Acute Stroke Therapy Beyond IV tPA: Clot Removal—Bust It Up or Pull It Out

Abstracts and Commentary

Synopsis: This phase 1 study shows that cerebral embolectomy with the Merci Retriever was safe and that successful recanalization could benefit a significant number of patients, even when performed in an extended 8-hour time window.

Sources: Gobin YP, et al. MERCI I: A Phase I Study of Mechanical Embolus Removal in Cerebral Ischemia. Stroke. 2004;35:2848; Alexandrov AV, et al. For the CLOTBUST Investigators. Ultrasound Enhanced Systemic Thrombolysis For Acute Ischemic Stroke. N Engl J Med. 2004;351:2170.

The treatment of acute ischemic stroke is in its most simple terms a matter of plumbing. There is an artery occluded by clot. Blood flow must be restored as quickly as possible to provide the best possible chance of salvaging ischemic brain, and optimizing clinical outcome. While intravenous or intra-arterial thrombolysis may achieve this goal, recanalization rates are incomplete, and there may be the complication of intracerebral hemorrhage. These 2 contemporaneous reports, while very different in methodology, both open the horizon of a different, perhaps more effective, and likely safer method of stroke treatment-mechanical clot disruption or removal.

In the CLOTBUST study, Alexandrov et al applied transcranial doppler (TCD) to patients under treatment with intravenous tPA for acute ischemic stroke. It is believed that ultrasound may enhance thrombolysis by improving drug transport, altering fibrin structure, and improving the binding of tPA to fibrin. The randomized CLOTBUST study was based on compelling pilot data, and is a bridge towards a larger Phase III efficacy study.

Target patients were exposed to continuous therapeutic TCD examination for 2 hours, while control patients had TCD probes applied with only brief diagnostic examinations at 30, 60, 90, and 120 minutes. Patients were eligible for the study if they had abnormal flow in the middle cerebral artery confirmed on initial screening TCD. The predetermined end point of the study was complete arterial recanalization, or dramatic clinical recovery, within 2 hours. This was reached in 31/63 patients in the target group (49%), compared with 19/63 patients in the control group (30%), P = 0.03. When purely clinical end points were used, at 2 hours, 24 hours, and 3 months, there was a non-significant trend towards benefit in the target group. Larger studies will be needed to confirm whether these trends are meaningful. There was no difference in rates of bleeding between the 2 groups.

In the MERCI trial, Gobin et al applied a newly developed endovascular device that is capable of snaring blood clots and removing them from intracranial vessels. The device is a coiled wire composed of nitinol (nickel titanium), a material capable of assuming a linear shape while inside a catheter and then a coiled shape upon deployment into a clot.

The MERCI Phase I trial included 28 patients, treated up to 8 hours after stroke onset (mean time 6 hours, 15 minutes). The study required an initial NIH Stroke Scale (NIH-SS) of 10, and for enrolled patients, the median initial NIH-SS was 22. Successful recanalization with embolectomy alone was achieved in 12 (43%) patients and with the addition of intra-arterial tPA in 18 (64%) patients. There was 1 technical complication, in which the tip of the retriever device became detached in the vessel. Another MERCI retriever was used to capture this, without any clinical consequences. There were no symptomatic hemorrhages. At 1 month, there were 9/18 revascularized patients, and none of the 10 non-revascularized patients had achieved significant recovery. Recovery was defined as a modified Rankin score of 0-2 for patients with initial NIH-SS of 10-20 and 0-3 for patients with an initial NIH-SS of greater than 20.

The MERCI I study is to be followed shortly by the MERCI II data, extending these results to a much larger number of patients. Unfortunately, none of the MERCI data is in any way blinded, randomized or placebo controlled, making it difficult to draw definitive conclusions about its efficacy.


The CLOTBUST and MERCI studies share much in terms of their strengths and also in their weaknesses. Unlike IV tPA, which is a drug that can be given quite indiscriminately to stroke patients, only limited by a 3 hour time window and fairly simple clinical and radiographic selection criteria, these technologies raise complex issues in terms of patient selection, operator skill, and overall generalizability. Both CLOTBUST and MERCI apply primarily to large strokes affecting the middle cerebral artery (MCA--M1 or M2 trunks). Patients who have these lesions must therefore be accurately and quickly identified. A vascular diagnosis can be imputed on clinical grounds, but ideally requires confirmation by tests such as CT or MR angiography. For patients with distal MCA disease or penetrator artery lesions, CLOTBUST and MERCI are of limited or no use.

Perhaps more importantly, both of these technologies are extremely user dependent. CLOTBUST requires that TCD be performed by an experienced sonographer who can accurately insonate the MCA, and reliably analyze its waveforms. The TCD technology itself is inexpensive and portable, but there is a significant human element than cannot be underestimated. For MERCI, there must exist a tertiary care center with immediate availability of an endovascular suite and team of specialists who can effectively navigate the embolectomy device into intracranial clots. Clearly, the safety of the MERCI device and the reproducibility of its efficacy will not be easily replicated by novice users.

Despite the many unanswered questions, CLOTBUST and MERCI represent important advances in our ability to treat acute ischemic stroke. Interestingly, although future studies are pending, both TCD (sold commercially by many manufacturers) and the MERCI retriever (recently FDA approved) are at present available for use should any clinician choose to employ them. Alan Z. Segal

Dr. Segal, Assistant Professor, Department of Neurology, Weill-Cornell Medical College, Attending Neurologist, New York Presbyterian Hospital, is Assistant Editor of Neurology Alert.