We find it fascinating that the views expressed by the protagonists of our first controversy reflect the trans-Atlantic dichotomy of opinion concerning the use of tissue plasminogen activator (tPA). As most clinicians know, tPA is now licensed in North America (United States since 1996) and a number of South American countries. Licensing has been slower in Europe, Australasia, and Asia. Indeed, there is only restricted licensing in a limited number of countries outside North America.
Perhaps understandably, the use of tPA has been taken up more enthusiastically in the United States where the only unequivocally positive trial(s) of therapy was conducted and, importantly, under the auspices of their national funding body. The main difficulty is that an equivalent trial with a 3-hour time window has not been conducted elsewhere, although the cumulative meta-analysis supports the findings of the NINDS trial. Of importance, open phase IV studies with tPA indicate that the positive benefit/risk ratio of the NINDS trial may be replicated, provided that the therapy is administered in expert centers with strict adherence to the NINDS protocol.
Given our current dependence on evidence based medicine, investigators only wish to advocate a therapy when it is of proven value. Only a few thousand patients have been enrolled in tPA stroke trials. In contrast, tens of thousands of patients were studied in the mega-trials of myocardial infarction before thrombolysis became accepted. This may be because the single end point of mortality in myocardial infarction necessitates larger sample sizes than stroke trials, where a combined end point of death and disability increases power. Some would say that unnecessary additional randomized trials of thrombolysis for myocardial infarction were performed, well after the benefits were obvious by cumulative meta-analysis (had it been used). Are we in the same position with trials of thrombolysis for stroke?
In essence, it comes down to a view of when is enough evidence enough. Clearly the view differs from physician to physician, country to country, and continent to continent.
Our view is that tPA should now be used in expert centers, according to NINDS guidelines, less than 3 hours after stroke onset. However, there is still uncertainty about some subgroups within this time frame, such as those with very severe neurological deficits or extensive early ischemic CT changes, and randomization here may be appropriate. Perhaps more importantly, there is a real need for randomized trials of thrombolysis beyond the 3-hour time window and identification of neuroimaging parameters, such as diffusion/perfusion MRI, that may predict better outcome. Herein lies the real challenge.
Geoffrey A. Donnan MD, FRACP
Stephen M. Davis MD, FRACP
The opinions expressed in this editorial are not necessarily those of the editors or of the American Stroke Association.