Letter by Kuller and Mackey Regarding Article, “Lipids and Lipoproteins and Risk of Different Vascular Events in the MRC/BHF Heart Protection Study”
To the Editor:
The results of the Heart Protection Study (HPS) comparing detailed lipoproteins versus low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C),1 and a subsequent editorial by Ockene,2 raise an important question about the value of measuring lipoprotein subfractions to predict risk of coronary heart disease (CHD) and therapeutic decisions. In the past, many believed that measuring the LDL-C subfraction was unnecessary because it was highly correlated with total cholesterol, a robust predictor of CHD events in the population. LDL-C replaced total cholesterol because total cholesterol was less closely associated with risk (1) for women with high HDL-C levels but not elevated LDL-C and (2) for individuals with diabetes mellitus or metabolic syndrome, who typically had low levels of both HDL-C and LDL-C.
The HPS included adults with prior cardiovascular disease, diabetes mellitus, or hypertensive men aged ≥65, for whom the extent of preexisting atherosclerotic disease and plaque characteristics are likely to be the primary determinants of risk of CHD. All the lipid and lipoprotein measurements were associated with very small relative risk (ie, a 1–standard deviation increase in LDL-C in the placebo group was only associated with a 10% increase in the risk of CHD). In these very high-risk individuals, measurements of lipids or lipoproteins as screening to determine treatment (ie, need for lipid-lowering drugs) may be less important (ie, secondary prevention of CHD). Practically all patients with clinical CHD are recommended to receive lipid-lowering therapy.
The current HPS report that LDL-C, apolipoprotein B, non-HDL-C, and LDL particles (LDL-P) have similar associations with CHD risk may be a function of their high correlation at the population level. Many individuals have concordant levels (ie, their population percentile is similar across all lipid and lipoprotein indices). We and others have reported that patients in certain subgroups (ie, metabolic syndrome, diabetes mellitus, and obesity) often have discordant lipid/lipoprotein levels, lower LDL-C without equivalently low apolipoprotein B, LDL-P, or non-HDL-C.3–5 This disconnect of low LDL-C but high atherogenic particle concentration (apolipoprotein B or LDL-P) is highly related to the increased prevalence of small, dense cholesterol-poor LDL-P among these individuals. Both Multi-Ethnic Study of Atherosclerosis3 and the Framingham Offspring Study4 have reported that for individuals with discordant levels of LDL-C and LDL-P, the risk of incident CHD corresponds to the LDL-P concentration, rather than to the LDL-C level. Similar discordance is seen with apolipoprotein B versus LDL-C and might affect treatment decisions.5 The challenge is to identify individuals who are discordant with regard to lipoprotein levels3 and then determine whether such differences in lipoprotein levels affect therapeutic decisions, possibly new pharmacological agents, genetic studies, and clinical outcomes.
The article and editorial raise a very important issue about the importance of lipoprotein measurements. We propose that screening by total cholesterol alone would be adequate for populations, but there are likely many individuals with discordant levels between lipoprotein measures that could impact therapeutic decisions. Furthermore, evaluation of the importance of lipoprotein analysis in research and clinical decision making should be a high priority.
Lewis H. Kuller, MD, DrPH
Rachel H. Mackey, PhD, MPH, FAHA
Department of Epidemiology
Graduate School of Public Health
University of Pittsburgh
Dr Mackey received a previous research grant from LipoScience, Inc., to the University of Pittsburgh. Dr Kuller has no disclosures.
- © 2012 American Heart Association, Inc.
- Parish S,
- Offer A,
- Clarke R,
- Hopewell JC,
- Hill MR,
- Otvos JD,
- Armitage J,
- Collins R
- Ockene IS