麻豆传媒

PHOENIX -- Sooner is better for intravenous thrombolysis (IVT) given before mechanical thrombectomy for acute ischemic stroke, a meta-analysis of patient-level data from randomized controlled trials showed.

While the overall showed noninferiority of thrombectomy alone to combination with thrombolytics for functional outcomes, timing of thrombolysis had a significant impact, such that each hour of delay gave the combination a 16% greater advantage (ratio of adjusted common OR 0.84, 95% CI 0.72-0.97, P=0.02 for interaction), reported Johannes Kaesmacher, MD, PhD, of University Hospital Bern in Switzerland, at the American Stroke Association's International Stroke Conference (ISC).

The study findings were also published simultaneously in .

Clot-busting drugs given within 1 hour of stroke onset had a significant 49% greater odds of a one-step-better modified Rankin Scale (mRS) score compared with thrombectomy alone. That advantage diminished to 25% if given within 2 hours, but remained significant. The advantage was gone at around 3 hours.

The absolute difference in likelihood of surviving with independence in daily living (mRS ≤2) was 9% if the thrombolytic was given in the first hour, 5% if given within 2 hours, and 1% at 3 hours.

Thus, the effect size is clinically important with a "reasonable" number needed to treat to shift the outcome for one patient, Kaesmacher said.

"The time intervals are important," he said at the session. "The decision to administer IVT before thrombectomy should take into account the time-dependent treatment effect, rather than the overall effect."

While the findings are the first good proof that giving IV tissue plasminogen activator (tPA) thrombolysis is better for patients who will go on to get mechanical thrombectomy, they are largely confirmatory for clinical practice, noted ISC session moderator Nils Petersen, MD, PhD, MSc, of Yale New Haven Hospital in New Haven, Connecticut.

"IV tPA is still the standard of care out to 4.5 hours," he told 麻豆传媒. Where the findings can be informative, he suggested, is when you're in that later time window and "on the fence" for a patient with some concern for risk with thrombolysis, such as trauma without life-threatening bleeding.

"I think the prior studies have been supportive that maybe you're not harming the patient by withholding tPA in a situation where you're unsure whether you should give it, because there may be some relative contraindications," he said. "Now, this analysis, I think, shifts that a little bit by showing that if you have a patient that's early in the window, there's an actual benefit of the combination therapy. ... And specifically in these patients, you shouldn't forego IV tPA."

While intuitive that faster is better in stroke care, Kaesmacher noted that quicker recanalization might not have been the mechanism for benefit.

Early recanalization before thrombectomy occurred in 2% to 5% of patients in the overall trial, he said at the session. "The interesting thing is, that was significant actually, the rate of early recanalization is more frequent if you give IVT later. So it's in the opposite direction. So certainly this cannot be the effect explaining the differences in outcome."

Other neuroprotective effects with early thrombolysis are possible, Petersen noted. "Maybe if you have the tPA onboard and then you achieve mechanical recanalization, then it helps with clearance of distal microthrombi or distal small vessel perfusion by clearing up maybe some residual thromboses."

Kaesmacher also emphasized that the findings apply only to patients directly admitted to a thrombectomy-capable stroke center, which was the setting for all the included clinical trials.

The meta-analysis pooled data on the 2,313 participants with an anterior circulation large-vessel occlusion in the DEVT, DIRECT-MT, DIRECT-SAFE, MR CLEAN-NO IV, SKIP, and SWIFT DIRECT trials, all of which randomized participants to IVT plus thrombectomy versus thrombectomy alone. Basilar artery occlusions were excluded.

How to estimate the time to IVT for patients who haven't received it is "a problem which has been dealt with in other trials as well," Kaesmacher explained. The researchers used the stroke onset-to-trial randomization time plus the timing of randomization to IVT administration in each trial to yield the time from onset to expected IVT for the non-IVT-treated patients.

Onset to expected IVT times were well-balanced and showed a large overlap across all the trials. "So you're really analyzing a cohort effect rather than a trial effect," he added.

The findings were also "really stable statistically," he said, with all the sensitivity analyses, including an analysis using just onset-to-randomization time showing a significant result.

Comment