IMS III: Endovascular Therapy Fails to Improve Stroke Recovery vs. t-PA Alone

Print  |  
Key Points:
  • Functional recovery no better with endovascular therapy than IV t-PA alone in acute stroke patients
  • Improved reperfusion with endovascular treatment does not translate to better outcomes
  • Interventional approaches for stroke treatment should move ahead at slower pace, outside expert says

By Jason Kahn
Thursday, February 07, 2013

Endovascular treatment—whether catheter-delivered thrombolysis or a stent retrieval device—does not improve outcomes for acute stroke patients when used as add-on therapy with tissue plasminogen activator (t-PA). Results of the Interventional Management of Stroke (IMS) III trial were presented February 7, 2013, at the International Stroke Conference in Honolulu, HI, and simultaneously published online ahead of print in the New England Journal of Medicine.

For IMS III, researchers led by Joseph P. Broderick, MD, of the University of Cincinnati (Cincinnati, OH), looked at patients with acute moderate-to-severe ischemic stroke who presented within 3 hours of symptom onset. Patients were randomly assigned to receive t-PA with or without endovascular therapy, which consisted of a thrombectomy device or catheter-based delivery of t-PA.

Trial Halted Early

A total enrollment of 900 patients was planned, but after an interim analysis of 656 patients (434 endovascular, 222 t-PA alone), the trial was halted when the results showed no substantive difference in benefit between the groups.

At 90 days, the primary outcome of functional independence, defined as a modified Rankin Scale score of 2 or less, was similar in the endovascular and t-PA groups (40.8% vs. 38.7%; absolute adjusted difference 1.5%; 95% CI -6.1 to 9.1%). Nor were there differences in the primary endpoint for subgroups with National Institutes of Health Stroke Scale (NIHSS) scores of 8 to 19, indicating moderately severe stroke (difference of 1.0%; 95% CI -10.8 to 8.8%; P = 0.83) or NIHSS scores of 20 or more, indicating severe stroke (difference of 6.8%; 95% CI -4.4 to 18.1%; P = 0.06).

After adjustment, patients treated within 2 hours of symptom onset showed a trend for better outcomes with endovascular therapy, but the difference was not significant. The same was true for patients treated within 90 minutes.

Overall, the proportion of patients with functional independence increased with greater reperfusion (P < 0.001), according to Thrombolysis in Cerebral Infarction (TICI) score:

  • 12.7% in patients with a TICI score of 0
  • 27.6% with a score of 1
  • 34.3% with a score of 2a
  • 47.9% with a score of 2b
  • 71.4% with a score of 3

Endovascular Treatment Improves Reperfusion to No Avail

Endovascular treatment provided an estimated increase of 40% in revascularization compared with t-PA alone, showing partial or complete recanalization rates at 24 hours in internal carotid artery occlusions, M1 and M2 occlusions of 81%, 86%, and 88%, respectively. These rates were 35%, 68%, and 77%, respectively, in patients treated with t-PA alone.

Yet despite the improved revascularization with endovascular therapy, this did not translate to any clinical benefit.

In terms of individual endovascular devices, the highest rate of reperfusion (TICI score of 2-3 in an occlusion of the internal carotid artery, M1, or both) was achieved with the Penumbra System (85%; 39 patients; Penumbra, Alameda, CA) followed by:

  • 75% for the Solitaire FR device (4 patients; Covidien, Irvine, CA)
  • 73% for the Merci Retriever (77 patients; Stryker, Kalamazoo, MI)
  • 71% for intra-arterial t-PA via standard catheter (51 patients)
  • 71% for the MicroSonic SV infusion system with intra-arterial t-PA (14 patients; EKOS, South Bothell, WA)

Rates of mortality at 7 and 90 days were equivalent between groups, as were rates of symptomatic intracerebral hemorrhage and parenchymal hematoma. Patients receiving endovascular therapy did experience higher rates of asymptomatic intracerebral hemorrhage (27.4% vs. 18.9%; P = 0.01) and subarachnoid hemorrhage (11.5% vs. 5.8%; P = 0.02). There was a 16% rate of device or procedural complication specific to the endovascular group.

“The trial showed similar safety outcomes and no significant difference in functional independence with endovascular therapy after intravenous t-PA, as compared with intravenous t-PA alone,” the researchers conclude.

Even the ‘Best Data’ Leave Questions

The authors note some key limitations of the study. First, the time to endovascular treatment in IMS III was 32 minutes longer than in the smaller IMS I study that preceded it. “This may be one important reason for the lack of clinical benefit, despite the finding of substantially better revascularization with endovascular therapy than with [IV] t-PA,” they write. And second, the modern class of stent retrievers was used in only a small number of patients in IMS III.

Nevertheless, after SYNTHESIS Expanded, the results represent the second randomized trial in as many days to show no benefit of endovascular therapy compared with IV thrombolysis alone in acute stroke patients.

According to Philip M. Meyers, MD, of Columbia University Medical Center (New York, NY), the growing body of data will hopefully convince proponents of endovascular therapy to move ahead at a more deliberate pace. “There are a lot of proponents of interventional stroke therapy who have a strong sense that they already know what they need to know and we just need to do it,” he told TCTMD in a telephone interview. “But you take a look at the best data we’ve got, and it leaves questions.”

A key lesson from IMS III, he noted, is that better reperfusion does not equate to better stroke outcomes.

“Just opening up the vessel does not mean the patient is going to get better,” Dr. Meyers said. “That’s a very different message from what cardiology has found, that if you open up the vessel there’s a pretty good likelihood the heart muscle is going to respond. For the brain, it’s just not as apt to do that.”

Overall, he expressed hope that the data will “galvanize the whole community” regarding stroke research. Agreeing with the study authors, Dr. Meyers concluded that “there may be an important future for stroke intervention, but it’s not clear yet,” and that the only way forward is to continue enrolling patients in clinical trials “so we can figure out who are we supposed to be treating, because [endovascular therapy] is not a benign thing—it’s expensive, and it can definitely create harm.”


Broderick JP, Palesch YY, Demchuk AM, et al. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke. N Engl J Med. 2013;Epub ahead of print.



  • The study was supported by grants from the National Institutes of Health (NIH) and the National Institute of Neurological Disorders and Stroke and by Boehringer Ingelheim, Concentric Medical, Cordis Neurovascular, EKOS, and Genentech.
  • Dr. Broderick reports receiving consulting fees from Photo-Thera, receiving lecture fees from Oakstone Publishing, and receiving a drug from Schering-Plough for use in a study funded by the NIH.
  • Dr. Meyers reports serving as an external monitor for the IMS III trial.


Related Stories:

Click here for a listing of companies that provide support to the Cardiovascular Research Foundation, owner and operator of TCTMD.

Related Content: