PDF: ESPS2
(1996)Objective
ALow-dose ASA versus dipyridamole versus their combination for secondary prevention of ischemic stroke.
Study Summary
• Both ASA and dipyridamole independently reduced the risk of recurrent stroke and stroke/death. The combination of ASA and dipyridamole was significantly more effective than either alone. Bleeding was more frequent with ASA.
Intervention
Randomized, double-blind trial of ASA 50 mg daily, modified-release dipyridamole 400 mg daily, their combination, or placebo. Follow-up: 2 years. 6,602 patients with prior stroke or TIA.
Study Design
Arms: Array
Outcome
• Stroke vs. placebo — ASA: 18% reduction (p=0.013), dipyridamole: 16% (p=0.039), combo: 37% (p<0.001)
• Stroke/death — ASA: 13% reduction (p=0.016), dipyridamole: 15% (p=0.015), combo: 24% (p<0.001)
• TIA — 36% reduction with combination (p<0.001)
• No significant effect on death alone
• Adverse events: Headache common with dipyridamole; bleeding (especially GI) more frequent with ASA
• Stroke/death — ASA: 13% reduction (p=0.016), dipyridamole: 15% (p=0.015), combo: 24% (p<0.001)
• TIA — 36% reduction with combination (p<0.001)
• No significant effect on death alone
• Adverse events: Headache common with dipyridamole; bleeding (especially GI) more frequent with ASA
Bottom Line
Both ASA and dipyridamole individually reduce stroke risk, but combination therapy provides superior protection with 37% risk reduction compared to placebo, demonstrating additive beneficial effects.
Major Points
- Large multicenter trial with 6,602 patients from 13 European countries
- 2×2 factorial design comparing ASA, dipyridamole, combination, and placebo
- ASA alone reduced stroke risk by 18.1% (p=0.013) compared to placebo
- Dipyridamole alone reduced stroke risk by 16.3% (p=0.039) compared to placebo
- Combination therapy reduced stroke risk by 37.0% (p<0.001) - significantly superior to either agent alone
- Combination therapy showed additive protective effects without significant statistical interaction
- Low-dose ASA (25 mg twice daily) was as effective as higher doses used in previous studies
Study Design
- Study Type
- Randomized, placebo-controlled, double-blind trial
- Randomization
- Yes
- Blinding
- Patients, investigators, and steering/monitoring committees were blinded to treatment assignment until study completion
- Sample Size
- 6602
- Follow-up
- 2 years
- Centers
- 59
- Countries
- Belgium, Finland, France, Germany, Ireland, Italy, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom
Primary Outcome
Definition: Stroke (fatal and non-fatal strokes, death from any cause allowed first event to be counted for 'survival' analysis but avoided patient being counted twice)
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 250/1649 (15.2%) | ASA: 206/1649 (12.5%), DP: 211/1654 (12.8%), Combination: 157/1650 (9.5%) | ASA: 0.819 (0.67-0.99), DP: 0.837 (0.69-1.02), Combination: 0.630 (0.52-0.76) (ASA: 0.67-0.99, DP: 0.69-1.02, Combination: 0.52-0.76) | ASA: 0.013, DP: 0.039, Combination: <0.001 |
Limitations & Criticisms
- Relatively high discontinuation rate due to adverse events
- Gastrointestinal side effects were significantly more common in all active treatment groups
- Bleeding risk increased with active treatments, particularly combination therapy
- Study was not powered to detect differences in mortality endpoints
- No formal sample size calculation reported in the paper
Citation
Journal of the Neurological Sciences 143 (1996) 1-13