Code Stroke Protocol

Abstract & Commentary

Source: Asimos AW, et al. Therapeutic yield and outcomes of a community teaching hospital code stroke protocol. Acad Emerg Med 2004;11:361-370.

With the publication of the National Institute of Neurological Disorders and Stroke (NINDS) trial in 1995,1 and the subsequent adoption of its findings by the American Heart Association starting in 1996,2 the use of intravenous tissue plasminogen activator (tPA) has been recommended for the treatment of acute stroke in a subset of patients who meet predefined criteria. Almost from its inception, however, there has been controversy surrounding its safety, efficacy, and applicability to a general emergency department (ED) population. A number of subsequent studies have called into question the findings of the original trial that documented a 6% rate of symptomatic intracranial hemorrhage (ICH) rate, a 17% three-month mortality, and a 12% absolute improvement in the Barthel score, testing functionality at three months. A chief concern of subsequent studies is whether these results can be duplicated in community EDs, where there are less specialized resources for the treatment of stroke.

This retrospective review, performed at Carolinas Medical Center, a large, community teaching hospital with an annual ED census of more than 100,000, looked at its experience with intravenous tPA for stroke during a 56-month time period (1997-2001) and compared its experience with that of the NINDS trial. At this institution, when a patient is identified as a potential code stroke patient, an emergency physician (EP) screens the patient for eligibility criteria. If found eligible, the EP works closely with a private practice neurology group, which will respond to see the patient. The decision to proceed with thrombolysis is made jointly by the EP and the neurologist.

During the specified time period, 255 patients were deemed to meet code stroke criteria. After evaluation, 60 were treated with intravenous tPA (24%). The 36-hour symptomatic ICH rate was 10% (vs NINDs = 6%). At three months, 60% of treated patients achieved greater than 95 on the Barthel Index (vs NINDS = 52%).

Mortality at three months in this study was 12% (vs NINDS =17%). Protocol violations, usually cited as a key feature in increasing risk of hemorrhage, occurred in 32% of patients in this study, but were equally common in those with and without symptomatic ICH. The study concludes that intravenous tPA can be given safely in the community hospital setting for stroke, with outcome measures that closely parallel those found in the NINDS trial. While protocol violations were relatively common at this community teaching hospital, the three-month functional outcomes and mortality rates were equal to or better than those found in the original NINDS study.

Commentary by Andrew D. Perron, MD, FACEP, FACSM

To thrombolyse or not to thrombolyse…

To say that the evidence is conflicting on this topic is a significant understatement. There is now a substantial body of literature on both sides of this topic, and the clinician can find support for either view with little difficulty. So where does this study leave me in terms of making a decision regarding thrombolysis in my community teaching hospital?

While handicapped by its retrospective nature, this study has direct applicability to my ED, and I suspect a large number of other EDs in the United States. Perhaps it is only because I practice in the same environment as described, but this study clearly lays out a logical pathway to developing a code stroke protocol at my institution. Asimos and colleagues are clear in their discussion of the mechanics, algorithms, and care protocols they use to get these results. If I were charged with creating a similar protocol at my institution, I would use this as a blueprint.

That being said, this study does not help me answer the bigger question of whether we should be offering thrombolysis as the standard of care in all EDs across the land. I am continually amazed at the reported rate of protocol violations in these studies, where we go to such extremes to specifically avoid them. It is to the point that any study that reports a rate of less than 30% is immediately suspect. I am somewhat heartened by the fact that in this study the rate of symptomatic ICH was the same in both groups; this has not been shown to be true in other studies. In fact, the one lesson I have learned from previous studies is that if you cannot do this well (i.e., you have significant protocol violations), you unquestionably will hurt patients. Whether this can be reproduced in other community settings remains to be seen.

Dr. Perron, Residency Program Director, Department of Emergency Medicine, Maine Medical Center, Portland, ME, is on the editorial board of Emergency Medicine Alert.


1. Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke. N Engl J Med 1995;333:1581-1587.

2. Adams HP Jr, et al. Guidelines for thrombolytic therapy for acute stroke: A supplement to the guidelines for the management of patients with acute ischemic stroke. A statement for healthcare professionals from a Special Writing Group of the Stroke Council, American Heart Association. Circulation 1996;94:1167-1174.