Subscribe to eTOC

No Improvement Found in Stroke Outcomes with Endovascular Treatment


Two major studies found that endovascular therapy did not improve functional outcomes for stroke compared with tissue plasminogen activator. But experts said more studies are needed with second-generation devices.

Eight years after the Food and Drug Administration (FDA) cleared the first endovascular clot retrieval device for acute stroke, several large studies that compared endovascular techniques with standard of care — intravenous tissue plasminogen activator (tPA) — have arrived at a surprising finding: endovascular treatment doesn't enhance the functional outcome for patients in the throes of a stroke. At least the first generation clot retrieval devices don't, according to two teams of investigators whose results appeared in the Feb. 7 and Feb. 8 online editions of The New England Journal of Medicine (NEJM).

One team of researchers — the MR RESCUE Investigators — hoped to determine if penumbral imaging would identify patients who would differentially benefit from endovascular therapy for acute ischemic stroke within eight hours after symptom onset; while the second team — the Interventional Management of Stroke (IMS) III Investigators — focused on whether combining endovascular therapy with tPA was better than tPA alone for patients with moderate-to-severe acute ischemic stroke.

No one involved in testing endovascular techniques for stroke is willing to strike a death knell for these devices, as there are many reasons that the studies did not show an added benefit. It could have been the limitations of the device itself, they said, noting that there are second-generation devices that are better at opening the vessel.

At this point, said Chelsea S. Kidwell, MD, professor of neurology and director of the Georgetown University Stroke Center, who led the MR RESCUE study, “we don't have evidence that the first-generation devices work, but I can't make recommendations for routine practice. It is hard to extrapolate to the latest retrieval devices.”

The FDA's device division cleared the MERCI clot retriever for use in stroke in 2004. That same year, Dr. Kidwell began a study to test outcomes in patients who received tPA versus those who underwent tPA and mechanical embolectomy.

In 2006, the IMS III team also began a large study comparing tPA versus endovascular therapy following tPA. Endovascular therapy included treatment with intra-arterial administration of tPA and the FDA-cleared thrombectomy devices — the Merci Retriever, the Penumbra aspiration system, and the Solitaire Retriever. However, only a handful of the stent retrievers had been used when the study was halted.


Joseph P. Broderick, MD, the Albert Barnes Voorheis chair of neurology and rehabilitation medicine at the University of Cincinnati and research director of the university's Neuroscience Institute, led the IMS III group, an international study of 656 patients (at 58 centers) that was halted midstream after the safety monitoring board could not find a signal that opening the arteries (either mechanically or medically) to restore blood flow was helping.

Patients who received both tPA and endovascular interventions showed no more improvement in function at 90 days than those randomized to tPA alone: 40.8 percent who had tPA plus endovascular therapy had a modified Rankin score of two or less compared with 38.7 percent of those who just received intravenous t-PA. Mortality was similar at 90-days: 19.1 percent for those who had both treatments versus 21.6 percent for tPA. There were similar safety outcomes. There was some signal of potential benefit for endovascular therapy in subjects with severe strokes (based on the NIH Stroke Scale score ≥ 20), treated earlier, and with clots in the distal part of the internal carotid arteries, however, these differences were not statistically significant in the pre-defined analyses.

“The findings mean that we have a lot more work to do to understand who benefits most from endovascular approaches. So many stroke physicians and interventionalists were convinced that endovascular therapy is beneficial,” said Dr. Broderick. “Based on these studies, the endovascular techniques should be ideally used in the context of trials rather than as the standard of care. There is no evidence that we are denying patients a proven benefit. We just don't have the data yet to say how we should be using these tools.”


Dr. Kidwell was interested in testing the benefits of a multi-modal MRI or CT scan to see whether identifying occlusions would improve outcomes in those patients with evidence of tissue at risk for damage on the heels of stroke. They enrolled and analyzed the results from 118 patients who arrived to the emergency room within eight hours of the initial ischemic event; those who arrived to the emergency room after the four-hour window were no longer candidates for tPA.

Patients underwent pretreatment CT or MRI scans and then were stratified into groups, depending on the amount of salvageable tissue. It was thought that endovascular techniques might be more useful in preventing further tissue damage in and around the penumbra.

They developed a model to determine whether there was a penumbra pattern — areas of brain that had abnormal perfusion but were not abnormal on diffusion-weighted MRI. The patients were stratified into four groups: those with and without a penumbra pattern, and those treated with a retrieval device versus those who were treated with tPA. (Most of the patients in this study did not receive tPA because they arrived at the hospital too late.)

As was seen in the IMS III study, there were no differences between the two treatment arms. The analysis also showed no difference in the penumbra pattern observed on imaging, said Dr. Kidwell. Imaging and targeted interventions did not result in better outcomes for patients.

Dr. Kidwell said that the rate of substantial revascularization in the embolectomy group was rather low, 67 percent, and that could have a lot to do with the first-generation technology. (The newer devices have better revascularization rates.) Still, time could also be a contributing factor to the neutral results of these studies, said Dr. Kidwell. When they separated out the groups that received treatment within three hours and after three hours, they observed that recanalization “was particularly beneficial in patients with large-vessel occlusions and poor collateral vessels,” they wrote in their NEJM paper.

The results from both studies, as well as a third study called SYNTHESIS that was led by a research team in Italy, were presented at the American Stroke Association/American Heart Association's International Stroke Conference in Honolulu. The SYNTHESIS trial randomized patients to either tPA or endovascular therapy and found no evidence of benefit in those treated with intra-arterial strategies.


Neurologists agree that more studies are needed to address the use of endovascular therapy. In an editorial in the same issue of the NEJM, Marc I. Chimowitz, MBChB, of the Medical University of South Carolina, said that “recruitment in new trials will still be challenging, particularly among patients with large disabling strokes and their concerned families who 'want everything done,' especially with new endovascular devices available and third-party payers willing to reimburse for these procedures.”

He said that should Medicare call for a moratorium on reimbursement for endovascular treatment “it would facilitate recruitment in these urgently needed trials.”

Walter Koroshetz, MD, deputy director of the National Institute for Neurological Disorders and Stroke, said that there have been many anecdotal reports of dramatic improvement with endovascular techniques following acute ischemic stroke, and certainly it may be useful in otherwise fatal or severely disabling cases where tPA doesn't dissolve the embolus. Blocked carotid or basilar arteries are life-threatening. Such cases initially ignited the idea for endovascular techniques for stroke. But over time, Dr. Koroshetz said, “we have generalized from what we know works for those who have no other chance at quality of life to other stroke patients in which the harm balances the benefit. We shifted from compassionate use to widespread use without the data.”

The studies do offer hints on how to proceed. “We have to be much faster and safer in the delivery of endovascular techniques,” said Dr. Koroshetz. Sixteen percent of patients in the IMS III had a procedural complication. “We have to learn how to protect the brain while these unstable patients are instrumented, sedated, and vulnerable in the post procedure period.”

The promise of the newer devices is that they are faster at opening arteries, Dr. Koroshetz added.

“The findings were a big shock,” added Dr. Koroshetz, who noted that the NINDS funded the studies. “What people thought would be a major success was a major disappointment.” IMS III cost $26 million, even though the study ended with only 650 of the 900 patients enrolled. Without this trial, the procedure would have continued without the data that are necessary to advance the science of reperfusion.

“These studies raise a lot of questions,” added Anthony J. Furlan, MD, a professor and chairman of neurology at Case Western Reserve School of Medicine. “This is not the death of endovascular therapy. We are still early in the field. But I think it tells us that the number of patients who benefit from endovascular approaches is smaller than we thought. We have to learn our lessons. We can't lump all these strokes together, or all of the devices. The key issue is homogeneity in our selection for these studies.”

Lawrence Wechsler, MD, chairman of the department of neurology at the University of Pittsburgh School of Medicine and director of the Stroke Institute, called the results of the trials “sobering. We need to do better at finding the right patients and improving the endovascular therapies.”

“We have felt all along that this approach required testing and evidence to figure out when and where it should be used. We hope that more patients will be entered into new trials so that we can make sense of what we are doing.”



Joseph P. Broderick, MD, the Albert Barnes Voorheis chair of neurology and rehabilitation medicine at the University of Cincinnati and research director of the university's Neuroscience Institute, led the IMS III group, an international study of 656 patients (at 58 centers) that was halted midstream after the safety monitoring board could not find a signal that opening the arteries (either mechanically or medically) to restore blood flow was helping. Here Dr. Broderick highlights the findings from this and other trials on endovascular therapy for stroke — and what they mean for clinicians:


• Kidwell CS, Jahan R, Saver JL, et al for the MR RESCUE Investigators. N Engl J Med 2013: E-pub 2013; Feb. 8.
    • Broderick JP, Palesch YY, Tomsick TA, et al, for the Interventional Management of Stroke (IMS) III Investigators. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med 2013; E-pub 2013 Feb 7.
      • Chimowitz MI. Endovascular treatment for acute ischemic stroke - still unproven. N Engl J Med 2013; E-pub 2013 Feb 8.
        Neurology Today archive on endovascular therapy: