol. This suggests that low HDL cholesterol can be used to monitor long-term average high triglycerides and remnant cholesterol, analogous to high HbA1c as a long-term monitor of average high glucose levels.

Keywords: cardiovascular disease; diabetes; lipids; lipoproteins; regression dilution bias.