Why are stroke patients excluded from TPA therapy?
An analysis of patient eligibility

We completely agree mild stroke and so called "neurologic improvement" are no guarantee of good recovery. [1]. Although a neurologic score may be suggestive of a mild stroke, the behavioral consequences can often be functionally disastrous.

In the case described by Fink et al, a patient with a right hemisphere improved (because the NIH score stroke had declined). This led to an initial decision not to treat the patient. Following transfer of this patient and sophisticated neuro imaging, although the patient improved, a large perfusion deficit was discovered. The decision was reversed and treatment was given, because of the large perfusion deficit, resulting in an extremely good neurologic outcome. Therefore, the patient was treated, despite the improvement, on the basis of persistent ischemia, rather than using the NINDS rules. The NINDS study is based on NIH scores which can be unevenly low for right hemisphere strokes (because the score is so dependent on language), i.e. a low score with right brain involvement belies the severity of the stroke.

There is an excess risk of symptomatic intracranial hemorrhage unless there is strict adherence to the NINDS criteria.[5]. By sticking to these rules we have maintained low risks of intracerebral hemorrhage and have extended our single center study of effectiveness to a national study (The Canadian Activase for Stroke Effectiveness Study - CASES) which has now collected over 1,000 under three hour stroke patients who have been treated with TPA. [6]. As described in our paper [1], we have become increasingly concerned about those patients who are excluded from TPA therapy on the basis of the NINDS rules. We found that no fewer than one -third of patients who were denied treatment because the stroke was either too mild or the neurologic score was improving ended up either dead or dependent at three months. We would presume they were at a relatively low risk of treatment induced intracerebral hemorrhage but there is no randomized data which suggests these patients would benefit from the intervention. Assuming the risk is low or lower, we believe that these patients should be studied in a randomized, prospective trial.

We too, have been very impressed with the eloquence of the CT scan. The unenhanced scan provides a surrogate for diffusion MRI. [7]. Our scoring system, the Alberta Stroke Program Early CT Score (ASPECTS) has been used both in Calgary and throughout Canada to help predict those patients who are most likely to benefit. [8]). Randomizing patients on the basis of an ASPECT score goes beyond the NINDS trial. It is our current thinking by using the ASPECT score [8]) to minimize risk, we could break out of the so called "NINDS box". For those with a good ASPECT score we could treat (or at least randomize to a new trial) patients, beyond three hours, those who wake up with a deficit or those who have an unknown time of onset. We would also propose that those patients who are judged to be either too mild or who are recovering (assuming the ASPECT score is good) might also benefit from intervention with low risks of hemorrhage.

What is exciting about the report of Fink et al, is it presages the use of a CT perfusion index (perhaps a P-ASPECT score) that will allow us to pick those patients who are too mild or who are improving but have a persistent ischemic attack (PIA) with a transient neurologic deficit. [9]). We predict that perfusion CT and careful quantitation of unenhanced CT (ASPECTS) will allow us to design trials that are more inclusive, affording safe TPA therapy to more stroke patients. It is our contention that we are excluding far too many stroke victims from TPA therapy and feel that our study of patient eligibility might predict what a new more inclusive trial might look like [1].

References:

5. Buchan AM, Barber PA, Newcommon N et al. "Effectiveness of t-PA in Acute Ischemic Stroke: Outcome Relates to Appropriateness." Neurology 2000;54:679-684.

6. Hill MD, Lawence K, Buchan AM, "Canadian Activase for Stroke Effectiveness Study (CASES): A Multi-Stakeholder Collaboration." Canadian Journal of Neurological Sciences 2001;28.

7. Barber PA, Demchuk AM, Hill MD et al. "A Comparison of CT versus MR Imaging in Acute Stroke using ASPECTS: Will the "New" replace the "Old" as the Preferred Imaging Modality?" Stroke 2001;32:325 (Abstract).

8 Barber PA, Demchuk, AM, Buchan AM. For the ASPECTS Study Group "Validity and Reliability of a Quantitative Computed Tomography Score in Predicting Outcome of Hyperacute Stroke before Thrombolytic Therapy." Lancet 2000;355:1670-1674.

9. Buchan AM, Aspects of Stroke Imaging. Canadian Journal of Neurological Sciences 2001;28:99-100.

The report of Barber et al. of the exclusion of stroke patients from treatment with tPA raises some important issues for stroke physicians.[1] The authors' finding that a third of the patients excluded from tPA treatment on the basis of clinical assessment of "mild" stroke severity or "improvement" ultimately had a poor outcome is of particular concern. We recently experienced an illustrative case:

An 82 year-old woman presented to an affiliated hospital 20 minutes after the sudden onset of left hemiplegia and neglect. The emergency physician contacted the stroke fellow at our hospital after his examination for advice. A CT scan was performed and reported negative. The treating physician was reluctant to proceed with tPA treatment after the CT scan because the patient had improved from complete paralysis to mild weakness on the left. The patient was immediately transferred to our hospital for expert assessment. On arrival, she was conversant. Motor deficits were mild. However, she had gaze deviation, neglect and hemisensory loss; NIHSS score 12. MRI was contraindicated due to pacemaker. Perfusion CT revealed an abnormality in the right temporal region and CT angiography (CTA) revealed occlusion of the inferior division of the right MCA. Intravenous tPA was administered. The next day, repeat CTA showed recanalization of the right MCA; her NIHSS score was 1.

It has been our impression that emergency clinical assessment of stroke severity may be particularly problematic for patients with right-hemisphere syndromes without major motor deficits, such as the patient resented. While the severity of the clinical deficits in right-hemisphere stroke may not be as immediately obvious as the dysphasia caused by left-hemisphere lesions, the resulting disability is at least as great.[2] The NIHSS has a greater weighting for aphasia than neglect.[3,4] For a given NIHSS score, the volume of right hemisphere strokes on chronic CT is larger than the volume of left hemisphere strokes.[4] We have found that 59 patients presenting with an NIHSS score 5 or less who had an MRI within 24 hours of stroke onset, the mean acute diffusion-weighted MRI lesion volume for right-hemisphere strokes was 8.8 cm [3], compared with 3.2cm [3] on the left (p=0.04). We would be interested to know what proportion of the patients in the authors' series who were excluded from treatment because of mild or improving deficits presented with right-hemisphere stroke, and whether right-hemisphere cases were over-represented in the group with subsequent poor outcome.

The inappropriate use of clinical exclusion criteria for tPA identified by the authors is of great concern. We believe that equal rigor must be used to ensure the inclusion of deserving patients for tPA treatment as that used to exclude those at greater risk of treatment complications. The current definition of "improving" stroke is too ambiguous. We agree with the authors that non-invasive neuroimaging techniques such as diffusion and perfusion MRI, MR angiography, CTA or transcranial Doppler ultrasonography can be useful in making the decision to treat with tPA when there is uncertainty on clinical grounds alone.

References

1. Barber PA, Zhang J, Demchuk AM, Hill MD, Buchan AM. Why are stroke patients excluded from TPA therapy? An analysis of patient eligibility. Neurology 2001;56:1015-1020.

2. Ween JE, Alexander MP, D'Esposito M, Roberts M. Factors predictive of stroke outcome in a rehabilitation setting. Neurology 1996;47:388-392.

3. Brott T, Adams HP, Olinger CP, et al. Measurements of acute cerebral infarction: a clinical examination scale. Stroke 1989;20:864-870.

4. Woo D, Broderick JP, Kothari RU, et al. Does the National Institutes of Health Stroke Scale favor left hemisphere strokes? Stroke 1999;30:2355-2359.