Low Cholesterol Linked to Higher Hemorrhagic Stroke Risk

Batya Swift Yasgur, MA, LSW

April 15, 2019

Women with very low levels of low-density lipoprotein cholesterol (LDL-C) and triglycerides may be more than twice as likely to have a hemorrhagic stroke than women with higher levels, a new study suggests.

Investigators at Harvard University prospectively studied a cohort of almost 28,000 women who had been enrolled in the Women's Health Study over a period of approximately 20 years.

They found that compared to those whose LDL-C levels were 100 – 130 mg/dL, women whose LDL-C levels were <70 mg/dL had more than twice the risk of experiencing hemorrhagic stroke, after adjusting for other factors that could affect stroke risk.

Similarly, women whose triglyceride level was in the lowest quartile had a significantly increased risk for hemorrhagic stroke, compared to women whose level was in the top quartile, after multivariable adjustment.

"I think the main take-home message for physicians is that women with low levels of LDL cholesterol are usually considered at low risk for heart attack and stroke, but they might still have increased risk of hemorrhagic stroke," lead author Pamela Rist, ScD, assistant professor, Harvard Medical School, Boston, Massachusetts, told Medscape Medical News.

"Given the morbidity and mortality associated with hemorrhagic stroke, it is important to monitor these women for other risk factors, such as hypertension and smoking, to lower the risk of hemorrhagic stroke," she said.

The study was published online April 10 in Neurology.

Focus on Women

High levels of LDL-C and low levels of HDL-C are associated with increased risk for ischemic stroke or myocardial infarction, Rist observed, but some prior work has suggested that there may be a link between very low levels of LDL-C and increased risk for hemorrhagic stroke.

"We were interested in doing a study specifically in our cohort, which is a large population of females, since most other studies enrolled both [males and females], but events in females were more limited than they were in our study," she explained.

"Since our study had almost 28,000 women, we wanted to see if the suggestion regarding hemorrhagic stroke would be seen in our cohort as well," she said.

The cohort consisted of women participating in the Women's Health Study. Although that trial ended in 2004, observational follow-up of participants is ongoing.

Of 28,345 fasting blood samples obtained from participants, 27,937 were analyzed for levels of LDL-C, high-density lipoprotein cholesterol (HDL-C), total cholesterol, and triglycerides.

The researchers adjusted for covariates that included age, smoking status, menopausal status, postmenopausal hormone (PMH) status, body mass index, alcohol consumption, history of diabetes and hypertension, physical activity, and treatment with cholesterol-lowering medications.

U-Shaped Curve

Participants in the lowest LDL-C category (<70 mg/dL) were younger and were less likely to have a history of hypertension or to use cholesterol-lowering medications, compared to those with cholesterol levels of 100 – 129.9 mg/dL.

They were also more likely to consume alcohol, be of normal weight, be physically active, and be premenopausal.

By contrast, women in the highest LDL-C category (≥160 mg/dL) were older and were more likely to smoke, to use cholesterol-lowering medications and PMHs, to have a history of hypertension or diabetes, and to be obese.

They were also less likely to consume alcohol or be physically active than those with cholesterol levels of 100 – 129.9 mg/dL.

During a mean of 19.3 years of follow-up, 137 incident hemorrhagic stroke events were confirmed.

Of these, the most common were intracerebral hemorrhage (ICH), followed by subarachnoid hemorrhage (SA) (n = 85 and n = 43, respectively).

Of the 1069 women with an LDL-C level <70 mg/dL, 0.8% had a hemorrhagic stroke, compared to 0.4% of women with cholesterol levels ≥70 mg/dL.

A U-shaped relationship between LDL-C and hemorrhagic stroke risk was found: after multivariable adjustment, compared to those with an LDL-C level of 100 – 129.9 mg/dL, those with an LDL-C level <70 mg/dL had 2.17 times the risk (95% confidence interval [CI], 1.05 – 4.48) of experiencing a hemorrhagic stroke.

There was a suggestion of elevated risk among those with an LDL-C level ≥160 mg/dL, but the increase was not statistically significant.

Participants with an LDL-C level of 70 – 99.9 mg/dL or 130 – 159.9 mg/dL did not have a significantly increased risk for hemorrhagic stroke (relative risk [RR], 1.25; 95% CI, 0.76 – 2.04; and RR, 1.14; 95% CI, 0.72 – 1.80, respectively).

Results for the ICH subtype were similar to those seen for total hemorrhagic stroke (ie, the highest risks for events were found in those with LDL-C <70 mg/dL, followed by those with LDL-C ≥160 mg/dL).

Naturally Low Levels

Women whose triglyceride level was in the lowest quartile (≤74 mg/dL for fasting and ≤85 mg/dL for nonfasting) had a significantly increased risk for hemorrhagic stroke compared to those whose level was in the top quartile, after multivariable adjustment (RR, 2.00; 95% CI, 1.18 – 3.39).

The other quartiles were not associated with significantly increased in risk.

Low triglyceride levels were associated with a significant risk for SAH but not for ICH.

A model that included LDL-C categories as well as triglyceride quartiles found that those women whose triglyceride level was in the lowest quartile had an increased risk for hemorrhagic stroke (RR, 2.14; 95% CI, 1.24 – 3.70), compared to those whose level was in the highest quartile, after controlling for LDL-C.

Conversely, those with LDL-C <70 mg/dL (RR, 2.04; 95% CI, 0.98 – 4.23) and those with LDL-C ≥160 mg/dL (RR, 1.75; 95% CI, 1.05 – 2.92) were at increased risk for hemorrhagic stroke compared to those with an LDL-C level of 100 – 129.9 mg/dL, after controlling for triglyceride levels.

Analyses that were restricted to women who were not taking cholesterol-lowering medications at baseline (n = 27,044) yielded results similar to the those of the main analyses, although the increased risk among those with LDL-C <70 mg/dL was no longer statistically significant.

No significant associations were observed between HDL-C or total cholesterol and risk for hemorrhagic stroke

Rist noted that the potential mechanism responsible for the increased risk for hemorrhagic stroke in women with very low cholesterol levels is not clear.

"It has been suggested that the mechanism may be connected with issues of vessel wall integrity," she speculated.

She noted that their cohort "is unique, in that our blood measurement was performed in the early 1990s, before lipid-lowering drugs, such as statins, were as widely used as they are today."

This is relevant because "we were looking at the long-term effects of low levels [of cholesterol], and women with already low levels are likely different than women who originally had high levels that were lowered through medication use," she said.

Flag of Long-term Safety

Commenting on the study for Medscape Medical News, Erin D. Michos, MD, MHS, associate professor of medicine and epidemiology and associate director of preventive cardiology, Ciccarone Center for the Prevention of Heart Disease, Division of Cardiology, Johns Hopkins School of Medicine, Baltimore, Maryland, who was not involved with the research, noted that there has been "a long-standing controversy whether low cholesterol levels place individuals at risk for hemorrhagic stroke."

The controversy "stems from many epidemiological studies which have found an association between low blood cholesterol and increased risk of hemorrhagic stroke," she said.

The "key point" is expressed by the authors themselves, she pointed out: "individuals with very low LDL-C levels may be less healthy than those with higher LDL-C levels, making them more vulnerable to brain bleeds."

The findings therefore "do not change my practice at all, since I remain skeptical about the residual confounding from a single lipid measurement in an epidemiology study," she said.

However, "if this association is later to be confirmed in Mendelian randomization genetic studies among individuals with lifetime very low LDL-C levels, or if it is found with RCTs of therapies that lower LDL-C, I'd be more convinced," she said.

In her practice, she "only treats patients with statins and lipid-lowering agents who are anticipated to get net benefit for ASCVD [atherosclerotic cardiovascular disease] reduction, such as those at highest risk, [because] statins are not for everyone."

Raffaele De Caterina, MD, PhD, professor of cardiology and director of cardiology at Azienda Ospedaliero-Universitaria Pisana, Italy, who was not involved with the research, had a different perspective.

"I think this is an important piece of evidence because of the long follow-up of this study, by far exceeding what is known from other ones," he told Medscape Medical News.

"It raises some concern about the long-term consequences of very profound LDL lowering, as it occurs with the PCSK9 [proprotein convertase subtilisin-kexin type 9] inhibitors," he said.

"We already had concerns about the cost-effectiveness of intensive LDL-lowering treatment in persons with LDL <70 mg/dL, and this would now be even more reinforced when a flag of long-term safety is now raised."

Rist added, "The impact of low cholesterol as a result of medication on hemorrhagic stroke risk is a different research question, and it's hard to extrapolate from these women who already had very low levels to begin with to people with high levels who have been treated and now have low levels."

Further research is necessary, she emphasized.

The study was supported by the National Institutes of Health. The authors, Michos, and De Caterina report no relevant financial relationships.

Neurology. Published online April 10, 2019. Abstract

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.