乌鸦传媒

NASHVILLE, Tenn. -- Endovascular clot retrieval for acute ischemic stroke improved outcomes over medical treatment with thrombolytics alone, three trials showed here in what was seen as verification that will change usual care.

In the EXTEND-IA trial, more brain was reperfused 24 hours after endovascular thrombectomy in patients getting alteplase (Activase) than those getting the thrombolytic alone (median 100% versus 37% of the initially ischemic area, P<0.001), , of Australia's Royal Melbourne Hospital, and colleagues found.

with the Solitaire FR stent retriever device deployed within 4.5 hours of stroke onset, as marked by 71% versus 40% of patients with a modified Rankin score of 0 to 2 (P=0.01).

In the ESCAPE trial, a similar comparison between usual medical care (mostly with alteplase) alone or with use of available thrombectomy devices within 12 hours of onset showed a significant downward shift in modified Rankin scores with the intervention (29.3% versus 53.0% scoring 0 to 2 at 90 days, P<0.001).

That trial also showed a with the endovascular procedure (10.4% versus 19.0%, P=0.04), , of the University of Calgary, Alberta, and colleagues found.

Both trials were reported at the American Stroke Association's International Stroke Conference and simultaneously online in the New England Journal of Medicine.

The same pattern of downshift in disability (P=0.0002) was seen in the , reported as an invited presentation at the late-breaking session because the trial was stopped just 7 days ago and the abstract didn't make the cutoff time.

The proportion with a good functional outcome at 90 days was 60.2% with Solitaire devices plus tPA versus 35.5% on tPA alone, , director of the UCLA Comprehensive Stroke Center, and colleagues found.

Numbers needed to treat were in the 3-4 range across trials to get one additional patient heading home functionally independent.

Mandate for Practice

These findings were seen as confirmation of last year's MR CLEAN trial, which showed a benefit of adding endovascular thrombectomy versus medical therapy alone.

But they came after a spate of neutral trials -- IMS III, MR RESCUE, and SYNTHESIS -- with the technology, leading some to call for further confirmation before health systems should change their standard practice.

Now with three additional positive trials, there is a clear validation and mandate to treat with stent-retriever procedures for patients with the large vessel ischemic strokes like those selected by CT angiography or perfusion imaging in the trials, said Kyra Becker, MD, press conference moderator and meeting program committee chair.

"I have been a doubter of that technology, but I am convinced," she said.

"This is a real breakthrough," agreed , director of the stroke program at Cedars-Sinai in Los Angeles, who chaired the data safety monitoring board of the largest of the prior failed endovascular thrombectomy trials.

"If you see it once, you think maybe it is a chance event, but when you see three or four times it's been replicated ...." he said. "There've been very few people as critical as me about this procedure, but even I have to take a look at the data and say we have to believe this."

"Behold the age of the stent retrievers," announced , director of the stroke program at the University of Vermont in Burlington, who called the new findings "absolutely amazing."

Lyden predicted a fast implementation, unlike with tPA, for which he led the pivotal trials that led to approval in 1996.

"We had a huge barrier to implementing that therapy; nobody thought before about treating acute stroke. Now we know that acute stroke treatment works, and for the first time we have data that these devices can salvage patients that sometimes the IV treatment can't save."

The technology isn't a barrier now, and comprehensive stroke centers are set up to do embolectomy and have some experience doing endovascular treatments in the brain, such as for aneurysms, Hill noted.

"This is not just about technology," he said. And while it will take some time to get the workflow and teams optimized, "this is not a situation where we have to wait several years. It can be offered today."

The Details

The reason for the shift to positive results was likely multifactorial, all agreed.

"Faster, better, and on-target," Saver summed up. The earlier trials didn't have the same focus on speedy "picture-to-puncture" times or the newer thrombectomy devices that have higher procedural success rates or select patients as accurately from CT angiography or CT perfusion characteristics.

ESCAPE included 316 patients before being stopped early for efficacy, who were randomized to open-label treatment with usual medical care (including alteplase if within the 4.5-hour window and meeting local guidelines for use) alone or along with use of any available endovascular thrombectomy device.

It allowed patients of any age and up to 12 hours after onset of a proximal intracranial occlusion in the anterior circulation and a small core and good collateral flow seen on noncontrast CT and CT angiography (preference for multiphase scans, MRI discouraged for timeliness).

EXTEND-IA included 70 patients before similar early termination for efficacy. Those patients already getting IV alteplase within 4.5 hours after stroke onset were randomized to that treatment alone or with the Solitaire FR device.

It required occlusion of the internal carotid or middle cerebral artery and evidence of salvageable brain tissue and an ischemic core of less than 70 mL on CT perfusion imaging done with the RAPID protocol.

SWIFT PRIME included 196 patients up to age 80 getting IV tPA within 4.5 hours of stroke onset for a small-to-moderate core, large vessel ischemic stroke with CT or MR angiography confirmation. The trial initially required perfusion imaging but switched to an ESCAPE-like protocol to include more centers.

Those patients were randomized to continued medical treatment alone or use of the Solitaire FR or newer generation Solitaire 2 device.

While there were differences among the three, "they're more similar than different," Lyden told 乌鸦传媒.

The vast majority of patients were randomized within 6 hours of stroke onset across trials, because those caught late were less likely to fit the treatment criteria, noted Yvo Roos, MD, PhD, of the Academic Medical Centre in Amsterdam, who was a co-author on the MR CLEAN trial.

However, he suggested that the determinants should be individualized rather than a hard time cutoff.

MR CLEAN didn't even require much by way of imaging to confirm that it was a small core infarct and still showed a benefit, Roos told reporters at the press conference. He suggested that this showed that advanced imaging wasn't necessary, and noted that even with good logistics, every picture costs time.

However, the advantage of thrombectomy over medical treatment alone were roughly doubled in the trials that did use CT selection than in MR CLEAN that didn't, Becker noted.

As to whether tPA is a necessary part of the equation, further studies are needed, Saver argued.

The 10% of patients in MR CLEAN and 25% in ESCAPE who didn't end up getting tPA didn't get less benefit from thrombectomy. The effect size was the same either way and the interaction wasn't significant, Roos and Hill noted.

While Becker concluded that the procedure could be offered to patients regardless of whether they're a candidate for tPA, Saver added that the evidence supports getting both treatments if eligible.

Limitations

The fact that all four trials were stopped early, for either efficacy or because of the emergence of the MR CLEAN results, might have led to overestimation of the benefit.

But there's no doubt that the benefit is real given confirmation across several trials, ESCAPE co-author Mayank Goyal, MD, also of the University of Calgary, told 乌鸦传媒.

And the benefit is likely only to grow as issues like prehospital notification and routing to comprehensive stroke centers get worked out, Roos added.

However, the speed with which the centers achieved treatment once patients arrived in these trials might not pan out as well outside the trial setting, Gorman cautioned. And a limited number of centers are going to be able to offer it, he added.

Saver estimated that 300 such centers would be needed across the country but only roughly 70 are available as yet.

And the procedure will only be applicable to a minority of stroke patients, perhaps bringing the treatable proportion to about a third of stroke patients, up from 20% to 25% with tPA at a well-organized stroke center, Lyden pointed out.

Risks of the procedure were called low across the trials.

EXTEND-IA and ESCAPE showed no differences in symptomatic intracerebral hemorrhage between arms. It actually tended to occur less often with the procedure in PRIME SWIFT (1.0% versus 4.1%), while overall serious adverse events were similar between groups.

Comment