Non-HDL-C Levels Linked to Risk for Cardiovascular Events

Fran Lowry

March 28, 2012

March 28, 2012 — Levels of non–high-density lipoprotein cholesterol (non-HDL-C) among users of statins is linked to the risk for a major cardiovascular event, such as a heart attack or stroke, more strongly than are levels of low-density lipoprotein cholesterol (LDL-C) and apolipoprotein B (apoB).

The finding, from a meta-analysis of data from 8 trials comprising a total of 62,154 patients, is published in the March 28 issue of JAMA.

Statins are a lynchpin of therapy for primary and secondary prevention of cardiovascular disease, and LDL-C has long been the target used to determine when to start and when to adjust lipid-lowering therapy.

But research is beginning to question whether LDL-C is the best lipid measure for predicting cardiovascular risk or to measure the atheroprotective effect of statin therapy, the study authors, led by S. Matthijs Boekholdt, MD, PhD, from the Academic Medical Center, Amsterdam, the Netherlands, write.

In the current study, Dr. Boekholdt and his colleagues sought to evaluate the relative strength of the associations of LDL-C, non-HDL-C (total cholesterol minus HDL), and apoB with cardiovascular risk in patients receiving statin therapy.

The meta-analysis looked at individual-patient data from randomized, controlled statin trials published between 1994 and 2008. All study participants had conventional lipids and apolipoprotein levels assessed at baseline and at 1-year follow-up.

Among the 38,153 patients who were randomly assigned to statin therapy, 158 (0.4%) had a fatal myocardial infarction and 1678 (4.4%) had a nonfatal myocardial infarction during follow-up. Death due to other types of coronary artery disease occurred in 615 (1.6%) of the study participants, and fatal or nonfatal stroke occurred in 1029 (2.7%) study participants.

In addition, 2806 (7.4%) patients were hospitalized for unstable angina, and 6286 major cardiovascular events (MACE) occurred in 5387 participants, for an event rate of 14.1%.

The analysis showed that all the studied lipid and apolipoprotein measures were associated with the risk for MACE and that these associations were statistically significant.

Table. Adjusted Hazard Ratios for MACE per 1–Standard Deviation Increase

Parameter Hazard Ratio (95% Confidence Interval) P Value
LDL-C 1.13 (1.10 - 1.17) < .001
Non-HDL-C 1.16 (1.12 - 1.19) < .001
ApoB 1.14 (1.11 - 1.18) < .001


Further analysis showed that the difference between LDL-C and non-HDL-C in predicting the risk for MACE for each 1–standard deviation increase was statistically significant (P = .002). The difference between non-HDL-C and apoB was also statistically significant (P = .02), but the difference between LDL-C and ApoB was not (P = .21).

"Given the fact that many other arguments for the clinical applicability of non-HDL-C and LDL-C are identical, non-HDL-C may be a more appropriate target for statin therapy than LDL-C," Dr. Boekholdt suggests.

Dr. Mauro Moscucci

Commenting on the study findings for Medscape Medical News, Mauro Moscucci, MD, MBA, chief of the Cardiovascular Division and professor of medicine at University of Miami Miller School of Medicine, Florida, said the authors did a nice analysis trying to tease out the differences in predictive value for adverse cardiovascular events between LDL-C, non-HDL-C, and ApoB.

"However," Dr. Moscucci noted, "there are still concerns related to pooling together multiple clinical trials with different designs, different treatments, and different targets, not to mention the added concerns related to collinearity between measurements, as non-HDL cholesterol includes both LDL and ApoB."

He believes that changing current practice, which is currently based on LDL-C targets, might be premature until guidelines become available.

"For example, we know today that an LDL target of less than 70 is appropriate in high-risk patients. The present study does not provide any information on the difference between non-HDL and LDL cholesterol in patients who are at this LDL target. Yet, we might still want to pay more attention to non-HDL cholesterol values in patients for whom the LDL target is less than 100 mg/dL and who have a non-HDL cholesterol greater than 130 mg/dL," he said.

"All that said, I think that based on this study, I will take an additional consideration of non-HDL cholesterol and decide whether I might want to be more aggressive with therapeutic interventions in intermediate-risk patients who might fall into the category of LDL less than 100 mg/dL and non-HDL greater than 130 mg/dL. This is definitely an easy measurement that does not require any additional test," Dr. Moscucci added.

Dr. Boekholdt reports a financial relationship with Pfizer. Dr. Moscucci has disclosed no relevant financial relationships.

JAMA. 2012;307:1302-1309. Abstract