Rigid Criteria May Exclude Patients From IV-TPA

July 2010

By Amy Rothman Schonfeld
Elsevier Global Medical News

TORONTO - By following new evidenced-based SMART criteria that reflect current knowledge of stroke management, almost 90% of ischemic stroke patients who met at least one or more common exclusion criteria for IV recombinant tissue plasminogen activator treatment were able to undergo successful thrombolysis, according to findings presented at the annual meeting of the American Academy of Neurology.

"Despite the results of randomized trials demonstrating the safety and efficacy of IV r-TPA [recombinant tissue plasminogen activator] for acute ischemic stroke, its use remains exceedingly low, about 1%-3%," said Dr. Jack Rose, associate director of Neurocritical Care at the California Pacific Medical Center in San Francisco, in his presentation.

He noted that at his center, by following the Simplified Management of Acute Stroke Using Revised Treatment, or SMART criteria, about 25% of all ischemic stroke patients now seen there receive some form of TPA.

Dr. Rose presented the results from 195 patients with ischemic stroke who were seen between July 2006 and December 2009 and treated with TPA. Of those, about 70% (140) received IV r-TPA alone. Favorable outcomes (modified Rankin Scale scores of less than or equal to 1) were achieved for 58% of the patients. The symptomatic intracerebral hemorrhage rate was 2.4%.

What was unusual was that 89% of the patients "violated" at least one common r-TPA treatment criterion. For instance, 42% were older than 80 years, 13% were older than 90 years, and 9% had prior stroke. Other patients had severe strokes; mild strokes with disabling symptoms; asymptomatic brain lesions, such as unruptured aneurysms; low National Institutes of Health Stroke Scale scores; prior trauma; bleeding problems or were on anticoagulant therapy; dementia; pregnancy; or abnormal blood glycemic levels. The mean contraindication rate was 1.4 per patient.

To be considered for treatment, patients must have had disabling neurologic symptoms caused by intracerebral ischemia. The only exclusion criterion was acute symptomatic intracranial hemorrhage. Most of the patients were seen within 4.5 hours of symptom onset, although a small subset received treatment beyond that time window if neuroimaging results suggested intervention would be beneficial.

Dr. Rose said that modification of IV r-TPA stroke treatment criteria is urgently needed. "Most of the criteria that have been followed are not evidenced based, and a lot of stroke centers use even more stringent criteria than the original trials suggested," he commented.

By giving IV r-TPA according to the SMART criteria, 20%-30% of stroke patients can potentially receive thrombolysis, he said, adding that at California Pacific Medical Center, the treatment rate went from 7% of ischemic stroke patients to 29% in the fourth quarter of 2009.

Dr. Rose said that he is careful to explain to patients when he obtains informed consent that they have a condition that might preclude them from receiving IV TPA at other centers, but that his hospital does not follow the same treatment criteria.

Dr. Rose said he had no conflicts to disclose.


Dr. Rose reports the results obtained at a single center where patients with acute ischemic stroke were treated with r-TPA using an approach that set aside several exclusion criteria from the NINDS trial on which FDA approval was based, thus constituting "off label" use of the drug. The only firm exclusion criterion indicated in this report is "acute, symptomatic intracranial hemorrhage." We cannot tell on what basis the treating physicians dispensed with other contraindications on a case-by-case basis.

This is not a study in which patients were randomized to a treatment decision based either on the established NINDS criteria or on the "SMART" criteria, so we cannot really know anything from this report about how the SMART criteria perform relative to the NINDS criteria. It is entirely possible that patients would have done better if the decision about thrombolysis had been based on NINDS criteria. Undoubtedly, Dr. Rose would scoff at such a notion; but the question is one of what the evidence tells us.

Fourteen years after the FDA approved r-TPA for the treatment of acute ischemic stroke, many emergency physicians and neurologists remain deeply skeptical about its safety and effectiveness and are therefore very conservative in patient selection. This report seemingly would encourage us to be much more liberal in selecting patients for fibrinolytic therapy, suggesting that established contraindications often can be disregarded.

In general, the exclusion criteria in common use today are in place because they were selected for NINDS or because they were thought for some other reason to make sense. Most of them have not been evaluated in a rigorous fashion, and so it is reasonable to think that not all of them would prove essential were such an evaluation to be conducted.

Unfortunately, this "study" at California Pacific Medical Center provides no such rigorous evaluation of the exclusion criteria that were abandoned in the development of the "SMART" protocol. Therefore, while it tells us that patients at this institution appeared to do fairly well with the use of an alternative protocol, this report really adds nothing to the evidence-based-medicine database and offers no rationale for widespread adoption of the SMART criteria.

Dr. Robert C. Solomon,
ACEP News Medical Editor 

Related Links

Click here to
send us feedback