The use of an advanced imaging shortly after the onset of acute stroke failed to identify a subgroup of patients who could benefit from a clot-removal procedure, a study has found.

The randomized controlled trial known as Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy (MR RESCUE) was funded by the National Institute of Neurological Disorder and Stroke (NINDS), part of the National Institutes of Health.  It was published online Feb. 8 in the New England Journal of Medicine and simultaneously presented at the International Stroke Conference. 

In patients with ischemic stroke, brain cells deprived of blood die within minutes to hours. Rapidly opening the artery can halt brain cell death.  Intravenous tissue plasminogen activator (t-PA) has been shown to improve outcomes in such stroke patients.  However intravenous t-PA is not effective in many patients with large clots blocking the major brain arteries that cause the most devastating strokes.  MR RESCUE investigators tested an invasive embolectomy strategy designed to remove clots from these large arteries.  Patients in the study were enrolled at 22 centers in the United States within approximately 5.5 hours of their stroke onset. Their ability to function independently was assessed at 90 days.

All MR-RESCUE patients underwent emergency computed tomography (CT) or magnetic resonance (MRI) perfusion imaging to identify regions of the brain with decreased blood flow, as well as regions that could not be salvaged.

The investigators hypothesized that patients in whom the scan suggested that less than 70 percent of the brain with decreased blood flow had already died would benefit from the clot-removal procedures.  Based on the imaging results, the 118 study patients were randomly assigned to receive a clot-removal procedure within eight hours of symptom onset (64 patients) or standard therapy (54 patients) according to medical protocols. The procedure involves threading a special catheter through an artery in the groin up to the site of a clot in a brain artery, then removing the blockage.  Tools used to remove the clot included the MERCI Retriever (a tiny corkscrew-like device) or the Penumbra System (an aspiration device).

"Despite a lack of evidence showing that these embolectomy devices improve outcomes, they are already widely used in patients who are not able to get to the hospital in time to receive t-PA," said Walter J. Koroshetz, M.D., deputy director of NINDS. "Though some patients have had dramatic improvements with clot removal, it has not been shown effective in this or another larger study, the Interventional Management of Stroke (IMS III), which was halted early because it did not find the procedure to be of significant benefit."

"The majority of patients were not eligible to receive intravenous t-PA because they arrived at the hospital too late for t-PA to be effective," said one of the primary investigators, Chelsea Kidwell, M.D., of Georgetown University Medical Center in Washington, D.C.

The patients with the presumed favorable imaging findings had the same level of disability at 90 days whether or not they had undergone the embolectomy procedure or had received standard therapy. More specifically:

• The level of disability 90 days after stroke was no different in patients who received medical care or clot removal.
• The rates of death (21 percent) and symptomatic brain bleeds (4 percent) were no different in the two treatment groups.
• Response to embolectomy was no better in the 58 percent of patients with a favorable imaging pattern that suggested salvageable brain tissue.

The researchers hypothesized that the lack of a treatment effect may reflect the fact that these patients had enough blood flow to the brain from secondary sources to support the brain tissue until spontaneous reperfusion occurred. Patients without the favorable imaging findings did not benefit from embolectomy.
However, the results of MR RESCUE are not consistent with the conclusions of a separate NINDS-funded observational study called DEFUSE-2 that suggested that a slightly different brain imaging strategy could predict patients who benefited from the clot-removal procedure.

In addition to other imaging techniques "it's possible that newer intra-arterial devices that were not available when the study started could improve functional outcomes," said Scott Janis, Ph.D., program director, NINDS. "But an important message from MR RESCUE is that those newer devices still need to go head to head with standard therapy."

"Advances in neuroimaging are promising and may someday help to identify who will benefit from a device-based approach. But the MR RESCUE results show that more work needs to be done," Dr. Koroshetz said.

Reza Jahan, M.D., associate professor of interventional neuroradiology at the University of California, Los Angeles, is co-principal investigator of the study. Other authors include: Jeffrey Gornbein, Dr.P.H.; Jeffry R. Alger, Ph.D.; Val Nenov, Ph.D.; Zahra Ajani, M.D.; Lei Feng, M.D., Ph.D.; Brett C. Meyer, M.D.; Scott Olson, M.D.; Lee H. Schwamm, M.D.; Albert J. Yoo, M.D.; Randolph S. Marshall, M.D., M.S.; Philip M. Meyers, M.D.; Dileep R. Yavagal, M.D.; Max Wintermark, M.D.; Judy Guzy, R.N.; Sidney Starkman, M.D.; and Jeffrey L. Saver, M.D. Disclosures from all authors are reported in the full text of the NEJM article and on the abstract presented at the International Stroke Conference.

Funding for MR RESCUE was provided by an NINDS grant (P50 NS044378). Concentric Medical, Inc. (Mountain View, Calif.) provided study catheters and devices from study start until August, 2007; thereafter, costs for all study catheters and devices have been covered by study funds or third party payers.