Aspirin can help stem the tide of stroke. Why aren’t more doctors prescribing it?

Cheap and widely available.
Cheap and widely available.
Image: AP Photo/Patrick Sison
We may earn a commission from links on this page.

First sold commercially 119 years ago, aspirin is a remarkable drug. Cheap and with few side effects, its benefits include the ability to prevent blood cells from clotting. As a result, it’s given to patients who have suffered heart attacks, which are caused by clots blocking arteries leading to the heart.

Aspirin can also help prevent strokes—specifically ischemic strokes, the most common kind, which are caused by clots that block the flow of blood in the brain. Giving ischemic stroke victims aspirin can reduce the chances of a second stroke after two to four weeks by about 12%, and by about 17% over the next three years.

But not all stroke patients receive aspirin. For hemorrhagic strokes—the other main form of stroke, cause by a ruptured blood vessel in the brain—patients need their blood to clot to stop uncontrolled bleeding. So doctors never give aspirin to hemorrhagic patients out of fear of causing another stroke.

For decades, if the type of stroke is unknown, standard practice has been to deny aspirin to prevent unintended harm to hemorrhagic stroke victims. CT scans are necessary to make the right diagnosis and guide the course of treatment—but they are often unavailable in rural hospitals and clinics across the developing world. That’s one reason fewer than 4% of stroke patients in low-income countries are on anti-platelet therapies like aspirin, compared to more than half in high-income countries.

Aaron Berkowitz, a neurologist specializing in treating stroke in poor countries, wants to turn that conventional wisdom on its head. In a 2014 paper, he argued the harm caused by giving aspirin to hemorrhagic stroke patients is overstated, and that it make sense to give aspirin to all stroke victims when the type of stroke is unknown.

Berkowitz started with data from pair of massive studies in the 1990s that looked at the effect of aspirin on a total of 40,000 stroke patients. While those studies were designed for ischemic stroke victims, 773 hemorrhagic victims were inadvertently included, enough to provide a baseline understanding of how those patients respond to aspirin.

He then ran computer simulations that modeled the impact of aspirin treatment on a large population of stroke patients when the type of stroke was unknown. In the model, Berkowitz dialed the percentage of hemorrhagic strokes up to 34%, the highest known rate in the world, found in sub-Saharan Africa (in the US, it’s more like 15%).

The results showed that, on balance, giving aspirin to stroke patients reduced a secondary stroke by about eight per 1,000 victims, and reduce deaths by about four per 1,000. While the numbers are small, Berkowitz says, they are statistically significant enough to reconsider how stroke treatment is approached in resource-poor settings.

Stroke is the second-leading killer globally, claiming 5.8 million lives a year, and is both more common and more fatal in low-income countries. Even if aspirin offers only slight improvements in patient outcomes, prescribing it widely could mean saving hundreds of thousands of lives over decades.

Unless there is reason to suspect a hemorrhagic stroke, such as a prior history, Berkowitz recommends giving aspirin after 24 hours when the type of stroke is unknown, even though professional medical associations recommend against it.

“The American Heart Association and American Stroke Association guidelines for treating stroke are very useful,” he says. “There are 100 pages for ischemic stroke and 100 pages for hemorrhagic stroke, but for most countries and for most patients in the world, you won’t know what guidelines to open.”

Until recently, Berkowitz headed the global neurology program at the Harvard-affiliated Brigham and Women’s Hospital, and spent time treating stroke patients in Haiti. He acknowledges there’s a difference between prescribing a treatment for patients at the population level and on an individual basis. While it may make sense to give aspirin to hundreds of stroke patients, giving it to the wrong patient, and seeing them suffer the consequences, can be difficult for a doctor. “Medicine is this constant tension between what you know is globally correct, and tailoring that to an individual patient,” he said.

There is no clear ethical choice, says Christine Mitchell, the director for the center of bioethics at Harvard Medical School. “If you took a direct, utilitarian approach, it’s pretty clear from an aggregate public health perspective, if you can save more lives then you should give aspirin,” she said.

But, according to a duty-based framework (also called deontological reasoning), a doctor’s ethical responsibility is to the patient in front of him or her, not to a theoretical aggregation of patients.

According to that reasoning, “you have a duty not to add to the harm of these patients,” she said. “If there’s substantial identifiable risk in advance, if you know a percentage of your patients will have had hemorrhagic strokes, you should not give them aspirin.”

While the actual risk of giving aspirin to hemorrhagic patient is unclear—the large 1990s trials suggest there was no adverse affect, and more recent trials suggests it may actually benefit patients—most doctors are taught it’s harmful. Changing the course of global treatment is a slow process that could take decades, even if they are endorsed by groups such as the Geneva-based World Health Organization. “People can sit in Geneva and write guidelines and put them online, but a lot of people will never know they exist,” Berkowitz says.

The challenge is clear in Zambia, a fairly typical low-income country with high rates of stroke. Deanna Saylor, an assistant professor at Johns Hopkins who leads a neurology training program in Lusaka, started giving aspirin to patients with unknown types of stroke in the last year. While they haven’t formally studied the results, she believes the current research justifies the new protocol. However, when she presented the evidence at a national stroke conference for Zambian physicians “there was a lot of dismay and reticence about this recommendation, as it is so counter-intuitive and against what we learned in medical school,” she said in an email.

Ultimately, she said, what is needed is a randomized, controlled trial in a setting like Zambia. Until then, a simple, effective drug that can prevent the recurrence of stroke may be overlooked.