The New England Journal of Medicine SEPTEMBER 25, 2008

Thrombolysis with Alteplase 3 to 4.5 Hours after Acute Ischemic Stroke (ECASS-3)

Werner Hacke, Markku Kaste, Erich Bluhmki, Miroslav Brozman, Antoni Dávalos, Donata Guidetti, Vincent Larrue, Kennedy R. Lees, Zsolt Medeghri, Thomas Machnig, Dietmar Schneider, Rüdiger von Kummer, Nils Wahlgren, and Danilo Toni

Bottom Line

The ECASS-3 trial demonstrated that intravenous alteplase administered between 3 and 4.5 hours after the onset of acute ischemic stroke symptoms significantly improved functional outcomes at 90 days compared to placebo, despite an increased risk of symptomatic intracranial hemorrhage.

Key Findings

1. A favorable functional outcome (modified Rankin Scale score 0–1) at 90 days was achieved by 52.4% of patients in the alteplase group compared to 45.2% in the placebo group (odds ratio 1.34; 95% CI 1.02–1.76; P=0.04).
2. Symptomatic intracranial hemorrhage occurred more frequently in the alteplase group (2.4%) compared to the placebo group (0.2%; P=0.008).
3. There was no statistically significant difference in mortality between the alteplase group (7.7%) and the placebo group (8.4%; P=0.68).
4. Safety analyses indicated a significantly higher rate of any intracranial hemorrhage in the alteplase group (27.0%) versus the placebo group (17.6%; P<0.001).

Study Design

Design
RCT
Double-Blind
Sample
821
Patients
Duration
90 days
Median
Setting
Multicenter, Europe
Population Adults 18 to 80 years old with acute ischemic stroke presenting 3 to 4.5 hours after symptom onset.
Intervention Intravenous alteplase (0.9 mg per kilogram of body weight, maximum 90 mg).
Comparator Placebo (saline solution).
Outcome Favorable functional outcome at 90 days, defined as a score of 0 or 1 on the modified Rankin scale.

Study Limitations

The study excluded patients with severe strokes (NIHSS >25), those older than 80 years, and individuals with a history of both prior stroke and diabetes.
There was an imbalance in baseline characteristics, with the alteplase group having lower mean NIHSS scores and a lower prevalence of prior stroke compared to the placebo group.
The benefit observed, while statistically significant, was more modest than that reported in earlier trials for the 0–3 hour window, emphasizing the time-dependent nature of thrombolytic efficacy.

Clinical Significance

The results of the ECASS-3 trial were instrumental in expanding the approved therapeutic window for intravenous thrombolysis in acute ischemic stroke from 3 hours to 4.5 hours, allowing a broader population of patients to receive evidence-based reperfusion therapy.

Historical Context

Prior to this study, the landmark 1995 NINDS trial established intravenous alteplase as the standard of care only for patients treated within 3 hours of symptom onset. Subsequent pooled analyses and the ECASS-3 trial provided the necessary prospective randomized data to safely extend this treatment window.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the biochemical mechanism of action of alteplase (rt-PA) in the treatment of acute ischemic stroke, and why does the benefit of this therapy rapidly diminish as the time from symptom onset increases?

Key Response

Alteplase is a recombinant tissue plasminogen activator that binds to fibrin in a thrombus and converts trapped plasminogen to plasmin, which then initiates local fibrinolysis. The 'time is brain' concept is rooted in the pathophysiology of the ischemic penumbra; as time passes, the salvageable brain tissue progresses to irreversible infarction, and the risk of reperfusion injury and hemorrhagic transformation increases due to blood-brain barrier degradation.

Resident
Resident

The ECASS-3 trial extended the thrombolysis window to 4.5 hours, but utilized more stringent exclusion criteria than the original 0-3 hour NINDS trial. Which specific patient subgroups were excluded from the 3-4.5 hour window in ECASS-3 that are often eligible in the <3-hour window?

Key Response

ECASS-3 excluded patients older than 80 years, those with a combination of prior stroke and diabetes mellitus, those taking any oral anticoagulants (regardless of INR), and those with a baseline NIH Stroke Scale score >25. While some of these are now managed more flexibly in clinical practice, they represent the specific trial-based safety constraints for the extended window.

Fellow
Fellow

In ECASS-3, the rate of symptomatic intracranial hemorrhage (sICH) was significantly higher in the alteplase group compared to placebo (2.4% vs 0.2% per the ECASS definition). How does the definition of sICH used in this trial compare to the SITS-MOST or NINDS definitions, and how does this affect the interpretation of safety across different stroke trials?

Key Response

ECASS-3 defined sICH as blood on imaging plus an increase in NIHSS of 4 points or more, or if the hemorrhage led to death. The definition used greatly impacts reported rates; for instance, the NINDS definition is broader (any neurological decline plus any blood), typically yielding higher ICH rates. Fellows must recognize that 'safety' is definition-dependent when comparing alteplase, tenecteplase, or endovascular trials.

Attending
Attending

Given that the Number Needed to Treat (NNT) for a favorable outcome (mRS 0-1) increases from approximately 4-5 in the 0-3 hour window to 14 in the 3-4.5 hour window, how should this evidence influence your bedside 'teaching' and 'process' regarding 'door-to-needle' targets?

Key Response

The attending must emphasize that while ECASS-3 provides a 'safety net' for late arrivals, it should not justify a lack of urgency. The efficacy of alteplase is highly time-dependent; earlier treatment results in significantly better outcomes. Clinical processes should still target <60 (or <45) minute door-to-needle times regardless of whether the patient arrives at 1 hour or 3.5 hours.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

Analyze the potential impact of baseline imbalances in the ECASS-3 trial, specifically the slightly higher baseline NIHSS and higher prevalence of previous stroke in the placebo group. To what extent might these imbalances have inflated the perceived efficacy of alteplase in the primary endpoint analysis?

Key Response

Although the differences were small, baseline stroke severity is the most powerful predictor of outcome. If the placebo group was slightly 'sicker' at baseline, the treatment effect of alteplase might be over-estimated. A PhD-level critique would focus on the sensitivity analyses and adjusted odds ratios provided by the authors to determine if the 1.34 OR for favorable outcome remains robust after controlling for these prognostic variables.

Journal Editor
Journal Editor

If you were the lead reviewer for the ECASS-3 manuscript, how would you address the discrepancy between the significant improvement in the primary endpoint (mRS 0-1) and the lack of significant difference in mortality (7.7% vs 8.4%) despite the increased rate of sICH?

Key Response

A critical reviewer would flag that while alteplase improves functional independence, it does not necessarily improve survival. The editor would ensure the authors do not overstate the benefit as a 'survival' benefit and would require a transparent discussion on the trade-off: a higher chance of a 'perfect' recovery balanced against the risk of an iatrogenic, potentially fatal hemorrhage.

Guideline Committee
Guideline Committee

The AHA/ASA guidelines currently recommend intravenous alteplase in the 3-4.5 hour window with a Class I, Level of Evidence B-R. Why is the level of evidence lower than the 0-3 hour window (Level A), and what specific data would be needed to harmonize these recommendations?

Key Response

The 0-3 hour window is supported by multiple large-scale RCTs (NINDS, ECASS-1, ECASS-2, ATLANTIS), whereas the 3-4.5 hour window relies heavily on ECASS-3 and meta-analyses. To reach Level A, more randomized data or larger registry-based evidence confirming the results in the general population (especially those excluded in ECASS-3, like the elderly) was historically required. Current guidelines emphasize that while effective, the evidence base is narrower for the extended window.

Clinical Landscape

Noteworthy Related Trials

1995

NINDS rt-PA Stroke Study

n = 624 · NEJM

Tested

Intravenous rt-PA (alteplase)

Population

Patients with acute ischemic stroke within 3 hours of onset

Comparator

Placebo

Endpoint

Favorable outcome at 3 months (Barthel index, modified Rankin scale, Glasgow outcome scale, NIHSS)

Key result: Patients treated with rt-PA were at least 30% more likely to have minimal or no disability at 3 months compared to placebo.
1995

ECASS-I

n = 620 · JAMA

Tested

Intravenous alteplase

Population

Patients with acute ischemic stroke within 6 hours of onset

Comparator

Placebo

Endpoint

Neurological improvement at 90 days

Key result: No significant benefit was found in the overall population, with higher rates of intracranial hemorrhage in the alteplase group.
2000

ATLANTIS Trial

n = 613 · JAMA

Tested

Intravenous alteplase

Population

Patients with acute ischemic stroke within 3 to 5 hours of onset

Comparator

Placebo

Endpoint

Excellent neurological recovery at 90 days (NIHSS score 0-1)

Key result: The trial failed to demonstrate significant benefit for alteplase administered between 3 and 5 hours after stroke onset.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis