研究：LDL并非总是预测中风风险最好的指标思维导图

By Caroline CasselsMedscape Medical News

Elevated levels of serum triglycerides and non–high-density lipoprotein (HDL) have been associated with an increased risk of large artery atherosclerotic (LAA) stroke, suggesting that low-density lipoprotein (LDL) may not always be the best predictor of atherosclerotic vascular risk.

A study by investigators at the University of California, Los Angeles Medical Center shows that individuals with the highest triglycerides are 2.7 times more likely to have LAA, while those with the greatest non-HDL levels had a 2.4-fold increased risk of large artery stroke.

"LDL has been the primary target for reducing the risk of stroke, but these results show other types of cholesterol may be more strongly linked with stroke risk," principal investigator Bruce Ovbiagele, MD, said in a statement from the American Academy of Neurology.

The study is published online December 26 in Neurology.

Conflicting Evidence

According to the authors, there is conflicting evidence with respect to the importance of circulating serum lipids in stroke, with some early studies reporting no association between stroke risk and elevated cholesterol. However, they note, this research generally did not distinguish between stroke subtype or lipid subfractions.

More recent studies have demonstrated an association between serum cholesterol levels and ischemic stroke, but few have examined the relationship between serum lipids and subtype of ischemic stroke most directly linked to lipid metabolism.

"This may be important in light of evidence that the effect of plasma lipids on atherogenesis in the cerebral vascular bed may be distinct from that in the coronary vascular bed," they write.

The authors point out that current national US guidelines, which emphasize the management of LDL cholesterol, are largely based on evidence from the cardiac literature.

However, they note, "there is mounting evidence that other serum lipid indices derived from the classic lipid profile may be better predictors of vascular risk than LDL."

Large Study

All patients had fasting lipid panels drawn early in the morning the day after hospital admission. Total cholesterol, triglycerides, HDL, and LDL were assessed. In addition, non-HDL cholesterol, total cholesterol-to-HDL ratio, triglyceride-to-HDL ratio, and LDL-to-HDL ratio were also assessed.

Based on their clinical syndromes, diffusion-weighted imaging data, and the results of vascular, cardiologic, and laboratory studies, patients were divided into 2 groups — LAA and non-LAA.

Overall, 247 patients were in the LAA group and 802 classified in the non-LAA group. The non-LAA group consisted of 395 patients with cardioembolism, 224 with small vessel disease (SVD) stroke, 89 patients with other determined etiology, and 94 patients with cryptogenic stroke.

LDL Should Not Be Discounted

The study showed lipid levels were similar between LAA and SVD patients. Furthermore, total cholesterol, triglycerides, LDL, non-HDL, and triglyceride/HDL ratio were significantly higher in LAA than in non-LAA, non-SVD patients.

After adjusting for age, hypertension, diabetes, smoking habits, body-mass index, and the prior use of statins before onset of symptoms, the investigators found a strong association between triglycerides and non-HDL and LAA. However, there was no association between LAA and LDL.

"Our data indicate that other traditional serum lipid measures such as the triglyceride level, as well as less commonly assessed indices derived from the routinely ordered serum lipid panel such as non-HDL cholesterol, may be more strongly associated with symptomatic large artery cervicocephalic atherosclerosis (vs all other ischemic stroke subtypes) than LDL level," the authors write.

In light of these findings, the authors suggest when treating patients with or at risk for atherosclerotic stroke, clinicians may want to consider focusing additional attention on other aspects of the lipid panel. However, they note, LDL and its role in patients' overall health should not be discounted.

The authors report no conflicts of interest.

Neurology. Published online December 26, 2007.