Pooled RCTs: In acute ischemic stroke with salvageable brain tissue, alteplase at 4.5 to 9 hours improved function

Abstract

Question In patients with ischemic stroke and pretreatment perfusion imaging, does IV alteplase at 4.5 to 9 hours after stroke onset or in wake-up stroke improve function compared with placebo? Review scope Included randomized controlled trials (RCTs) compared alteplase, 0.9 mg/kg (maximum 90 mg), with placebo in patients 18 years of age who had hemispheric ischemic stroke, symptom onset 4.5 to 9 hours before randomization or wake-up stroke with sleep midpoint <9 hours before randomization, and computed tomography (CT) perfusion imaging or perfusiondiffusion magnetic resonance imaging (MRI) before treatment. Primary outcome was excellent functional outcome at 3 months. Other outcomes included functional independence and all-cause mortality at 3 months, early neurologic improvement at 72 hours, and symptomatic intracerebral hemorrhage at 36 hours after treatment. PROSPERO CRD42019128036. Review methods PubMed (2006 to Mar 2019), Clinical Trials.gov, and references from a previous systematic review were searched for English-language RCTs. 2 RCTs (EXTEND and ECASS4-EXTEND) and 1 RCT subgroup (EPITHET) met inclusion criteria. Individual patient data were provided for patients with age, pretreatment National Institutes of Health Stroke Scale score, and 3-month modified Rankin Scale score data (n =414, mean age 73 y, 57% men, wake-up stroke 50%, large vessel occlusion 61%). All patients in EXTEND and ECASS4-EXTEND and most in EPITHET* had trial-defined perfusion mismatch. EXTEND and ECASS4-EXTEND were stopped early because positive results from other trials were published. 99% of patients had outcome data at 3 months. The 3 RCTs had low risk of bias for allocation concealment; blinding of patients, study personnel, and outcome assessors; and completeness of outcome data. Main results The main results are in the Table. Conclusion In acute ischemic stroke with salvageable brain tissue, alteplase at 4.5 to 9 hours after symptom onset or in wake-up stroke improved function at 3 months. Alteplase vs placebo at 4.5 to 9 h after symptom onset or in wake-up stroke in patients with acute ischemic stroke with salvageable brain tissue Outcomes Event rates RBI (95% CI) NNT (CI) Alteplase Placebo Excellent functional outcome at 3 mo 36% 29% 49% (10 to 89) 8 (4 to 34) Functional independence at 3 mo 49% 44% 31% (4 to 57) 8 (5 to 53) Early neurologic improvement at 72 h 28% 16% 105% (40 to 182) 7 (4 to 17) RRI (CI) NNH (CI) All-cause mortality at 3 mo 14% 9.0% 48% (18 to 152) Not significant Symptomatic intracerebral hemorrhage at 36 h after treatment 4.7% 0.5% 830% (23 to 5456) 25 (4 to 875) Abbreviations defined in Glossary. RBI, RRI, NNT, NNH, and CI calculated from placebo event rates and odds ratios, adjusted for age and National Institutes of Health Stroke Scale score, in article based on individual patient data analysis of 3 randomized controlled trials (n =414). Modified Rankin Scale score 0 to 1. Modified Rankin Scale score 0 to 2. National Institutes of Health Stroke Scale score 0 to 1 or reduction 8 points. Commentary The meta-analysis by Campbell and colleagues included 3 randomized placebo-controlled trials (EXTEND, ECASS4-EXTEND, and EPITHET) and found that tissue plasminogen activator (t-PA) at 4.5 to 9 hours after stroke improved the odds of achieving an excellent functional outcome (modified Rankin Scale score 0 to 1), functional improvement, or functional independence at 3 months, and promoted early neurologic improvement at 72 hours. Most patients had salvageable brain tissue based on CT perfusion or MRI perfusiondiffusion mismatch assessment. A prespecified subgroup analysis based on automated reassessment of perfusion mismatch did not show a mismatchtreatment interaction for the primary outcome (P =0.43), although the analysis was underpowered for this comparison. Lack of such an interaction may relate to the small proportion of patients who did not have a mismatch. Also, the methodology to statistically show the utility of the mismatch may require improved penumbra-based predictive models (1). Although the role of penumbral imaging in stroke treatment requires further study, the meta-analysis results may influence the current practice of using automated mismatch criteria to inform treatment decisions in the 4.5- to 9-hour window. First, stroke centers without thrombectomy capability could consider expanding the t-PA treatment window. Second, t-PA could be given before transfer to comprehensive stroke centers for thrombectomy. Third, it could be used when distal thrombi are inaccessible to thrombectomy or when vascular pathology prevents access to the affected vessels. Because current guidelines suggest immediate thrombectomy for eligible patients who receive t-PA (2), access to a thrombectomy-capable facility should be expedited for eligible patients treated with t-PA in an extended time window. For patients who lack immediate access to, or are ineligible for, mechanical thrombectomy, the meta-analysis supports the use of IV t-PA at 4.5 to 9 hours after stroke onset to protect salvageable brain tissue.