(Journal Watch is from the Publishers of New England Journal of Medicine)
anuary 19, 2010 | Mitchell S. V. Elkind, MD, MS
Lipid Levels and Stroke
In a large meta-analysis, non-HDL cholesterol was associated with risk for ischemic but not hemorrhagic stroke; triglycerides, HDL, and apolipoproteins were not associated with stroke risk.
For decades, lipid levels have been accepted as risk factors for atherosclerotic heart disease, but the relative importance of different lipid parameters and their association with ischemic stroke has been controversial, despite evidence that statin therapy reduces the risk for recurrent stroke. This controversy stems from the heterogeneity of ischemic stroke, which is caused primarily by nonatherosclerotic mechanisms.
This individual-patient meta-analysis, aimed at providing clear estimates of the vascular risks associated with lipid levels, included 68 prospective studies with data on 302,430 people without vascular disease at baseline; of these, 32 studies provided data on ischemic stroke outcomes in more than 173,000 people. Non-HDL cholesterol level (i.e., low-density, intermediate-density, and very low-density lipoprotein cholesterol) was modestly associated with ischemic stroke risk, but triglyceride and HDL cholesterol levels were not associated with either ischemic or hemorrhagic stroke risk. Both non-HDL and HDL cholesterol levels were associated with cardiac risk. Measurement of the more-novel cholesterol-related biomarkers apolipoprotein B and apolipoprotein AI did not add predictive value. Use of nonfasting (vs. fasting) levels did not materially change the associations. The authors conclude that HDL cholesterol is unlikely to relate to stroke risk, although it plays an important role in cardiac disease.
Comment: This large, well-conducted meta-analysis provides convincing evidence that non-HDL cholesterol levels are modestly associated with ischemic stroke but not with hemorrhage. Levels of other lipid markers, including novel apolipoproteins, do not have independent predictive value. These findings appear to validate the approach to stroke prevention of using statin therapy targeted at LDL levels, and they also help allay the concern that low LDL levels might be associated with increased risk for cerebral hemorrhage. They do not support use of niacin, an HDL-raising treatment, for stroke prevention. However, in stroke patients at high risk for cardiac disease (e.g., those with carotid disease), HDL-directed therapies may help reduce cardiac risk. A limitation to the study is that many strokes could not be classified as either hemorrhagic or ischemic, much less by ischemic stroke subtype, confirming the concern that heterogeneity among strokes contributes to difficulty in assessing their association with lipids. Further clinical trials among patients with clearly defined stroke subtypes may help resolve these questions.
Mitchell S. V. Elkind, MD, MS
Dr. Elkind is Associate Professor of Neurology, Columbia University, New York City. He is a coinvestigator on a separate study with this study’s corresponding author.
Published in Journal Watch Neurology January 19, 2010
Citation(s):
The Emerging Risk Factors Collaboration. Major lipids, apolipoproteins, and risk of vascular disease. JAMA 2009 Nov 11; 302:1993.
Original article (JAMA Subscription required)
Medline abstract (Free)