In a new randomized trial that investigated the question of whether thrombolysis can be omitted for patients with stroke who are undergoing endovascular thrombectomy for a large-vessel occlusion, results were similar for both approaches.
and functional outcomes were not significantly different. In addition, hemorrhage rates with or without intravenous alteplase administration before endovascular treatment were similar.
“From the MR CLEAN NO IV results, we cannot change standard practice, as we failed to show superiority of the direct endovascular approach, and we also didn’t meet the noninferiority criteria. So, the standard practice of giving tPA to those eligible still holds,” said co–lead investigator Yvo Roos, MD.
“But I think we can say that these results suggest that there may also not be such a need for tPA in patients who can go straight for endovascular therapy,” said Dr. Roos, who is professor of neurology at Amsterdam Medical Center.
“If we are not sure whether a patient is suitable for tPA because they have a higher bleeding risk, I think we can be reassured about missing the tPA out and going straight to endovascular treatment. So, if in doubt, leave it out,” he added.
Results of the MR CLEAN NO IV trial were presented at the International Stroke Conference sponsored by the American Heart Association.
“If in doubt, leave it out”
Dr. Roos noted that three trials have investigated the question regarding dropping thrombolysis for patients who can receive thrombectomy quickly. These are the DIRECT MT, SKIP, and DEVT studies. All of these trials were conducted in Asian countries, and none found differences in functional outcomes between the two approaches.
The largest of these studies – the DIRECT-MT trial, from China, which was a sister study to MR CLEAN NO IV – did show noninferiority of the direct endovascular approach to tPA plus endovascular treatment.
But because of differences in health care logistics and trial populations, the benefits and risks of dropping thrombolysis in Western countries are not known, explained Charles Majoie, MD, who is co–lead investigator of the current trial and is chair of neuroradiology at Amsterdam Medical Center.
The MR CLEAN NO IV trial was designed to show superiority of the direct endovascular approach with noninferiority for hemorrhage. It enrolled 540 European patients who were eligible for both thrombolysis and thrombectomy and who presented to a thrombectomy-capable center. They were randomly assigned to receive thrombolysis plus endovascular therapy or direct endovascular therapy alone.
The mean time from stroke onset to groin puncture (the start of endovascular therapy) was very fast in both groups – 130 minutes in the direct group, and 135 minutes in the tPA group.
The primary outcome was a shift analysis of the Modified Rankin Scale (mRS). On that outcome, the trial failed to show significant superiority of the direct approach (odds ratio, 0.88; 95% confidence interval, 0.65-1.19).
A good functional outcome (mRS, 0-2) was achieved in 49% of the direct thrombectomy group and in 51% of the tPA group (OR, 0.95; 95% CI, 0.65-1.40).
Safety results showed no difference in any of the hemorrhage endpoints between the two groups. The rate of symptomatic intracranial hemorrhage was actually numerically higher in the direct thrombectomy group (5.9% vs. 5.3%).
“One of the most intriguing results of this study is that there was no increase in hemorrhage in the tPA group,” Dr. Roos commented. “This is very surprising, as we have always thought thrombolysis causes an increased bleeding risk. But after these results, we may have to rethink that idea – perhaps it is not the tPA itself that causes bleeding risk but rather the opening up of the vessel.”
On the failure to show noninferiority of the direct approach, Dr. Roos suggested that the trial may have been underpowered in this respect.
“Our sister trial, DIRECT-MT, was a noninferiority study. They had 650 patients, and they just reached noninferiority,” he said. “In MR CLEAN NO IV, we were aiming for superiority, and we had fewer patients – 540. We didn’t show superiority, and we didn’t have quite enough patients to show noninferiority.”
He added that, considering all the four studies together, the results look very similar and suggest no difference between the two approaches.