Ischemic stroke patients with lower levels of low-density lipoprotein cholesterol, including those on statins, may be at increased risk of developing a brain hemorrhage after recanalization therapy, regardless of other risk factors, according to a March 6 study in Neurology (68:737–742).
Researchers at the University of California-Los Angeles (UCLA), and Sungkyunkwan University and Samsung Medical Center, in Seoul, Korea, evaluated the independent effect of statin use and admission cholesterol level on the risk of symptomatic hemorrhagic transformation (sHT) after recanalization therapy for acute ischemic stroke.
Ischemic stroke patients with lower LDL cholesterol had a significantly higher risk of sHT after treatment with tissue plasminogen activator (t-PA), angioplasty, or clot retrieval, independent of concurrent statin therapy or smoking. A 3.2 percent decrease in risk was found for every 1 mg/dL increase in LDL cholesterol.
The findings should be interpreted with caution until further research has been conducted because of limitations in the study design, according to lead author Bruce Ovbiagele, MD, assistant clinical professor of neurology at the UCLA Stroke Center.
“The benefits of statin therapy far outweigh any risk. We're not in any way advocating not taking statins, but our study suggests that something is going on and we'll need new studies to see if the risk is real and if there is an LDL threshold below which the risk increases,” he told Neurology Today. “We don't want to alter patient care or treatment protocols, we just want to find out if there is a risk.”
Prior studies have shown that pre-stroke statin use may improve ischemic stroke outcomes, yet there is also evidence that statins and extremely low cholesterol levels may increase the risk of intracranial hemorrhage.
Some treatments for ischemic stroke aimed at dissolving the occluding blood clot can increase the likelihood of cerebral hemorrhage. Hemorrhagic transformation occurs when blood vessels weakened by ischemic stroke rupture and cause bleeding. This can happen even without antithrombotics.
The researchers analyzed data on 104 consecutive ischemic stroke patients, in a prospectively maintained registry, who underwent IV or intra-arterial fibrinolysis or endovascular embolectomies during an eight-month period in 2006.
Male sex, hypertension, statin use, low total cholesterol and low LDL levels, current smoking, elevated glucose, and higher admission NIH Stroke Scale (NIHSS) scores were all associated with a greater risk of sHT, they found. However the elevated risk in low LDL patients was independent of any of these other factors after univariate analysis.
Low LDL cholesterol was associated with an odds ratio (OR) of 0.968 per 1-mg/dL increase) current smoking (OR, 14.568), and higher NIHSS score (OR, 1.265 per 1-point increase) were independently associated with sHT risk.
“How low is too low in these patients? That's the big question,” said Dr. Obviagele. “In cardiac literature, LDL has been reduced to the mid-40s to 50 without bleeding problems, but cardiac patients are different from stroke patients. Stroke patients are more susceptible to bleeding.”
The researchers are next conducting a prospective study in a larger randomized patient population.
FURTHER RESEARCH NEEDED
Eric Smith, MD, assistant professor of neurology and associate director of Acute Stroke Services at Massachusetts General Hospital and Harvard Medical School, agreed that the findings are limited in terms of clinical use.
“The lower a patient's LDL, the greater the risk of symptomatic bleeding is an interesting observation, but it was in a small number of patients.” He told Neurology Today in a telephone interview, “These patients [with low LDL] were also older, had more diabetes, more severe strokes, and more anti-platelet and anti-coagulant therapy. All of these might affect bleeding, but even after they adjusted for these factors, LDL was a strong predictor.”
Dr. Smith noted that the findings “are especially interesting” in light of the recent Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial, in which statin therapy was associated with an increase in hemorrhagic strokes in ischemic stroke and TIA patients (N Engl J Med 2006;355:549–559).
“In SPARCL there was higher risk of hemorrhages in the patients treated with high-dose atorvastatin, and that study, together with epidemiological studies, suggests some sort of link between low LDL cholesterol and bleeding, but we don't understand the biological mechanism involved. That's the next frontier,” he said.
As many as 30 percent of stroke patients are taking statins at the time of admission, Dr. Smith observed. “I don't think the potential risk of brain hemorrhage after reperfusion therapy is a reason for not staying on LDL-lowering drugs, but we'll want to know if that's true. It's really hard to sort out whether the risk of bleeding is due to low LDL levels or statins because the two go hand-in-hand.”
He also noted that the researchers “imputed” missing data on stroke-related variables in patients when such information was unavailable, a major limitation of the study.
“‘Imputation’ is a statistical term for guessing, and is not a good substitute for the actual information,” he commented. “These findings are preliminary. The highest risk of reperfusion-related hemorrhage seems to be within the first 24 hours, so it may be reasonable to withhold statins during that time period. But I don't think we should make any changes yet in whether these patients should remain on statins during the rest of the hospital stay and at discharge.”
Larry B. Goldstein, MD, professor of neurology at Duke University and Durham VA Medical Center and director of the Duke Center for Cerebrovascular Disease and the Duke Stroke Center, in Durham, NC, wrote an editorial accompanying the study in Neurology.
“An observational study can be useful for identifying potential risk factors, especially when the finding is supported by other studies,” he said. “Whether risk factor modification alters outcomes can only be established through properly controlled, prospective studies.”
Data from other studies support a relationship between low LDL cholesterol and the risk of hemorrhagic infarction, he added, noting that epidemiologic studies generally find that lower cholesterol levels are associated with an increased risk of brain hemorrhage, even without recanalization therapy.
For example, the Asia-Pacific Cohort study showed a 20 percent decreased risk of hemorrhagic stroke per 4.5 mg/dL increase in cholesterol levels.
“It is not possible, however, to determine whether the effect reported from epidemiologic studies is related to an increase in hemorrhagic transformation of ischemic infarctions, primary hemorrhages, or both,” he observed. “Unless sequential brain scans are available, differentiating the two types of parenchymal hemorrhages can be difficult.”
If the new findings can be verified in other studies, understanding the mechanism by which low cholesterol (and low LDL cholesterol) increases the risk of hemorrhagic infarction might lead to treatments to reduce this complication, he noted.
BENEFITS OF STATINS
A 2005 meta-analysis of over 90,000 subjects, predominately with coronary heart disease, enrolled in statin trials found a 19 percent proportional reduction in ischemic strokes with no difference in hemorrhagic transformation. In addition, the 2006 SPARCL trial found high dose statin therapy was associated with an increase in primary hemorrhage in patients with recent stroke or TIA and no known coronary heart disease.
“But this, too, was based on a post-hoc analysis,” he said. “Therefore, there remain no data showing that statin treatment increases the risk of hemorrhagic infarction.”
Dr. Goldstein, an investigator in the SPARCL study, said: “In SPARCL, there was not only a significant 16 percent reduction in the combined risk of nonfatal and fatal stroke with treatment, but also profound reductions in other major vascular events”.
“The available data clearly support the use of statins in patients with a history of cerebrovascular disease,” he told Neurology Today in a telephone interview. “Whether low LDL cholesterol levels increase the risk of hemorrhagic infarction in patients undergoing recanalization therapy requires verification. Even if true, the increased chances of bleeding associated with this and other identified contributors to risk such as current cigarette smoking would need to be balanced against the benefits of restoring circulation to ischemic brain,” he said.
“There may be increased risk, but the risk is far overwhelmed by the data we have showing statin use, with or without low LDL, significantly reduced risk, including subsequent cardiovascular events in first-time ischemic stroke patients. If you don't have a stroke in the first place, you don't have the problem.”
ARTICLE IN BRIEF
Ischemic stroke patients with lower LDL cholesterol had a significantly higher risk of symptomatic hemorrhagic transformation after treatment with tissue plasminogen activator, angioplasty, or clot retrieval, independent of concurrent statin therapy or smoking. A 3.2 percent decrease in risk was found for every 1 mg/dL increase in LDL cholesterol.