Background: Patients with stroke may have cognitive deficits that impact their capacity to provide informed consent for research. Some institutional review boards restrict surrogate consent to persons who have specific legal authority to provide it. We examined the importance of surrogate consent in the National Institute of Neurological Disorders and Stroke (NINDS) rt-PA Stroke Trial, the study that led to the only US Food and Drug Administration–approved treatment for acute ischemic stroke.

Methods: The NINDS rt-PA Stroke Trial randomized subjects with ischemic stroke to treatment with recombinant tissue plasminogen activator (rt-PA) or placebo. We compared the baseline characteristics and clinical outcomes of subjects enrolled by self-consent with those of subjects enrolled by surrogate consent.

Results: Surrogate consent was used to enroll 439 of 624 (70%) subjects. Subjects enrolled by surrogate consent were older (68.5 vs 63.4 years, p < 0.001), had more severe strokes (median NIH Stroke Scale score 17 vs 9, p < 0.001), and were less likely to make a good recovery (p < 0.001 for all measures) than patients who provided their own consent. There was no interaction between method of consent and response to rt-PA. If the trial had used the same sample size and recruited at the same rate but excluded patients who could not provide their own consent, it would have taken 12.5 years to complete.

Conclusions: The National Institute of Neurological Disorders and Stroke recombinant tissue plasminogen activator (rt-PA) Stroke Trial would not have been completed in a timely fashion without subjects enrolled by surrogate consent. Furthermore, exclusion of subjects who could not provide their own consent would have severely limited the generalizability and value of trial results.