Anticoagulants versus antiplatelet agents for acute ischaemic stroke

Antiplatelet agents produce a small, but worthwhile benefit in long-term functional outcome and survival, and have become standard treatment for acute ischaemic stroke. Anticoagulants are often used as an alternative treatment, despite evidence that they are ineffective in producing long-term benefits. We wanted to review trials which have directly compared anticoagulants and antiplatelet agents, to assess whether any anticoagulant regimen offers net advantages over antiplatelet agents, overall or in some particular category of patients (e.g. patients with atrial fibrillation).

a) To assess the effectiveness of anticoagulants compared with antiplatelet agents in acute ischaemic stroke b) To assess whether the addition of anticoagulants to antiplatelet agents offers any net advantage over antiplatelet agents alone.

We searched the Cochrane Stroke Group Trials Register, the Cochrane Controlled Trials Register (Central/CCTR), the trials register held by the Antithrombotic Therapy Trialists' Collaboration, MEDLINE (1966-2000), and EMBASE (1980-2000). All searches were performed during April and May 2001.

Truly unconfounded, randomised-controlled trials comparing anticoagulants with antiplatelet agents, or anticoagulants and antiplatelet agents with antiplatelet agents alone, given within 14 days of onset of presumed or confirmed ischaemic stroke.

Both reviewers independently selected trials for inclusion in the review, assessed trial quality and extracted data.

A total of 16,558 patients from four trials contributed to the analyses. The methodological quality was high in all four trials. The anticoagulants tested were unfractionated heparin (UFH) and low molecular-weight heparin. Aspirin was used as control in all trials. Overall, there was no evidence that anticoagulants were superior to aspirin in reducing 'death or dependency' at long-term follow-up (odds ratio [OR] 1.07, 95% confidence interval [95% CI] 0.98-1.15). Compared with aspirin, anticoagulants were associated with a small but significant increase in the number of deaths at the end of follow-up (OR 1.10, 95% CI 1.01-1.29), equivalent to 20 more deaths (95% CI 0-30) per 1000 patients treated; a significant increased risk of symptomatic intracranial haemorrhage (OR 2.35, 95% CI 1.49-3.46); and a non-significant increased risk of 'any recurrent stroke' during treatment (OR 1.20, 95% CI 0.99-1.46). These neutral or adverse effects outweighed a small, but significant effect on symptomatic deep vein thrombosis (OR 1.20, 95% CI 0.07-0.58), equivalent to 10 fewer (95% CI 0-30) DVTs by 14 days per 1000 patients treated with anticoagulants instead of aspirin. Subgroup analysis could not identify any type, dose, or route of administration of anticoagulants associated with net benefit, or any benefit in patients with atrial fibrillation. Overall, the combination of UFH and aspirin did not appear to be associated with a net advantage over aspirin alone. A subgroup analysis showed that, compared with aspirin, the combination of low-dose UFH and aspirin was associated with a marginally significant reduced risk of 'any recurrent stroke' (OR 0.75, 95% CI 0.56-1.03) and a marginally significant reduced risk of death at 14 days (OR 0.84, 95% CI 0.69-1.01), and with no clear adverse effect on death at end of follow-up (OR 0.98, 95% CI 0.85-1.12).

Anticoagulants offered no net advantages over antiplatelet agents in acute ischaemic stroke. The combination of low-dose UFH and aspirin appeared in a subgroup analysis to be associated with net benefits compared with aspirin alone, and this merits further research.