Conference Coverage

CT opens extended window for stroke thrombolysis

View on the News

EXTEND makes delayed thrombolysis feasible for more U.S. centers

Ever since results from the WAKE-UP trial came out earlier in 2018, we at the University of Cincinnati have been imaging acute ischemic stroke patients who presented outside the standard 4.5-hour time limit for thrombolysis with MRI to see if they qualify for an extended window for thrombolysis. But this has been a cumbersome and redundant process because our default imaging method is CT, so we have been imaging potential candidates for an extended thrombolytic window twice, first with CT and then later with MRI.

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

Dr. Pooja Khatri

The new findings from EXTEND now appear to make MRI redundant after perfusion CT. The results give all U.S. stroke centers the opportunity to treat patients with thrombolysis outside of the conventional time frame even when MRI is not available.

The EXTEND findings also provide a further reason for U.S. stroke centers to purchase and use some type of imaging processing software if they don’t already have it, either the RAPID software that was used in EXTEND or one of the several similar software packages that are now available. Several primary stroke centers in my area still do not currently use this software, although its use has been quickly spreading and it will now be increasingly hard for these centers to hold off acquiring it. Fortunately the increased competition among software vendors who sell this type of software has meant that the price has been dropping.

Pooja Khatri, MD , is a professor of neurology and director of acute stroke at the University of Cincinnati. She has been a consultant to Biogen, Greenwich, and PTC Therapeutics. She made these comments in an interview.



– An extended time window for thrombolytic treatment of acute ischemic stroke patients using tissue plasminogen activator out to 9 hours from stroke onset was safe and effective using CT perfusion imaging and automated imaging processing software to select suitable patients in the EXTEND trial. This result matches the groundbreaking finding reported earlier in 2018 that used MRI to select patients for extended thrombolysis.

Dr. Henry Ma

“To reproduce our results you need to set up CT perfusion” as well as the RAPID software for automated image processing to identify patients with a small infarct core and a large area of salvageable brain, said Henry Ma, MD, a stroke neurologist at Monash University, Melbourne, as he reported results from the trial at the World Stroke Congress. “EXTEND is the first positive thrombolysis trial in an extended time window using automated penumbral imaging.”

The new finding, from a trial with 225 randomized stroke patients, was especially notable because, by showing the validity of CT imaging for patient selection, it makes applying the extended time window for thrombolytic therapy more feasible for U.S. and Canadian stroke centers where CT imaging is much more common than MRI. A report from European investigators published in August 2018 from the WAKE-UP trial showed that thrombolysis with tissue plasminogen activator (tPA) was safe and effective when administered to patients who woke up with an acute ischemic stroke that had occurred more than 4.5 hours before treatment, but this study exclusively used MRI for patient selection (N Engl J Med. 2018 Aug 16;379[7]:611-22).

“In North America, our systems are more equipped for using CT,” commented Ashfaq Shuaib, MD, a professor of medicine and neurologist at the University of Alberta, Edmonton. Based on the WAKE-UP results, “MR would be preferred, but what we’ve been doing [since the WAKE-UP report] is if we see a CT scan that’s good we go ahead” with thrombolysis.

Dr. Jeffrey Saver

“Biologically, it doesn’t matter whether you use MR or CT; they both index the same underlying pathology. We’ve been hesitant to go beyond the MR finding from WAKE-UP, where there were data, but the findings from EXTEND were right in line with the WAKE-UP results, and that’s all we need to be reassured” that CT perfusion imaging also works for patient selection, commented Jeffrey L. Saver, MD, professor of medicine and director of the Comprehensive Stroke Center at the University of California, Los Angeles.

CT perfusion imaging and automated image processing “worked to select stroke patients” for an extended time window for treatment with mechanical thrombectomy in the DAWN (N Engl J Med. 2018 Jan 4;378[1]:11-21) and DEFUSE 3 (N Engl J Med. 2018 Feb 22;378[8]:308-18) trials, a history that makes the new finding of successfully using CT imaging to select patients who qualify for extended use of thrombolysis “a convincing result,” Dr. Saver said in an interview. The new EXTEND findings “will have a major impact” on using an extended time window for thrombolysis in U.S. practice, he predicted.

The EXTEND trial (Int J Stroke. 2012 Jan 1;7[1]:74-80) ran at 22 sites in Australia, 11 sites in Taiwan, and 1 center in New Zealand. Recruitment of patients into the study stopped early, after enrolling 225 patients, in June 2018, when results from WAKE-UP came out.

The EXTEND investigators enrolled patients who were either 4.5-9 hours out from the onset of their stroke or patients with a wake-up stroke with an uncertain onset. Participating centers could use either CT perfusion or MRI to identify candidates for treatment, and all used the RAPID software for image processing to identify patients with a perfusion lesion of at least 10 mL and an ischemic core volume no greater than 70 mL. Dr. Ma did not report what percentage of patients underwent imaging with each of these methods, but hinted that clinicians had used CT for a majority of the cases. The study randomized patients to receive either 0.9 mg/kg tPA or placebo, and by the trial protocol none of the enrolled patients received treatment with mechanical thrombectomy.

The trial’s primary endpoint was the percentage of patients with a modified Rankin Scale score of 0 or 1 at 90 days after their stroke, which was achieved by 44% more patients in the tPA group relative to the placebo arm after adjustment for age and baseline stroke severity, a statistically significant difference. The results were also positive for several secondary endpoints, such as recanalization 24 hours after treatment, which occurred in 67% of patients treated with tPA and 37% of the control patients, a statistically significant 68% relative improvement with thrombolysis.

Mortality at 90 days was similar in the two arms – 9% among the placebo patients and 12% among those who received tPA. The rate of symptomatic intracranial hemorrhage 36 hours after treatment was significantly higher among patients treated with tPA at 6%, compared with 1% in the placebo group, but the magnitude of this adverse effect was consistent with rates of intracranial hemorrhages previously reported in other studies of thrombolytic treatment for acute ischemic stroke, Dr. Ma said. The small number of increased intracranial hemorrhages “was not associated with increased mortality, and did not negate the positive result of an improved rate of excellent functional outcomes.”

These findings will likely spur further adoption of imaging processing software of the type used in EXTEND by U.S. stroke centers, Dr. Saver predicted.

“More and more centers have been getting this [software], and now they have two reasons to have it: to identify patients for an extended window for mechanical thrombectomy and to identify patients for an extended window for thrombolysis. It is a compelling case to have the imaging software as widely disseminated as possible. Centers that want to do the best for patients should have this imaging-processing software,” Dr. Saver said.

Dr. Ma and Dr. Shuaib reported no disclosures. Dr. Saver has received research funding and personal fees from Medtronic-Abbott and Neuravia.

SOURCE: Ma H et al. Int J. Stroke. 2018 Oct;13(2S):235, Abstract 1014.

Next Article: