ARAIS
(2023)Objective
To assess the efficacy of argatroban plus alteplase compared with alteplase alone for acute ischemic stroke
Study Summary
• Primary outcome occurred in 63.8% vs 64.9% (RR 0.98, P=0.78)
• No significant safety concerns with similar bleeding rates between groups
Intervention
Argatroban (100 μg/kg bolus over 3-5 minutes followed by 1.0 μg/kg/min infusion for 48 hours) within 1 hour after alteplase vs alteplase alone
Inclusion Criteria
Adults 18-80 years with AIS, NIHSS score >6, within 4.5 hours of symptom onset, premorbid mRS ≤1
Study Design
Arms: Argatroban plus alteplase vs Alteplase alone
Patients per Arm: 364 vs 396 (full analysis set)
Outcome
• Secondary outcomes showed no differences in favorable outcome, early neurologic improvement, or vascular events
• Symptomatic ICH rates similar: 2.1% vs 1.8%
Bottom Line
Among patients with acute ischemic stroke, treatment with argatroban plus intravenous alteplase compared with alteplase alone did not result in a significantly greater likelihood of excellent functional outcome at 90 days.
Major Points
- Multicenter, open-label, blinded endpoint RCT conducted at 50 hospitals in China
- 817 patients randomized with median age 65 years, median NIHSS score 9, 29.1% women
- 93% trial completion rate (760/817 patients)
- Primary outcome (mRS 0-1 at 90 days) occurred in 63.8% with argatroban plus alteplase vs 64.9% with alteplase alone (RD -1.0%, 95% CI -8.1% to 6.1%, P=0.78)
- No significant differences in any secondary outcomes including favorable outcome, early neurologic improvement/deterioration, or vascular events
- Similar safety profile between groups with symptomatic ICH rates of 2.1% vs 1.8%
- Lower proportion of patients with large artery occlusion (20.8%) compared to previous studies may explain neutral results
- Only 23.2% of patients met target APTT at 2 hours, taking approximately 5 hours to reach target
- No evidence of effect modification across prespecified subgroups
Study Design
- Study Type
- Multicenter, open-label, blinded endpoint randomized clinical trial
- Randomization
- Yes
- Blinding
- Open-label for treatment assignment; blinded endpoint assessment for primary outcome (mRS) and adverse events by assessors unaware of treatment assignment
- Sample Size
- 817
- Follow-up
- 90 days
- Centers
- 50
- Countries
- China
Primary Outcome
Definition: Excellent functional outcome, defined as modified Rankin Scale score of 0 to 1 at 90 days
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 238/367 (64.9%) | 210/329 (63.8%) | - (RR 0.98 (0.88-1.10); RD -1.0% (-8.1% to 6.1%)) | 0.78 |
Limitations & Criticisms
- Sample size not fully achieved in argatroban group (n=364 vs planned n=367) due to higher dropout rate, potentially reducing statistical power
- Open-label design could introduce bias, though primary endpoint was assessed by blinded evaluators
- Lower proportion of patients with large artery occlusion (20.8%) compared to previous positive studies (51.1%-100%), which may explain negative results
- Only 23.2% of patients met target APTT at 2 hours; took approximately 5 hours to reach target APTT, suggesting suboptimal argatroban dosing
- Large difference between assumed (21%) and observed (64%) rates of excellent functional outcome suggests power calculation may have been based on overly pessimistic assumptions
- Endovascular thrombectomy used infrequently, limiting generalizability to comprehensive stroke centers
- Conducted only in Chinese population, limiting generalizability to other ethnic groups
- Higher dropout rate in argatroban group (21 withdrew consent) vs control group (12 withdrew consent) may introduce attrition bias
- Per-protocol analysis excluded many patients (39 from argatroban group, 29 from alteplase group) due to protocol deviations
- Low-dose argatroban regimen (1 μg/kg/min) compared to high-dose studies (3 μg/kg/min) may have limited efficacy
Citation
JAMA. 2023;329(8):640-650. doi:10.1001/jama.2023.0550