Stroke Prevention: A Lifetime of Lessons — What 44 Years of Research Teaches Us About Protecting Your Brain

Table of Contents

• Key Points
• Why This Research Matters
• Controlling Resistant Hypertension
• Measuring Carotid Plaque Burden
• Nutrition and Stroke Prevention
• Clinical Implications
• Limitations of This Research
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• In a Nigeria study, renin/aldosterone-guided therapy controlled systolic blood pressure in 85% of patients versus 15% with usual care.
• Carotid plaque burden predicted 5-year cardiovascular risk better than risk factor scores in a 2002 report.
• In patients with stiff arteries, diastolic pressure below 60 mm Hg with pulse pressure above 60 mm Hg increased stroke risk 5.85-fold in one study.
• For eGFR below 66 mL/min per 1.73 m², avoiding egg yolk and red meat may reduce toxic intestinal metabolites linked to atherosclerosis.
• Cyanocobalamin may harm kidney patients; methylcobalamin or oxocobalamin appear safer and more effective for lowering homocysteine.

Why This Research Matters

Dr. J. David Spence's journey into stroke prevention began in 1973, during his neurology residency with Dr. Henry Barnett at a time when stroke was devastating families in ways that were entirely preventable. He realized that of the roughly 1,000 patients he had seen with stroke, half were due to hypertension (high blood pressure).

Beyond the small vessel strokes caused by high blood pressure, his team saw approximately 200 patients per year with hypertensive intracerebral hemorrhage — bleeding inside the brain caused by uncontrolled hypertension. "Patients and their families were being devastated by something completely preventable," he recalls. That realization set him on a path to learn how to prevent strokes, leading him through training in internal medicine and a fellowship in clinical pharmacology at the University of California, San Francisco.

In 1976, he returned to London, Ontario, Canada, to practice stroke prevention. Over the next 44 years, he would see more than 40,000 patients at high risk of stroke, and his research would reshape how doctors understand hypertension treatment, atherosclerosis (hardening and narrowing of the arteries), and nutrition's role in brain health.

Controlling Resistant Hypertension

A Very Different Time in Medicine

Until about 1970, medical students were actually taught that older people needed high blood pressure to force blood through their blocked arteries. As a result, hypertension was seldom treated, and a condition called malignant hypertension — severe high blood pressure that damages the kidneys and causes pressure to spiral dangerously upward — was common. This condition carried a very high 2-year mortality rate.

In his first year of practice, Dr. Spence saw 10 patients with hypertensive encephalopathy (brain dysfunction caused by extremely high blood pressure). Two of those patients walked into his clinic without symptoms despite having systolic blood pressures above 300 mm Hg and diastolic pressures of 170 mm Hg. By comparison, a normal systolic reading is below 120 mm Hg.

Remarkably, within just 5 years of opening his hypertension clinic, strokes in London, Ontario, had been reduced by half.

Why Measuring Stimulated Renin Matters

In 1977, Dr. Spence opened the third hypertension clinic in Canada. His approach was based on a strategy proposed by John Laragh: using plasma renin levels — a hormone produced by the kidneys that helps regulate blood pressure — to identify the right therapy for each patient. His Chief of Medicine, Dr. Adam Linton, pointed him to a paper from Keith Dawson's group in Montreal explaining that renin needed to be measured in a stimulated condition (after a dose of the diuretic furosemide) to accurately distinguish between different causes of hypertension.

This technique made it possible to tell whether suppressed renin levels were due to an underlying medical condition or simply from a high salt intake. The following year, in 1978, the Department of Family Medicine launched a major community program to improve hypertension detection and treatment. The results were impressive: by 1983, in Middlesex County surrounding London, Ontario, 94% of hypertensives were detected, 92% were receiving treatment, and 72% were controlled. This virtually eliminated hypertensive intracerebral hemorrhages and markedly reduced lacunar strokes (small strokes caused by high blood pressure damaging tiny blood vessels in the brain).

The Critical Role of Aldosterone

When measuring plasma aldosterone — a hormone that regulates salt and water balance — became possible alongside renin testing, blood pressure control improved even further. Once rare causes of hypertension are ruled out (such as pheochromocytoma, a tumor of the adrenal gland; coarctation of the aorta, a narrowing of the main artery; licorice consumption; and certain kidney tumors), virtually all hypertension is driven by the physiology of the renin/aldosterone/angiotensin system.

Dr. Spence discovered that primary aldosteronism — excessive secretion of aldosterone — was far more common than previously suspected, and that his Black patients were much more likely to have this condition. Among the first 10 patients in his clinic who required adrenalectomy (surgical removal of the adrenal gland) for primary aldosteronism, 4 were Black — a huge disproportion. One came from Africa; the other three were from North Buxton, a nearby settlement established in 1849 for escaped slaves who came to Canada via the Underground Railroad.

Two main types of low-renin hypertension exist, both more common in Black patients:

• Primary aldosteronism phenotype (low renin, high aldosterone): Best treated with aldosterone antagonists such as spironolactone and eplerenone. At least 6 gene variants predispose to this condition: CYP11B2, KCNJ5, ATP1A1, ATP2B3, CACNA1D, and ARMC5.
• Liddle phenotype (low renin, low aldosterone): Caused by overactivity of the renal tubular epithelium sodium channel, leading to salt and water retention. At least 6 gene variants predispose to this condition: SCNN1B (true Liddle syndrome), GRK, NEDD4L, CYP4A11, NPPA, and UMOD. The specific treatment is amiloride, which blocks the sodium channel and is more effective than aldosterone antagonists for this phenotype.

Experts commonly recommend simply adding spironolactone for resistant hypertension, and the PATHWAY-2 trial did show that amiloride is as effective as spironolactone. However, Dr. Spence emphasizes that it is not advisable to simply add spironolactone without identifying the physiological driver — measuring both renin and aldosterone in a stimulated condition is essential to choose the best therapy.

Physiologically Individualized Therapy: Striking Results

Research by Howard and colleagues showed that Black patients are more likely to have their blood pressure diagnosed, more likely to have it treated, and more likely to have it treated intensively — yet less likely to have it controlled. Dr. Spence attributes much of this disparity to what he calls diagnostic inertia: the failure to ask, "What are the physiological drivers of the hypertension in this patient who is not responding to usual care?"

In 2006, Dr. Spence suggested that racial disparities in stroke in the United States could largely be eliminated by physiologically individualized therapy. That clinical trial still hasn't happened in the U.S., but he tested the hypothesis in a study in Africa. Patients with uncontrolled hypertension were randomly assigned to either usual care or physiologically individualized therapy based on renin/aldosterone phenotyping. The results at the Nigerian site were striking:

• Systolic blood pressure controlled: 15% with usual care vs 85% with individualized therapy (P=0.0001)
• Diastolic blood pressure controlled: 45% vs 75% (P=0.11)
• Both systolic and diastolic controlled: 15% vs 75% (P<0.0001)

Dr. Spence states plainly: "It is past time that such a study be performed in the United States."

A Critical Caution: Don't Lower Diastolic Blood Pressure Too Far

After the SPRINT trial (Systolic Blood Pressure Intervention Trial), many guidelines began recommending lower systolic blood pressure targets. However, Dr. Spence warns that there is a subgroup of patients who will be harmed by aiming for a systolic pressure below 120 mm Hg — specifically, patients with stiff arteries who have a marked diastolic J curve.

McEvoy and colleagues reported that patients with a diastolic blood pressure (DBP) below 60 mm Hg and a pulse pressure above 60 mm Hg (DBP<60/PP>60) had a doubling of subclinical myocardial ischemia (reduced blood flow to the heart without noticeable symptoms). Park and Ovbiagele found that this same pattern increased the risk of stroke 5.85-fold.

There are several reasons for this danger:

1. More than half of cerebral blood flow occurs during diastole (the relaxation phase of the heartbeat), so low diastolic pressure means less blood flow to the brain.
2. There is a large pressure gradient within the brain: when blood pressure in the brachial artery (arm) is 117/75 mm Hg, it drops to 113/73 mm Hg in the lenticulostriate artery but plummets to only 59/39 mm Hg in small branches in the posterior parietal subcortex — areas vulnerable to small vessel disease.
3. Patients with stiff arteries are more likely to have a large cuff artifact, meaning the intraarterial diastolic pressure is actually much lower than the cuff measurement suggests.

This understanding of blood pressure gradients in the brain has important implications for treating small vessel disease and avoiding overtreatment in older patients with stiff arteries.

Measuring Carotid Plaque Burden

A Better Predictor of Risk

The measurement of carotid total plaque area was invented in Dr. Spence's lab in 1986 by Maria DiCicco, RVT. The carotid arteries are the main blood vessels supplying the brain, and plaque buildup in them is a powerful indicator of overall cardiovascular health. In 2002, his team reported that by quartile (four equal groups) of plaque area, the 5-year risk of stroke, heart attack, or vascular death was:

• 1st quartile (least plaque): 5.6%
• 2nd quartile: 10.7%
• 3rd quartile: 13.9%
• 4th quartile (most plaque): 19.5%

These figures were adjusted for age, sex, systolic blood pressure, smoking, serum cholesterol, plasma total homocysteine, diabetes mellitus, and treatment of blood pressure and cholesterol. Measuring carotid plaque burden is thus much more predictive of cardiovascular risk than scores based on risk factors alone. It is as predictive as coronary calcium scores — and for several reasons, better than a coronary calcium scan.

Treating Arteries Instead of Treating Risk Factors

In that same 2002 report, Dr. Spence's team made another sobering discovery: despite usual care based on guidelines, half of their patients had plaque progression in the first year of follow-up. Those with plaque progression had twice the risk of cardiovascular events during follow-up.

This recognition led to a paradigm change implemented in their clinics in 2003: treating arteries instead of treating risk factors. The objective of therapy is not simply to achieve target levels of blood pressure and cholesterol — it is to stop progression of plaque or achieve regression of plaque.

A major part of this process involves showing patients images of their own plaque and graphing how severe their atherosclerosis is compared with healthy people of the same age. This visual feedback has been shown to markedly improve medication compliance and improve risk scores. As Dr. Spence puts it: "Treating atherosclerosis without measuring plaque is like treating hypertension without measuring blood pressure."

By 2010, the evidence was compelling. In high-risk patients with asymptomatic carotid stenosis (narrowing of the carotid artery without symptoms), treating arteries was associated with:

• Reduction of microemboli (tiny clots detected by transcranial Doppler, a strong predictor of stroke risk) from 12.6% to 3.7% of patients
• Reduction in the annual rate of plaque progression from 69±96 mm² to 23±86 mm² (P<0.001)
• Reduction of the 2-year risk of stroke from 8.8% to 1%
• Reduction of the 2-year risk of myocardial infarction (heart attack) from 7.6% to 1%

This approach has been adopted in Switzerland, China, and across Argentina, where prevention clinics operated by Blossom DMO in 7 large health maintenance organizations use this strategy. Among patients aged 65 and older, the annual rate of cardiovascular events declined from 5.86% to 2.35% between 2011 and 2015.

Dr. Spence notes that a clinical trial comparing usual care versus treating arteries is still needed, and a grant application for that purpose has been submitted.

Resistant Unexplained Atherosclerosis

In 2012, Dr. Spence described extreme phenotypes of atherosclerosis. In a linear regression model where plaque burden was the dependent variable and coronary risk factors were the predictors, the R² (a measure of how much variation is explained) was 0.57. By saving residual scores for each patient — essentially measuring how far off the regression line each patient fell — his team could identify two remarkable groups:

• Protected phenotype: Patients with high levels of risk factors but little or no plaque buildup
• Unexplained atherosclerosis: Patients with low levels of risk factors but a very high plaque burden

Studying patients at such extremes of atherosclerosis reduces by three-quarters the sample size required for genome-wide association studies, making genetic research far more efficient.

Dr. Spence became interested in homocysteine (an amino acid linked to cardiovascular risk) because of these unexplained patients. In 2016, however, his team reported that high levels of plasma total homocysteine accounted for only 20% of the excess atherosclerosis associated with impaired renal (kidney) function. They then hypothesized that toxic metabolites produced by the intestinal microbiome — and normally excreted by the kidneys — might account for much of the excess atherosclerosis seen in kidney failure.

Toxic Metabolites and the Intestinal Microbiome

After treating arteries in more than 4,000 patients, Dr. Spence's team reported in 2017 that some patients were resistant to therapy, showing plaque progression despite very low levels of LDL-C (low-density lipoprotein cholesterol, the "bad" cholesterol). Among patients with LDL-C below 1 mmol/L (38 mg/dL) — an extremely low level — half still had plaque progression. The reason their LDL-C was so low, ironically, was that their plaque had been progressing, prompting progressively more intensive treatment.

From that study database, they recruited approximately 100 patients in each phenotype (protected, explained, and unexplained). Collaborator Dr. Bradley Urquhart measured 7 intestinal metabolites using ultra-performance liquid chromatography coupled to quadrupole time-of-flight mass spectrometry:

• Trimethylamine-N-oxide (TMAO)
• p-cresyl sulfate
• Hippuric acid
• Indoxyl sulfate
• p-cresylglucuronide
• Phenylacetylglutamine
• Phenyl sulfate

Four of these metabolites were significantly higher in patients with unexplained atherosclerosis: TMAO, p-cresyl sulfate, p-cresyl glucuronide, and phenylacetylglutamine. This finding was not explained by differences in kidney function or by differences in dietary intake of the precursor foods. In backward linear regression analysis, both TMAO and p-cresyl sulfate were significant predictors of plaque burden.

TMAO is formed when the liver processes trimethylamine, which intestinal bacteria produce from carnitine (largely from red meat) and phosphatidylcholine (largely from egg yolk). Hazen's group at the Cleveland Clinic recently confirmed that phenylacetylglutamine is associated with atherosclerosis and elegantly showed that it works by activating adrenergic receptors on platelets, promoting blood clotting.

Nutrition and Stroke Prevention

Egg Yolk, Red Meat, and Kidney Function

The study of intestinal metabolites revealed a critical finding: plasma levels of all 7 metabolites were significantly higher with even moderate impairment of kidney function, defined as an estimated glomerular filtration rate (eGFR) below 66 mL/min per 1.73 m².

Whereas TMAO is produced largely from red meat and egg yolk, most of the other toxic metabolites are produced from amino acids consumed in dietary protein. This leads to a clear dietary recommendation: patients with impaired kidney function should avoid egg yolk and red meat and limit their intake of animal flesh.

This may help explain why the Mediterranean diet — which is mainly vegetarian — significantly reduces the risk of stroke, as do vegetarian diets. Hazen's group reported that replacing red meat with white meat or a nonmeat diet significantly reduced plasma TMAO levels within one month.

Here's the sobering part for older adults: the average eGFR above age 75 is below 66 mL/min per 1.73 m², and it's even lower above age 80. This means the dietary recommendation applies to a large proportion of elderly patients at risk of stroke.

B Vitamins for Lowering Homocysteine

The evidence that high plasma levels of total homocysteine (tHcy) increase cardiovascular risk — along with the many biological mechanisms involved — made it seem implausible that B vitamin supplementation wouldn't reduce stroke risk. But the story turned out to be complicated.

Three major trials published conflicting results:

• VISP trial (Vitamin Intervention for Stroke Prevention, 2004): Reported no benefit of high-dose folic acid, B6, and B12 (given as cyanocobalamin).
• NORVIT trial (Norwegian Vitamin Trial, 2006): Not only reported no benefit, but actually found slight harm in the arm containing high-dose cyanocobalamin.
• HOPE-2 trial (Heart Outcomes Prevention Evaluation 2, same issue of the New England Journal of Medicine): Reported a statistically significant 25% reduction of stroke.

The first author of NORVIT declared that "homocysteine is dead." Dr. Spence writes: "Unfortunately, that pronouncement and its widespread acceptance as truth have caused many strokes that could have been prevented."

The VISP trial had three significant problems that likely masked a true benefit:

1. Folate fortification: Folic acid fortification of the grain supply in North America began at the same time the trial began, partially negating the benefit of supplemental folic acid.
2. No true placebo: Concerned that patients given placebo would take extra vitamins on their own, researchers gave low-dose vitamins instead of placebo.
3. B12 supplementation in both arms: Out of ethical concern, the low-dose vitamin contained the Recommended Daily Intake of B12 (6 µg/day), and patients with low B12 levels at baseline received monthly B12 injections in both study arms — effectively negating any B12 benefit in the very participants who would have benefited most.

To address these issues, a hypothesis-driven subgroup analysis was designed. Researchers excluded patients who received B12 outside their randomized assignment. They also — for what turned out to be the wrong reason — excluded patients in the lowest decile of eGFR, which was 46 mL/min per 1.73 m². (The "wrong reason" was the assumption that dialysis patients wouldn't respond to B vitamins; the researchers never dreamed cyanocobalamin itself could be harmful.)

The results were revealing. Stratifying by the median B12 level at baseline (322 pmol/L), they found that VISP participants with a baseline serum B12 above the median who received high-dose vitamins had a 34% reduction in stroke, heart attack, or vascular death compared with those who received low-dose vitamins and had a baseline B12 below the median (P=0.02).

Then in 2010, two more clues emerged. A trial in patients with diabetic nephropathy (kidney disease caused by diabetes) comparing placebo versus high-dose B vitamins with 1000 µg daily of cyanocobalamin actually showed harm: acceleration of the decline in kidney function and a doubling of adverse events (a composite of heart attack, stroke, revascularization, and all-cause mortality). Additionally, recent research has associated high serum B12 levels — probably due to supplementation with cyanocobalamin — with increased cardiovascular risk in the elderly.

The key insight is that metabolic B12 deficiency is an important cause of hyperhomocysteinemia, and it is often missed because many doctors don't realize that a serum B12 level in the reference range doesn't mean the B12 level is adequate. Only 6% to 20% of serum B12 is active.

Dr. Spence's conclusion is clear: we should be using B vitamins to prevent stroke, but we should use methylcobalamin or oxocobalamin instead of cyanocobalamin — especially in patients with kidney impairment. Studies have shown that in patients with renal failure, methylcobalamin effectively reduces both homocysteine and asymmetric dimethylarginine (another marker of cardiovascular risk), whereas cyanocobalamin did not.

Clinical Implications

What does all this mean for patients and their families? Several important messages emerge from Dr. Spence's decades of work:

• Resistant hypertension needs a personalized approach. If your blood pressure remains uncontrolled despite multiple medications, ask your doctor about renin/aldosterone testing to identify which physiological driver is at work. This is especially important for Black patients, who are more likely to have primary aldosteronism or Liddle phenotype, and who are more likely to have uncontrolled blood pressure despite intensive treatment.
• Measuring plaque is better than guessing. Carotid plaque measurement provides a far more accurate picture of cardiovascular risk than risk factor scores alone, and it can guide treatment decisions. "Treating arteries instead of risk factors" was associated with dramatic reductions in stroke and heart attack risk.
• Don't overtreat blood pressure in patients with stiff arteries. Lowering diastolic pressure too far (below 60 mm Hg) in these patients can actually increase stroke risk significantly.
• Diet matters enormously in the right patients. For those with even modestly impaired kidney function, avoiding egg yolk and red meat and limiting animal flesh could reduce levels of toxic metabolites that drive atherosclerosis.
• Not all B vitamins are created equal. B vitamin supplementation can prevent stroke, but the formulation matters. Cyanocobalamin may actually be harmful in patients with kidney disease, while methylcobalamin or oxocobalamin appear to be safer and more effective.

Limitations of This Research

This is a single-author special report reflecting Dr. Spence's personal experience and research — not a systematic review. Several important limitations should be noted:

• The Africa study showing the benefits of physiologically individualized therapy was conducted at a single site in Nigeria; a U.S.-based clinical trial has not yet been performed.
• A formal clinical trial comparing usual care versus "treating arteries" has not yet been completed, though a grant application has been submitted.
• Some findings, particularly around diet and gut metabolites, come from observational studies that can show associations but cannot definitively prove cause and effect.
• The finding that tHcy explains only 20% of excess atherosclerosis in kidney impairment is based on one study, and the role of intestinal metabolites in atherosclerosis is still an evolving area of research.
• The original article text was partially truncated at the end, so some details of the B12 discussion may be incomplete.

Recommendations for Patients

Based on this research, here are practical steps patients can discuss with their healthcare providers:

1. If your blood pressure is resistant to treatment (uncontrolled despite 3 or more medications), ask your doctor about measuring stimulated plasma renin and aldosterone levels to identify whether you need spironolactone/eplerenone or amiloride.
2. If you are at high risk of stroke or heart attack, ask about carotid plaque measurement to better understand your true risk and whether your treatment is actually stopping plaque progression.
3. If you are over 65 or have known kidney impairment (eGFR below 66 mL/min per 1.73 m²), talk to your doctor about limiting egg yolk, red meat, and animal flesh in your diet.
4. If your homocysteine is elevated, discuss B vitamin supplementation with your doctor — but specifically ask about methylcobalamin rather than cyanocobalamin, particularly if you have kidney problems.
5. If you have stiff arteries (known arteriosclerosis, wide pulse pressure, or advanced age), be cautious about overly aggressive blood pressure targets that push diastolic pressure below 60 mm Hg.
6. Ask about your B12 status more carefully. A serum B12 in the "normal" range doesn't guarantee adequate active B12. Metabolic B12 deficiency is underdiagnosed and treatable.

Frequently Asked Questions

My blood pressure is still high on three medicines. What should I ask my doctor about?

Ask about measuring stimulated plasma renin and aldosterone levels. In a study in Nigeria, patients with uncontrolled hypertension who received therapy based on these measurements had systolic blood pressure controlled in 85% of cases, compared with 15% for usual care. This testing helps identify whether you need spironolactone, eplerenone, or amiloride.

What is carotid plaque burden and why does it matter more than risk factor scores?

Carotid plaque burden measures the amount of plaque in the neck arteries supplying the brain. In a 2002 report, the 5-year risk of stroke, heart attack, or vascular death rose from 5.6% in the lowest plaque quartile to 19.5% in the highest. This measurement is more predictive than scores based on risk factors alone.

Can lowering my blood pressure too much be dangerous?

Yes, for some people. If you have stiff arteries, pushing diastolic pressure below 60 mm Hg with a pulse pressure above 60 mm Hg may be harmful. One study found this pattern doubled subclinical myocardial ischemia, and another found it increased stroke risk 5.85-fold. More than half of cerebral blood flow occurs during diastole, so low diastolic pressure reduces brain blood flow.

I have kidney impairment. Should I change my diet to prevent stroke?

If your eGFR is below 66 mL/min per 1.73 m², consider avoiding egg yolk and red meat and limiting animal flesh. In a study of patients with unexplained atherosclerosis, four intestinal metabolites—TMAO, p-cresyl sulfate, p-cresyl glucuronide, and phenylacetylglutamine—were significantly higher. These metabolites are produced from red meat, egg yolk, and dietary protein, and are normally excreted by the kidneys.

Are all B vitamins the same for stroke prevention?

No. In patients with kidney impairment, cyanocobalamin may be harmful; a trial in diabetic nephropathy showed it accelerated kidney function decline and doubled adverse events. Methylcobalamin or oxocobalamin appear safer and more effective. In a subgroup of the VISP trial, participants with baseline B12 above the median who received high-dose vitamins had a 34% reduction in stroke, heart attack, or vascular death.

What does 'treating arteries instead of risk factors' mean?

It means the goal is to stop plaque progression or achieve regression, not just reach target blood pressure and cholesterol numbers. In high-risk patients with asymptomatic carotid stenosis, this approach reduced the 2-year risk of stroke from 8.8% to 1% and heart attack from 7.6% to 1%. It also cut microemboli from 12.6% to 3.7% of patients.

Why is a 'normal' serum B12 level not always enough?

Only 6% to 20% of serum B12 is active. A level in the reference range does not guarantee adequate active B12. Metabolic B12 deficiency is an important cause of high homocysteine and is often missed. If your homocysteine is elevated, ask your doctor about methylcobalamin rather than cyanocobalamin, especially if you have kidney problems.

My blood pressure is still uncontrolled on three medications — when should I seek a second opinion about resistant hypertension?

When blood pressure stays uncontrolled despite three or more medications, a second opinion can help identify the physiological driver. Stimulated plasma renin and aldosterone testing distinguishes low-renin, high-aldosterone primary aldosteronism, treated with aldosterone antagonists such as spironolactone or eplerenone, from the low-renin, low-aldosterone Liddle phenotype, treated with amiloride. Simply adding spironolactone without measuring both hormones may miss the correct therapy. Physiologically individualized treatment controlled systolic pressure in 85% of patients versus 15% with usual care. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Spence 2020 Stroke Prevention Lifetime of Lessons

Author: J. David Spence, CM, MD, FRCPC, FAHA, Stroke Prevention & Atherosclerosis Research Centre, Robarts Research Institute, Western University, London, Ontario, Canada

Journal: Stroke, 2020;51:2255–2262

DOI: 10.1161/STROKEAHA.120.029679

Key words: aldosterone, atherosclerosis, homocysteine, hypertension, renin

This patient-friendly article is based on peer-reviewed research published by the American Heart Association.