Rosenson - Figure 2 - Meta-analysis of HDL and CHD Text

High-density lipoprotein (HDL) cholesterol has received so much attention because of its independent association with cardiovascular disease (CVD). Unlike low-density lipoprotein (LDL) cholesterol, however, where both the scientific and medical explanations are fairly consistent, for HDL cholesterol there are a lot of issues that are undefined, poorly defined, and not yet developed. As a result, several areas require further study:

• How is HDL defined in terms of its subfractions?
• How is HDL evaluated by the different analytical methods, how do they relate to each other, and how are they different (if they are different) with regard to prediction of CVD risk?
• How do/should we think about HDL function?

The most important functional aspect of HDL cholesterol is reverse cholesterol transport (RCT). But what does this mean? It is based on John Glomset’s theories in 1964[36][37] concerning the lecithin–cholesterol acyltransferase (LCAT) reaction. Is it relevant to humans, or again, is it a throwback to the experimental animal models? This raises some major issues that need to be addressed if the field of lipidology is to move forward and translate some of these very important experimental studies into clinical practice. As a biomarker of risk, HDL cholesterol is associated with an inverse relationship with incident CVD. A recent meta-analysis from the Emerging Risk Factors Collaboration[38] showed that HDL cholesterol remains an independent predictor of CVD risk, even after adjusting for major nonlipid risk factors and for baseline serum levels of LDL cholesterol and triglycerides.

As shown in the Figure, among the >302,000 individuals in the Emerging Risk Factors meta-analysis, there were almost 13,000 events. The fully adjusted model shows that the risk associated with an increased level of HDL cholesterol is 0.78, meaning that the higher the serum HDL cholesterol level, the lower the CVD risk (22% lower).

The data from the Cholesterol Treatment Trialists[39][40] show that regardless of baseline LDL cholesterol level – ie, whether it is ≥130, 100-129, 70-99, or <70 mg/dL, low levels of serum HDL cholesterol are predictive of incident CVD risk in statin-treated patients. Furthermore, after stratification into categories of LDL cholesterol, HDL cholesterol levels are still predictive of CVD events in men and in women. Thus the predictive value of HDL cholesterol levels is robust across a broad range of LDL cholesterol levels.

Rosenson RS. J Clin Lipidol. 2011; 5(6).