Strokes in COVID-19

Learn what experts and the evidence have found so far.


For a little while, stroke was high on the list of COVID-19 concerns for experts. On April 28, the New England Journal of Medicine published a case series of five young patients whose presenting symptom of the virus was a large-vessel occlusion stroke.

Image by Getty Images
Image by Getty Images

“That made everybody very worried that there was going to be a massive number of strokes in young people with COVID. And that actually didn't materialize,” said Brett Cucchiara, MD, associate professor of neurology at the University of Pennsylvania in Philadelphia.

Neurologists' relief at this good news was tempered by several other findings. The novel virus is clearly associated with increased risk of stroke, by mechanisms that aren't yet fully understood. And the need for infection control during the pandemic has hampered every step in stroke care, from diagnosis to discharge.

Overcoming these new challenges to continue providing high-quality care for stroke requires the involvement of hospitalists, stroke experts said.

Risks and recognition

Recent research has tried to quantify the risk of stroke associated with COVID-19. In a study published by Stroke on May 20, researchers in New York looked at how many of 3,556 consecutive COVID-19 inpatients had had an ischemic stroke. They came up with a rate of 0.9%.

“This was slightly lower than other studies, but other studies were more or less in the same ballpark, about 1.5%, 2.2%,” said Shadi Yaghi, MD, lead author of the study and director of vascular neurology at NYU Langone Hospital Brooklyn. “This is reassuring.”

The study also compared the strokes associated with COVID-19 to a control group of strokes treated before the pandemic, and the results were less reassuring. “In our study, they were more likely to be severe strokes, and unfortunately, patients who had strokes in the setting of COVID-19 were more likely to die in the hospital,” Dr. Yaghi said.

Some experts are concerned that mild strokes are being underdiagnosed among inpatients with COVID-19. Dr. Cucchiara estimated the stroke risk in this population at around 1% to 3% from his review of the research.

“It's quite likely that some patients who have mild stroke and are in the hospital with COVID may not be recognized and diagnosed because of the desire to limit nurse and physician exposure to patients. Neurology consultation is often limited or truncated,” he said. “You're really trying to not have a lot of different people having to suit up to go in to see the patients.”

In severely ill patients, physicians face the added difficulty that symptoms may be hard to spot even when in the room. “If they're very sick, you don't have much of an exam, because they are being sedated to tolerate ventilator settings,” said Thabele (Bay) Leslie-Mazwi, MD, assistant professor at Harvard Medical School and a neurocritical care and neuroendovascular specialist at Massachusetts General Hospital in Boston.

The pandemic may also have affected the threshold for acting on slight suspicion of a stroke. “Early on, there was a lot of hesitancy to do MRI scans, which are ultimately the best diagnostic tool we have for stroke. And that hesitancy was due to concerns about wheeling patients through the hospital and potentially contaminating other regions of the hospital, [or] contaminating an MRI scanner,” said Dr. Cucchiara.

At the same time, when COVID-19 cases were first on the rise, many hospitals also had significant competing priorities to stroke diagnosis. “The number of patients they had was just tremendous. They were really focused on keeping people alive, managing their respiratory failure,” said Dr. Cucchiara. “All of those factors kind of contribute a little bit to the uncertainty around how often you see stroke in COVID.”

One particularly confusing aspect is that overall, hospitals saw significantly fewer strokes than normal. In the 10 weeks following declaration of the COVID-19 national emergency, ED visits for stroke declined 20%, according to a study published by Morbidity and Mortality Weekly Report on June 22.

“I think it was related to the fact that people were afraid to come to the hospital because they didn't want to get infected, and so they only came if they were severe,” said David Hess, MD, dean of the Medical College of Georgia and a professor of neurology in Augusta, Ga. “I talked to a patient that didn't want to come in with a [transient ischemic attack] a few weeks ago, but they eventually came in.”

That dynamic is problematic in a highly time-sensitive condition like stroke, noted Dr. Yaghi. “By the time they come, they're outside of this window for treatment. So we have launched an education platform to try to educate people that stroke, heart attack is an emergency,” he said.

Identify and act

Hospitalists, meanwhile, have to educate themselves to provide speedy response to stroke in COVID-19 inpatients.

“Pupil checks should be part of routine evaluation of a patient that's heavily sedated for the ventilator,” said Dr. Leslie-Mazwi. “There should be a threshold that is low to image these folks if there is a neurologic concern, and that concern could be something focal, like a weak limb, for example. It could also be something global, like a patient who is acting delirious or is slow to wake up after the sedation is weaned following improvement in their [acute respiratory distress syndrome].”

Dr. Cucchiara agreed. “Don't be scared of involving your neurologist and getting MRI scans in these patients,” he said. “The faster you get your stroke team involved, the better, particularly in the setting of COVID, because there are some unavoidable challenges and delays in dealing with the patients because of infection control issues.”

Precautions against the spread of COVID-19 have affected many aspects of emergency stroke response. “Organizations are trying to figure out how we can still maintain the time efficiency which is critical for having a favorable outcome, while valuing the precautionary measures as well,” said Adnan Qureshi, MD, a neurologist with the University of Missouri in Colombia, Mo.

“There are a lot of logistics involved in the protection of the staff, the cleaning of the scanner,” said Dr. Leslie-Mazwi. “There will inevitably be some delay built into the process if we have to use PPE [personal protective equipment] and take precautions for COVID-19.”

The delay applies both to inpatients who have a stroke and those who arrive with stroke symptoms and uncertain viral status. For incoming patients, even rapid tests don't provide results soon enough to guide emergency care. “You would like to know if they're COVID-positive or not to know how much you have to clean the room when you're done. It becomes important when you're taking them for interventions, so you eventually want to get them tested, but you can't wait for the answer,” Dr. Hess said.

Limitations on hospital visitors may contribute to the difficulty of finding out whether a patient has had COVID-19 symptoms, noted Dr. Qureshi. “If the patient is unable to provide information because of the stroke, then we try to get in touch with people who have been in contact with the patient,” he said.

Of course, the safest solution is to treat incoming patients as positive. “Here in Massachusetts we're in an endemic area. We assume everyone has it until they rule out,” said Dr. Leslie-Mazwi. “Everyone's in protection from the anesthesia team all the way down.”

One point of debate has been whether to intubate all stroke patients before interventional therapy. “There are groups that have made that move, the idea being that you control the airway, you control what's being aerosolized a little bit better,” said Dr. Leslie-Mazwi. Downsides of that strategy are the aerosols released during intubation and the need to keep patients on a vented unit after the procedure. “We opted not to change our approach, which is the use of conscious sedation for thrombectomy,” he said.

Another possible tweak to the process for patients whose COVID-19 status is unknown at the time of stroke evaluation uses scanning to solve both questions at once. “We get the neuro imaging, but as part of the imaging, we also do imaging of the chest or lungs to see if there's any signs that would make us cautious that the patient may actually have a COVID-19 infection,” said Dr. Qureshi.

But in general, COVID-19 stroke patients get the usual stroke treatment. “There's no contraindication to use tPA [tissue plasminogen activator], which is the standard of care for stroke patients,” said Dr. Hess. Patients with COVID-19 are also candidates for thrombectomy, although there's some concern that their clots are more fragile and prone to fragmentation, reported Dr. Leslie-Mazwi.

That's one of many uncertainties in this area, the experts said. “While today the protocols are essentially the same, so COVID-19-infected patients with stroke will get essentially the same treatment, what we don't know is whether these treatments will still be as beneficial,” said Dr. Qureshi.

There's also uncertainty about the causes of the strokes, and therefore, what kind of antithrombotic therapy is appropriate as primary or secondary prevention. It seems likely that some of the strokes in COVID-19 patients relate to the hypercoagulability that's been found with the disease generally, the experts said, but that's not the whole story. Patients with COVID-19 have shown a variety of other neurological symptoms.

On the other hand, the causes of these strokes are often obvious. “One of the things that we found in our research at Penn is that in most of the patients with stroke related to COVID-19, it's actually not something super exotic and specific to COVID-19 that caused the stroke,” said Dr. Cucchiara. Patients often had atrial fibrillation, diabetes, or other typical risk factors.

“Not to say that COVID isn't contributing in some way, but just that it's very important to do the same things you would do for every stroke patient,” he added. “Do the usual diagnostic testing to figure out why they had the stroke. You shouldn't just chalk it up to COVID.”

The reverse is true, too. If your stroke patient might have COVID-19, remember to look beyond the brain.

“We should make sure that our evaluation is not a focal evaluation, like we've done for other stroke patients. . . . The recommendations are to actually make sure we have a good evaluation of the lungs, the liver, the kidneys, so all of that information is available,” said Dr. Qureshi, who was lead author on international advice on stroke management in COVID-19, published May 10 by the American Journal of Emergency Medicine.

Ongoing care

Of course, the virus complicates all types of evaluation, before and after stroke treatment. “For all the subsequent tests that patients have to go through, transthoracic echo, for instance, and the 24-hour scan, the impact of a pending COVID diagnosis is felt. That gums up the workings of the hospital in general,” said Dr. Leslie-Mazwi. “Physical therapy, occupational therapy, speech and swallow evals—again, you have a lot more PPE use.”

COVID-19 status may also affect the discharge destination after a stroke hospitalization. “There has been some variation in how that's handled,” said Dr. Qureshi. “We are seeing a probably longer hospitalization to make sure the patient's infection has resolved.”

Many rehabilitation facilities are prepared to deal with the virus, however. “Our rehab actually is not only taking patients with COVID-19 who have neurological problems, but they're also rehabilitating patients who have debility from COVID-19,” said Dr. Yaghi.

That's one of many ways in which the stroke care system is gradually catching up with the virus. For example, the time added to stroke response by infection control is shrinking and should continue to do so, according to Dr. Leslie-Mazwi.

“The same way that there's a door-to-needle time that's acceptable, there will be an amount of time that's added to your process, say 20 or 30 minutes, because you are screening and evaluating for potential COVID-19,” he said. “Like any process, the more you do it, the more lubricated it becomes, the faster it becomes.”

Improvement could be accelerated by collaboration among researchers and clinicians. Recent North American guidance on protecting stroke teams from infection “caused quite a stir in our community” when it was published by Stroke on April 1, Dr. Leslie-Mazwi said. “But actually the Chinese Interventional Federation had a very similar set of recommendations that they put out before the Canadian/American publication, and that few of us were aware of. It was a good reminder that when you have a problem this widespread . . . collaboration reduces the need to each reinvent the wheel.”