Stroke care continues to evolve thanks to AI, cardiologists and more
Stroke care has made big advances in the past decade, moving from thrombolytics to interventional thrombectomy. The time to reperfusion and clinical outcomes have also been improved thanks to artificial intelligence and the creation of acute care stroke teams.
Cardiovascular Business spoke with Gregg C. Fonarow, MD, director of the Ahmanson-UCLA Cardiomyopathy Center, co-director of the UCLA Preventative Cardiology Program and Eliot Corday Chair in Cardiovascular Medicine and Science at UCLA, who explained this movement to interventional stroke care and how the American Heart Association (AHA) Get With The Guidelines-Stroke program evolved to include interventional thrombectomy to speed large vessel occlusion (LVO) strokes. He also examined the growing role of artificial intelligence (AI) and how cardiology is making an impact on stroke care teams.
Stroke patients have traditionally received intravenous tissue plasminogen activator (tPA) to dissolve clots. But the drug needs to be administered in a short time period or it in not effective. Using thrombectomy catheters to remove clots directly from LVOs in the brain can be a complementary therapy used with tPA, Fonarow said.
“The real problem was a lot of patients would arrive too late. The clock starts at symptom onset, and for IV thrombolytics, the window is very short—within four and a half hours," Fonarow explained. "The early data was really not compelling, but then a series of trials came out that were really were compelling. So it became a standard of care. We were able to integrate that into Get With the Guidelines-Stroke set time metrics with door-to-reperfusion time and we were able to see improvements."
The shift was relatively quick as a result of more and more research being released.
"Four trials hit, all within a 12-month period, all individually showing benefit for functional outcomes. It has been a much more compressed timeframe than percutaneous coronary intervention (PCI), which evolved slowly over decades,” Fonarow said.
But these procedures are more sophisticated than just having a cath lab; you need qualified interventional radiologists or neurologists who know how to use the various tools and deal with complications. He said this is where cardiology is making contributions to stroke care and may be key to helping expand these stroke interventions beyond large academic research hospitals in urban areas.
"There are so many more interventional cardiologists than interventional radiologists, so there's going to have to be further investment in ensuring a broader geographic approach so this therapy can be offered to not just highly resourced hospitals and only those patients fortunate enough to be adjacent who can really benefit from it," Fonarow said.
He said scaling the infrastructure, training staff and ensuring that best practices are shared are all important components. Pre-hospital coordination is also critical, so EMS personnel need training to identify LVO stroke symptoms and quickly route patients to the right hospital that can offer thrombectomy. “Time lost is brain lost,” Fonarow emphasizes.
AI accelerating door-to-reperfusion times
Another game-changer in stroke care has been the rise of AI. Among some of the first commercially successful AI tools cleared by the FDA were stroke alerts systems offered by several vendors. The AI reads scans directly off a CT scanner and can send alert notifications with patient imaging to the entire stroke team before the images are even loaded into the picture archiving and communication systems (PACS). The AI flags potential stroke cases in real time and greatly speeds up the diagnosis by a human physician.
“Hospitals using AI are seeing a real impact on reducing door-to-intervention times,” Fonarow said. “We’re in the process of analyzing outcomes across multiple centers, but early indications suggest that AI can play a vital role in decision support and care coordination.”
He added that AI has shown potential to improve care in many other ways.
"AI can be of assistance for clinician judgment and interpretation, but also provide more real-time prompts about where their care deviations and to help them intervene before that patient's missed that window for getting the right therapy at the right time at the right dose," he explained.
AHA now has a pilot program to review the use of AI stroke alert systems to see from a data standpoint it they are making a difference in outcomes. Fonarow said the goal is to look collectively beyond the single center data to make meaningful insights across stroke care at various centers.
Stroke teams are a model of interdisciplinary collaboration
Modern stroke care is defined by multidisciplinary collaboration. “It’s not just neurologists and cardiologists,” Fonarow said. “Emergency physicians, nurses, radiologists, neurointerventionalists, intensivists and rehabilitation teams all play essential roles.”
Fonarow says today’s stroke care teams are “teams of teams” supported by real-time data, shared learnings and a culture of continuous improvement.
“Through Get With The Guidelines Stroke, we have national webinars, data sharing, and benchmarks that allow teams to learn what’s working elsewhere and adapt it locally,” he said.
Another trend to watch is the fact that cardiologists are playing a much more significant role in stroke care, because many stroke patients have cardiac comorbidities or cardio-embolic sources of stroke. He said cardiologists bring valuable expertise in managing these risks and improving secondary prevention.