Am Fam Physician. 1998;57(1):73-80
Successful treatment of patients with ischemic stroke depends on the ability to treat within three hours of onset, because tissue plasminogen activator has not yet been proved effective beyond this time frame. The two major causes of delay in treatment are failure, on the part of the patient or family, to recognize stroke symptoms and failure to access the medical system most efficiently—by calling 911. Hospital stroke teams can shorten the time between patient arrival at the emergency department and treatment. Guidelines for the evaluation and treatment of potential stroke patients are presented, along with goal times for arrival-to-treatment intervals.
In June of 1996, the U.S. Food and Drug Administration approved the use of intravenous recombinant tissue plasminogen activator (t-PA) for use in appropriately selected patients with ischemic stroke if administered within three hours from onset of symptoms. Approval was based on two studies funded by the National Institute of Neurological Diseases and Stroke (NINDS).1 These studies represent the first demonstrated effective treatment for ischemic stroke.
Animal models of thrombolytic therapy for acute ischemic stroke indicate that thrombolytic therapy should be administered within the first hour after stroke onset to provide the greatest chance for success.2 Evidence for effectiveness of thrombolysis beyond three hours after onset of ischemia in animals is minimal.2–5 With regard to human studies, the two NINDS studies1 of t-PA are the only ones that indicate a convincing benefit with t-PA, but only if t-PA is given within three hours of onset to appropriately selected patients. Time to treatment is the most important determinant of success in treating ischemic stroke with t-PA.
The European Cooperative Acute Stroke Study of t-PA (ECASS)6 showed no benefit and an increased risk of bleeding complications with t-PA. It has been suggested that the reason for this disappointing outcome rests in the selection of patients in this trial. Patients were treated in a six-hour rather than a three-hour window and may have had more extensive ischemic damage on the baseline computed tomographic (CT) scan than is now thought to be acceptable.
Current Logistics of Stroke Treatment
Because treatment within three hours of onset is crucial, successful treatment of stroke patients in the general population is dependent on educating stroke patients to arrive at a hospital within this time frame. The major components of the process of evaluation and treatment of the stroke patient include: recognition of stroke by patient or family, contact with and access to medical care, recognition of a “possible stroke” by the 911 dispatcher, identification of stroke as the highest level of emergency by the local Emergency Medical System (EMS), rapid transport of the stroke patient to the emergency department, evaluation of stroke in the emergency department and initiation of appropriate treatment.
In addition, stroke damage often impairs either perception of a problem (parietal lobe) or the ability to communicate (temporal, frontal and parietal lobes).
However, the major delay in time to treatment is probably caused by the lack of recognition of stroke signs and symptoms and failure to make the proper response. Only 58 percent of patients in a recent Gallup Poll of persons at least 50 years of age recognized weakness of an arm or leg as symptoms of stroke, and only 32 percent recognized difficulties with speaking as a warning sign of stroke (unpublished data). In a study conducted in greater Cincinnati,10 only 37 percent of stroke patients and their family members knew that their symptoms were caused by a stroke at the time of symptom onset. Of the 163 patients in the study, 62 (38 percent) did not know a single warning symptom of stroke, and another 45 (28 percent) could identify only one symptom.
The second most important cause of delay in evaluation of stroke patients is the call made to access the medical system.7 In three U.S. communities, potential stroke patients who first called their primary physician arrived at the emergency department within a mean time of 379 minutes, compared with a mean time of 155 minutes for patients who called 911, the emergency access number.7 It should be noted that in many communities, stroke is not considered by the EMS to be a level-one emergency, equivalent to trauma and myocardial infarction. Stroke, or “brain attack,” should be treated with the same urgency as trauma and myocardial infarction. Family physicians should support efforts to label stroke “the highest level of emergency” for their community EMS.
Family physicians should educate their patients about the warning signs and symptoms of stroke and instruct them to call 911 if a stroke is suspected. This advice is particularly important for patients at higher risk for stroke, such as persons with a previous transient ischemic attack or small stroke.
Stroke Evaluation in the Emergency Department
Only 3.6 percent of more than 17,000 potential stroke patients were eligible for treatment in the two NINDS randomized studies on t-PA for acute stroke.11 In contrast, approximately 33 percent of patients with an acute myocardial infarction in the National Registry of Myocardial Infarction II12 are currently treated with a thrombolytic agent (26 percent) or alternative reperfusion methods (7 percent) such as recanalization by invasive cardiac catheterization or angioplasty.
It is likely that improving public awareness and expediting prehospital care will increase the number of stroke patients eligible for thrombolytic treatment. Current differences in the prehospital approach to stroke and myocardial infarction are compared in Table 1. The major difference at present is the perception among physicians and hospitals that myocardial infarction is an emergency, while stroke is not. For example, in the NINDS studies, the most ambitious stroke study thus far in terms of time to treatment, the goal for time from arrival at the emergency department to treatment was 55 minutes. In contrast, current guidelines for myocardial infarction talk about a time from arrival to treatment of 30 minutes.13
| Factor | Stroke | Myocardial infarction |
|---|---|---|
| Public awareness | Poor public knowledge of warning signs | Excellent public awareness of warning signs |
| EMS designation | Inconsistent EMS designation as highest level of emergency | EMS designation as highest level of emergency |
| Presenting symptoms | Multiple possible presenting symptoms | A few presenting symptoms |
| Complicating clinical features | Impairment of communication and perception | No impairment |
| Pain as a clinical clue | No pain | Often prominent pain and shortness of breath |
| Treatment window | Treatment window of three hours | Treatment window of at least six hours |
| Treatment goal | Recommended time from emergency department arrival to treatment: 60 minutes (new goal—rarely reached at present) | Recommended time from emergency department arrival to treatment: 30 minutes (goal frequently attained) |
In Houston, the notification and presence of a designated “stroke team” shortened the time to examination of a stroke patient by a physician by 13 minutes and the time to CT scan by 63 minutes.14 This study illustrates the importance of stroke teams in the early evaluation and treatment of stroke patients. The term “stroke team” refers to an organized group of physicians, nurses and allied health personnel committed to rapid treatment, with expertise in the evaluation and treatment of stroke patients.
In December 1996, the NINDS-sponsored National Symposium on Rapid Identification and Treatment of Acute Stroke15 recommended guidelines for the evaluation and treatment of potential stroke patients (Table 2). Two key times are a time from arrival to CT scan of 25 minutes and a time from arrival to treatment (with t-PA or as-yet potential agents) of 60 minutes, with a goal of 80 percent compliance. These times should be viewed as ideal goals for all hospitals treating patients with acute stroke.
| Task | Time from emergency department arrival to completion of task |
|---|---|
| Evaluation by physician | 10 minutes |
| “Stroke expertise” or “neurologic expertise” contacted (i.e., stroke team) | 15 minutes |
| CT scan of the head | 25 minutes |
| Interpretation of CT scan | 45 minutes |
| Start of treatment | 60 minutes |
| Monitored bed in unit | 3 hours |
This conference also provided a recommended pathway for triage and evaluation of stroke patients in the emergency department (Figure 1). Triage is one way to focus resources. For example, stroke patients with a mild or moderate and stable neurologic deficit who appear more than six hours after symptom onset should be evaluated promptly but not with the same urgency that is required to treat a patient who arrives within a very early window (e.g., less than three hours after onset) with t-PA.
Treatment of ischemic stroke patients with a thrombolytic agent requires that a physician experienced in the care of stroke patients decide that a given patient has a significant neurologic deficit due to stroke and not to another cause. This patient must then meet all the inclusion and exclusion criteria for treatment with t-PA that were used in the NINDS study or in the package insert for t-PA in acute ischemic stroke (Table 3). CT scan of the brain must not show acute hemorrhage or a large, already-developing cerebral infarction. Finally, treatment must be administered within three hours of stroke onset.
| Eligibility for IV treatment with t-PA | |
| Age 18 or older | |
| Clinical diagnosis of ischemic stroke causing a measureable neurologic deficit | |
| Time of symptom onset well established to be less than 180 minutes before treatment would begin | |
| Patient selection: contraindications and warnings | |
| Evidence of intracranial hemorrhage on pretreatment CT† | |
| Clinical presentation suggestive of subarachnoid hemorrhage, even with normal CT† | |
| Active internal bleeding† | |
| Known bleeding diathesis, including but not limited to: | |
| • Platelet count less than 100,000 per mm3 (100 × 109 per L)† | |
| • Patient has received heparin within 48 hours and has an elevated aPTT (greater than upper limit of normal for laboratory)† | |
| • Current use of oral anticoagulants (e.g., warfarin [Coumadin]) or recent use with an elevated prothrombin time greater than 15 seconds†‡ | |
| Within 3 months: any intracranial surgery, serious head trauma or previous stroke† | |
| On repeated measurements, systolic blood pressure is greater than 185 mm Hg or diastolic blood pressure is greater than 110 mm Hg at the time treatment begins or patient requires aggressive treatment to reduce blood pressure to within these limits† | |
| History of intracranial hemorrhage† | |
| Known arteriovenous malformation or aneurysm† | |
| Patient was observed to have seizure at the same time the onset of stroke symptoms was observed§ | |
| History of gastrointestinal or urinary tract hemorrhage within 21 days§ | |
| Recent arterial puncture at a noncompressible site§ | |
| Recent lumbar puncture§ | |
| Abnormal blood glucose level (<50 or >400 mg per dL [<2.8 or >22.2 mmol per L])§ | |
| Only minor or rapidly improving stroke symptoms§ | |
| Postmyocardial infarction pericarditis§ | |
Patients who are treated with t-PA must be cared for in an experienced intensive care unit or a monitored unit and should have access to neurosurgical expertise if a hemorrhagic complication occurs. No heparin or anti-platelet agents should be administered until 24 hours after initiation of t-PA treatment (because of the increased risk of bleeding) and a 24-hour safety CT has ruled out intracranial hemorrhage. Arterial blood pressure should be kept just below a systolic pressure of 185 mm Hg and a diastolic pressure of 105 mm Hg during the first 24 hours to minimize the risk of intracerebral hemorrhage. Appropriate antihypertensive agents include small doses of intravenous medications such as labetalol (Normodyne, Trandate). A suggested plan of baseline laboratory studies and physician orders for the first 24 hours of treatment after ischemic stroke is shown in Table 4.
| Continue emergency department orders for t-PA infusion and monitoring vital signs and neurologic checks until two hours after start of t-PA infusion. | |
| Check vital signs (blood pressure, pulse, respiration) and make neurologic checks (level of consciousness, arm/leg weakness) every 30 minutes for 6 hours, then every 60 minutes for 16 hours after starting t-PA. | |
| Bleeding precautions: check puncture sites for bleeding or hematomas. Apply digital pressure or pressure dressing to active compressible bleeding sites. Evaluate urine, stool, emesis or other secretions for blood. Perform occult blood testing (guaiac) if there is evidence of bleeding. | |
| Call stroke-team physician (pager #_______) immediately to seek evidence of bleeding, neurologic deterioration or vital signs outside the following parameters: | |
| Systolic blood pressure greater than 185 mm Hg or systolic blood pressure less than 110 mm Hg | |
| Diastolic blood pressure greater than 105 mm Hg or diastolic blood pressure less than 60 mm Hg | |
| Pulse less than 50 | |
| Respirations greater than 24 | |
| Decline in neurologic status or worsening of stroke signs | |
| 0.45 normal saline or normal saline IV, to keep vein open, at 50 mL per hour for 24 hours | |
| Oxygen at 2 L per minute by nasal cannula | |
| Continuous cardiac monitoring | |
| Intake and output | |
| Diet: nothing by mouth except medications for 24 hours | |
| Bed rest | |
| Medications: acetaminophen, 650 mg orally for pain every 4 to 6 hours, as needed | |
| Patient's previously prescribed regular medications, if appropriate. | |
| No heparin, warfarin or aspirin for 24 hours. After 24 hours: CT to exclude intracranial hemorrhage before any anticoagulants are administered. | |
Each hospital should develop a treatment plan for acute stroke that reflects its abilities and limitations. Hospitals without brain imaging capacity should never treat patients with a thrombolytic agent. Hospitals with easy-access brain imaging, radiologic expertise and an experienced physician should be able to treat appropriate patients with t-PA. However, if an active intensive care unit or neurosurgical expertise is lacking, patients who are treated with t-PA should be transferred urgently after treatment has begun to another hospital where these capabilities are present.
Teleradiography will probably play an increasingly important role in the treatment of stroke patients at rural and small outlying hospitals. In the future, treatment of ischemic stroke patients with neuroprotective agents (currently in development) that are safe and do not require a CT scan before treatment could theoretically be done at any hospital, possibly even in the field by paramedics. However, this possibility depends on results of ongoing and future randomized trials of such agents.
Benefits and Risks of t-PA in Ischemic Stroke
Patients with an ischemic stroke who are appropriately selected and treated with t-PA within three hours from onset are at least 33 percent more likely to have a normal or near-normal neurologic examination score (the National Institutes of Health [NIH] stroke scale) or normal or near-normal Barthel index score (a measure of the activities of daily living) at three months after treatment. In terms of absolute benefit, one patient, who otherwise would not, returns to normal or near normal at three months for every six to nine patients who receive t-PA rather than placebo.16 However, t-PA is not a cure for ischemic stroke: 50 percent of patients who were treated with t-PA in the two NINDS studies had mild or greater disability as measured by a Barthel index score of less than 95 at three months (100 is normal and 0 is the worst score).1
The major risk of t-PA is symptomatic intracerebral hemorrhage, which occurred in 6.4 percent of patients who received t-PA, as compared with 0.6 percent of patients who received placebo.1 These figures represent an absolute difference in the risk of symptomatic intracerebral hemorrhage of 6 percent. Seventy-five percent of the patients with a symptomatic intracerebral hemorrhage were dead at three months. Yet despite the risk of intracerebral hemorrhage, the mortality at three months was insignificantly lower in patients treated with t-PA (17 percent) than in placebo-treated patients (21 percent). Two reasons may underlie the similar mortality of t-PA–treated and placebo-treated patients despite a higher risk of symptomatic intracerebral hemorrhage among patients treated with t-PA. First, most patients who had an intra-cerebral hemorrhage had large strokes and were likely to do poorly regardless of the presence of intracerebral hemorrhage within the damaged brain. Secondly, t-PA probably makes some “big” strokes much smaller. The patients with small strokes are less likely to die.
Patients treated with t-PA who have very large strokes (very severe neurologic deficits, meaning an NIH stroke scale score greater than 20) and who already have evidence of a large acute ischemic stroke on baseline CT have an increased risk of symptomatic intra-cerebral hemorrhage.17 However, patients in these two subgroups who receive t-PA are more likely to return to normal than patients who are treated with placebo. The decision to use t-PA in these patients should be made only after frank discussion of the potential risks and benefits with both the patient and the family.