Current issues of ACP Journal Club are published in Annals of Internal Medicine


Low-dose aspirin for cerebrovascular ischemic events reduced the risk for stroke or death

ACP J Club. 1992 Mar-April;116:41. doi:10.7326/ACPJC-1992-116-2-041

Related Content in this Issue
• Companion Abstract and Commentary: Low-dose aspirin after transient ischemic attack prevented vascular events as well as medium-dose aspirin and had fewer adverse effects

Related Content in the Archives
Antiplatelet agents reduce risks for death, stroke, myocardial infarction, deep venous thrombosis, and arterial occlusion

Source Citation

The SALT Collaborative Group. Swedish Aspirin Low-dose Trial (SALT) of 75 mg aspirin as secondary prophylaxis after cerebrovascular ischaemic events. Lancet. 1991 Nov 30;338:1345-9.



To examine the efficacy of a 75-mg daily dose of aspirin for the prevention of stroke or death after transient ischemic attacks or minor stroke.


Randomized, placebo-controlled, double-blind trial.


16 clinical centers in Sweden between December 1984 and January 1989.


Patients were eligible if they were between 50 and 79 years of age and had had a transient ischemic attack, minor ischemic stroke, or retinal artery occlusion within the previous 3 months. Exclusion criteria included a new or recurrent qualifying event during a 3- to 5-week run-in period with anticoagulant therapy or aspirin; a potential cardiac source of emboli; previous or planned carotid surgery; other causes for symptoms; other disorders affecting prognosis or compliance; and contraindications to aspirin.


Patients were allocated to receive either aspirin, 75 mg/d (n = 676), or placebo (n = 684), as 1 tablet taken ≥ 30 minutes before breakfast. Follow-up assessments were made every 4 months, and no patients were permanently lost to follow-up.

Main outcome measures

First stroke or death from any cause. All analyses were intention-to-treat.

Main results

The mean time between the qualifying event and randomization was 57 ± 24 (SD) days for aspirin and 58 ± 25 days for placebo. 17% of patients on aspirin and 23% of patients on placebo discontinued treatment for reasons other than an outcome event. Mean length of follow-up was 32 ± 15 months for all patients. Mean duration of treatment was 31 ± 17 months for aspirin and 28 ± 17 months for placebo. Fewer patients who received aspirin had strokes or died then did patients who received placebo (20% vs 25%, P = 0.02). {This absolute benefit increase of 5% means that 22 patients would need to be treated with aspirin (rather than aspirin) to prevent 1 additional stroke or death, 95% CI 11 to 75%, the relative benefit increase was 18%, CI 0.5% to 33%.}* Results were similar for all first events (stroke, myocardial infarction or death) (27% vs 22%, P = 0.03). Bleeding episodes were higher, however (3% vs 7%, P = 0.001). {This absolute risk increase (ARI) of 4% means that for every 25 patients who received aspirin (NNH) (rather than placebo), 1 additional patient would have bleeding, CI 15 to 59; the relative risk increase (RRI) was 125%, CI 39% to 267%.}* The results were similar for severe bleeding (1% vs 3%, P = 0.03): ARI 2%; NNH 61, CI 30 to 876; RRI 125%, CI 5% to 382%}.* Results were similar for both men and women.


Aspirin, 75 mg/d, lowered the risk for stroke or death after a cerebrovascular ischemic event and raised the risk for bleeding.

Sources of funding: Swedish National Association against Heart and Chest Diseases; Medical Products Agency; Swedish Medical Research Council; Leo AB.

Address for article reprint: Dr. B.Norrving, Department of Neurology, University Hospital, S-221 85 Lund, Sweden.

*Numbers calculated from data in article.


Although it is perhaps not as good as getting "more for less," the latest Dutch and Swedish trials of aspirin confirm that lower doses of aspirin are effective in reducing the risk for major fatal and nonfatal vascular events after initial minor cerebrovascular events and that the risk for adverse events is less with the lower doses. In addition, both studies found equal benefits for men and women from treatment with very low doses of aspirin.

The close timing of publication of these two studies was auspicious. The Dutch trial has an important limitation: There was no placebo group and both doses of aspirin were far below the dose level of 1300 mg per day that was originally shown to be effective (1). Thus, one possible interpretation is that neither dose was effective. The Swedish trial, however, does have a placebo control group and documents the effectiveness of 75 mg of aspirin per day after initial treatment with aspirin or anticoagulants.

The Swedish trial provides an interesting ethical twist. A placebo-controlled trial of aspirin for patients with mild qualifying cerebrovascular events could not be conducted today after previous trials showed the efficacy of aspirin (2). However, the Swedish study was in its final phase when the efficacy of aspirin became clearly defined, and the demonstration by this study that lower doses of aspirin than previously tested are also effective will likely lead to many more patients enjoying aspirin's benefits at a lower risk for adverse effects. This is not to say that low-dose aspirin is benign. Both studies show about a 3% risk for severe gastrointestinal bleeding, and the Swedish study provides a comparison risk of 1.3% for severe bleeding among patients receiving placebo (for a relative risk of about 2). It is likely that this complication rate cannot be lowered without losing the effectiveness of aspirin: Presumably both the bleeding and the prevention of ischemic vascular events are the result of the same process—inhibition of platelet cyclo-oxygenase. There may be a small increase in the risk for hemorrhagic stroke, although this is more than offset by the risk reduction for ischemic stroke.

Do the lower doses of aspirin give a benefit comparable to larger doses? The relative risk reduction for vascular events in trials comparing aspirin in the dose range of 900 mg to 1500 mg daily with placebo from the Antiplatelet Trialists' Collaboration meta-analysis was 22% (2), which is consistent with the 18% relative risk reduction seen in the Swedish trial. This is not surprising if the mechanism of aspirin's effect is mediated through cyclo-oxygenase inhibition: Complete inhibition occurs with a daily aspirin dose as low as 30 mg.

Although these studies began well before the demonstration of benefit for carotid endarterectomy for high-grade internal carotid artery stenosis, it is regrettable that little or no information was collected about the effect of aspirin for various degrees and locations of vascular stenosis. It is important that patients with minor vascular events in the anterior circulation of the brain receive further investigation to determine the need for endarterectomy. Despite the use of aspirin by patients with high-grade carotid stenosis, their risk for subsequent events remains unacceptable and can be reduced substantially by surgery.

Finally, judging by the stocks of aspirin in pharmacies, it will be difficult for patients to reduce their aspirin dose without splitting standard aspirin pills into small bits or paying the exorbitant price of children's aspirin. It is high time for pharmaceutical manufacturers to provide low-dose aspirin in economical packaging.

Preston C. Calvert, MD
The Neurology Center, P.A.Falls Church, Virginia, USA
R. Brian Haynes, MD
McMaster UniversityHamilton, Ontario, Canada

Preston C. Calvert, MD
The Neurology Center, P.A.
Falls Church, Virginia, USA

R. Brian Haynes, MD
McMaster University
Hamilton, Ontario, Canada


1. The Canadian Cooperative Study Group. A randomized trial of aspirin and sulfinpyrazone in threatened stroke. N Engl J Med. 1978;299:53-9.

2. Antiplatelet Trialists' Collaboration. Secondary prevention of vascular disease by prolonged anti-platelet therapy. BMJ. 1988;296:320-31.