Intravenous recombinant tissue plasminogen activator (rtPA) is believed to provide effective thrombolysis for ischemic stroke if given within three hours of symptom onset. More recent studies have raised the possibility of a longer therapeutic window, but the benefits of starting rtPA more than three hours after the clinical onset of stroke remain unclear. A meta-analysis of all available trials aimed to establish whether time is a critical factor in thrombolytic therapy for ischemic stroke.

The researchers examined data from six major clinical trials, which included 99 percent of all patients in randomized controlled clinical trials of intravenous rtPA in patients with acute ischemic stroke. All trials had similar inclusion criteria based on neurologic deficit and computed tomography. The time allowed to start treatment varied between trials from 90 to 360 minutes. Some trials had age limitations for participants, and others excluded patients who were rapidly improving or had only minor symptoms. The dose of rtPA used varied from 0.9 to 1.1 mg per kg, but all studies gave a 10 percent bolus during the first minute, followed by an infusion lasting one hour. The trials also differed in output measures, and the duration of follow-up ranged from 30 to 90 days.

The meta-analysis included 2,775 patients who were randomized to receive rtPA thrombolysis or placebo. The patients were predominantly white (85 percent), and the median age was 68 years. The median score on the National Institutes of Health Stroke Scale (NIHSS) was 11. The median onset-to-treatment time was 243 minutes, and 67 percent of patients were treated for longer than three hours after symptom onset. The researchers adjusted for age, blood glucose level, NIHSS score, diastolic blood pressure on admission, and previous hypertension.

At three months, 14 percent of the patients had died or had the worst assigned score. An additional 25.5 percent were severely disabled. The odds of a favorable outcome at three months were related to prompt treatment. Patients treated within 90 minutes had odds of a good three-month outcome of 2.8, compared with 1.6 in those treated 91 to 180 minutes after symptom onset; 1.4 in those treated 181 to 270 minutes after symptom onset; and 1.2 in those treated 271 to 360 minutes after onset. Hemorrhage, which occurred in 5.9 percent of treated patients compared with 1.1 percent of control patients, was not related to timing of rtPA treatment.

The authors conclude that early administration of rtPA is associated with greater chance of good outcomes, but that the window for potential benefit extends beyond the suggested limit of three hours from onset of symptoms.