Skipping tPA Before Stroke Thrombectomy May Be Okay: DIRECT-MT

Among patients presenting directly to a thrombectomy-equipped center with a large-vessel ischemic stroke, going directly to the endovascular suite and forgoing initial IV thrombolytic therapy provides noninferior outcomes compared with the guideline-recommended approach of giving alteplase to eligible patients before the procedure, the DIRECT-MT trial shows.

There was no significant difference in the distribution of modified Rankin Scale (mRS) scores at 90 days between the two groups (adjusted common odds ratio 1.07; 95% CI 0.81-1.40), with noninferiority established because the lower end of the confidence interval was at least 0.80, according to researchers led by Pengfei Yang, MD (Naval Medical University Changhai Hospital, Shanghai, China).

The trial, published online May 6, 2020, ahead of print in the New England Journal of Medicine, “is the first major study that is conclusively able to show that, for patients with acute ischemic stroke due to an anterior circulation large-vessel occlusion, endovascular thrombectomy alone is essentially equivalent to endovascular thrombectomy preceded by intravenous alteplase, because there were comparable clinical outcomes between two groups,” Yang told TCTMD in an email.

He predicted that the results will have a major impact on stroke care around the world. “Patients will no longer have to wait to receive intravenous thrombolysis. We have the potential to further improve efficiency of stroke care, [because] as you know in acute stroke: time is brain. The faster we can treat, the better it is for the patient,” Yang said. “Meanwhile, we can now avoid costs and complications associated with alteplase, such as hemorrhages and allergy. We think it will change practice and also the guidelines certainly when the results of our trial are confirmed by the other ongoing trials.”

But Gregory Albers, MD (Stanford Stroke Center, Palo Alto, CA), who wrote an accompanying editorial, is not so sure. Speaking with TCTMD, he raised a couple of issues to consider when interpreting the results. First, the trial included only patients who presented directly to a comprehensive stroke center with thrombectomy capabilities. The more common scenario in practice is for patients to initially go to a local hospital, have IV thrombolytic therapy started, and then get transferred to a thrombectomy center, allowing the alteplase time to work. Second, the noninferiority margin was “generous” and the difference in the primary endpoint could not exclude the possibility of a 20% advantage for administering tPA before thrombectomy, he said.

“We need more data. This one study alone is not enough to change guidelines or change the procedure,” Albers commented, noting that the SKIP trial, presented earlier this year at the International Stroke Conference, was unable to show that going directly to thrombectomy was noninferior to the standard approach. The community needs to see the results of at least three major ongoing trials exploring this issue—SWIFT DIRECT, DIRECT-SAFE, and MR CLEAN-NO IV—before making any changes to practice, he said.

DIRECT-MT

The randomized trials showing that mechanical thrombectomy improves functional outcomes in patients with acute ischemic strokes caused by large-vessel occlusions all provided initial IV thrombolytic therapy in patients who were eligible. There is, however, some uncertainty about the relative risks and benefits of using IV tPA before the procedure.

“Alteplase may increase early reperfusion of the ischemic area and dissolve residual distal thrombi after endovascular thrombectomy,” Yang et al note in their paper. “For large proximally located thrombi, however, the lytic effect of intravenous alteplase is limited, and partial lysis could fragment the target thrombus or cause it to migrate distally, potentially complicating endovascular thrombectomy.” Moreover, alteplase comes with a financial cost and may raise the risk of intracranial hemorrhage.

Some, but not all, observational data suggest that performing thrombectomy without initial IV thrombolytic therapy may provide outcomes comparable to those obtained using the guideline-recommended approach, but thus far there have been few randomized trials to inform the issue.

DIRECT-MT, conducted at 41 tertiary care centers in China, included 656 patients (median age 69; 56.4% men) who had acute ischemic strokes caused by large-vessel occlusions in the anterior circulation and who presented within 4.5 hours of symptom onset. They were randomized to receive IV alteplase administered at a dose of 0.9 mg/kg before thrombectomy or to undergo thrombectomy without initial thrombolytic therapy. Clot removal was done using stent retrievers as the primary modality, with aspiration devices allowed as secondary options if initial reperfusion failed.

Median time from stroke onset to randomization was 167 minutes in the direct thrombectomy arm and 177 minutes in the tPA arm, with median times from randomization to groin puncture of 31 and 36 minutes, respectively.

Of the patients who received alteplase, only a minority completed the infusion before the thrombectomy procedure began: 86.5% of patients had the infusion completed during the procedure.

The trial established that going directly to thrombectomy was noninferior to IV thrombolytics followed by thrombectomy, although because the lower boundary of the 95% confidence interval was just above the noninferiority threshold, “the results do not rule out a benefit of alteplase,” the researchers say.

For secondary outcomes, the patients who received IV alteplase before thrombectomy had numerically higher rates of successful reperfusion both before thrombectomy (7.0% vs 2.4%) and on final angiography after the procedure (84.5% vs 79.4%). The 90-day mortality rate did not differ between groups (18.8% vs 17.7%).

‘Puzzling’ Results

Albers said the results were puzzling because the potential disadvantages of alteplase, including delays in getting to endovascular therapy and increased risk of intracranial hemorrhage, did not occur and the potential advantage—namely, improved reperfusion—did. In that scenario, he said, one would expect to see better functional outcomes in the tPA arm, but that was not observed.

The reason for the disconnect is unclear, he said. It could be the thrombolytics did not have a lot of time to work on the clot before the thrombectomy procedure began, he said, pointing out that it’s typical in everyday practice for patients to have a complete infusion of tPA before undergoing the procedure.

Considering these issues, DIRECT-MT will not, on its own, have an impact on practice, Albers said, adding that it could have more of an effect if the results are replicated in the ongoing trials. But any impact would be confined to patients presenting directly to thrombectomy centers because one would still want to initiate IV thrombolytic therapy in those who need to be transferred, he said. The tPA will dissolve the clot before thrombectomy in about 10% of cases, he noted, “so you wouldn’t want to give up on that chance.” Albers pointed out, too, that another thrombolytic, tenecteplase (TNKase; Genentech), has been shown to double that proportion to about 20%.

Asked why functional outcomes were not better in the tPA arm, Yang said that the reperfusion rates, though numerically higher with thrombolytic therapy, were not significantly different due to the lack of adjustment for multiple comparisons in the study. Moreover, he said, there were numerically but not significantly more symptomatic intracranial hemorrhages in the alteplase arm (6.1% vs 4.3%), which may partially explain the lack of an advantage for that approach.

Yang agreed with Albers that the trial results apply primarily to patients who present directly to thrombectomy centers. For this group, “go directly to the angio suite for thrombectomy,” Yang said. However, for the so-called “drip-and-ship” patients who require a transfer, he said, “we still suggest using current standard treatment.”