Conference Coverage

Patient selection for acute stroke thrombectomy stirs controversy



– A little more than a year ago, results from the DAWN and DEFUSE 3 trials substantially broadened the time window for endovascular thrombectomy of acute ischemic stroke by selecting patients using brain imaging. Stroke clinicians are now trying to reconcile widespread, routine use of this life-changing treatment against an uncertain need to replicate the higher-end perfusion CT and analytical software imaging that these landmark trials used for patient selection. This has produced a schism in what experts advise for using endovascular thrombectomy on acute ischemic stroke patients.

Dr. Michael D. Hill, professor of clinical neursciences, University of Calgary, Canada

Dr. Michael D. Hill

“Go open the artery, people!” Michael D. Hill, MD, exhorted during a talk at the International Stroke Conference, sponsored by the American Heart Association. “Don’t get over-selective; more people will benefit than you think,” said Dr. Hill, a professor of clinical neurosciences at the University of Calgary (Alta.).

“We are over-selecting, and depriving patients,” commented Raul C. Nogueira, MD, professor of neurology at Emory University in Atlanta and a lead investigator of the DAWN trial, speaking from the audience during a discussion at the session where Dr. Hill spoke.

Dr. Raul G. Nogueira

Dr. Raul G. Nogueira

“The prevalence of treatable [acute ischemic stroke] patients is far higher than the prevalence of patients who are not good candidates, so you just want to exclude the ‘wipe-outs’; that’s what we do,” Dr. Hill explained. “Fortunately, endovascular therapy is very safe, and you’re not going to harm many patients. With other treatments [used routinely in medicine] some patients don’t benefit, but when you have a large effect size we use the treatment on almost everyone. The effect size from thrombectomy is so large it’s an argument to treat almost everyone, although the patients in the trials were selected by imaging.”

Dr. Hill repeatedly stressed that for most patients a non-contrast CT image is usually adequate to identify patients with salvageable brain tissue and a low risk for hemorrhage from intervention, and he endorsed also doing CT angiography to further inform the diagnosis. But he dismissed CT perfusion imaging as unnecessary. “Noncontrast CT and CT angiography are more than adequate to make treatment decisions,” he said. “The prevalence of poor collaterals is quite low, about 10%,” which means that about 90% of acute ischemic stroke patients will have more slowly progressing infarction,” making them amenable to treatment in an expanded time window and boosting the volume of salvageable tissue.

Dr. Maarten G. Lansberg, neurologist, Stanford (Calif.) University Mitchel L. Zoler/MDedge News

Dr. Maarten G. Lansberg

But these appeals for more liberal use of thrombectomy without the perfusion CT imaging used in DAWN (N Engl J Med. 2018 Jan 4;378[1]:11-21)and DEFUSE-3 (N Engl J Med. 2018 Feb 22;378[8]:708-18) received push back. Maarten G. Lansberg, MD, a co-investigator on the DEFUSE 3 trial, highlighted the speed and simplicity of CT perfusion imaging, and its utility in helping to better target thrombectomy to the right patients. It’s “speedy, simple, and safe,” it “excludes patients who will not benefit” from thrombectomy, and it helps when the patient’s history and noncontrast CT images are inconclusive, said Dr. Lansberg, a neurologist at Stanford (Calif.) University.

Dr. Marc Fisher, professor of neurology, Harvard Medical School, Boston Mitchel L. Zoler/MDedge News

Dr. Marc Fisher

“Clinical presentation will only tell you so much.” With imaging that includes CT perfusion, “you can find out, in 5, 10 minutes, whether there is an occlusion, its location, the extent of dead tissue – that’s all really helpful,” said Marc Fisher, MD, professor of neurology at Harvard Medical School in Boston. “There is a tension now between doing treatment really fast and the concept of slow and fast evolvers. For slow evolvers, the concern about speed is irrelevant because it can take days” for their brains to have substantial damage. “For the fast evolvers, time matters, but they could also possibly be harmed; that’s why we need more data.”

Dr. Pooja Khatri of the University of Cincinnati Mitchel L. Zoler/MDedge News

Dr. Pooja Khatri

A pitch for more data also came from Pooja Khatri, MD, who also spoke at the session. “There is a real tension now between personalizing the imaging and figuring out exactly the right patients against the time trade off for doing that. Some argue to keep it simple and move fast, and by doing that you’ll wash out any difference from doing more fancy stuff. Plus some places, even in developed countries, can’t afford the image-processing software” used in the DAWN and DEFUSE 3 trials. The correct approach remains unclear and has created “an area ripe for a trial,” declared Dr. Khatri, professor of neurology at director of acute stroke at the University of Cincinnati.

Dr. Hill has received honoraria from Merck and received research funding from Boehringer Ingelheim, Covidien, Medtronic, and Stryker. He has an ownership interest in Calgary Scientific and holds a patent on acute stroke triage methods. Dr. Nogueira has financial relationships with many companies. Dr. Lansberg and Dr. Fisher had no disclosures. Dr. Khatri has been a consultant to Lumosa and has received research funding from Cerenovus/Johnson & Johnson, Genentech, and Nervive.

Next Article: