To some extent they were justified. Aspirin cannot claim to prevent primary strokes and has never been proven effective as a prophylaxis in low risk patients with no history of cardiovascular disorders (N E J Med, 1992; 327:175-81; BMJ, 1988; 296:313-6). Even as secondary prevention, for which there is some evidence, there is disagreement about the optimum dose (Lancet, 1991; 338:1345-9). One trial comparing treatment with aspirin, warfarin, or no treatment at all was revealing. In patients with a low risk of stroke there was no significant difference in life expectancy between the three groups. In high risk patients, life expectancy over the following 10 year period was 6.27 years for the aspirin group, 6.51 for warfarin users and 6.01 for those receiving no treatment at all a statistically significant difference but close enough to provide food for thought (JAMA, 1995; 274:1839-45).

With hindsight, the BMJ's response in the editorial which prefaced the ATC reports was telling: "aspirin seems as effective as any other single agent or combination of agents [our emphasis]." (BMJ, 1994; 308:71-3). In the end, that's not saying much since most other single agents or combination of agents produce devastating side effects particularly in low to medium risk patients (see box p2).

Since the ATC trial, the Cochran Database of Systematic Reviews has published on disc the first set of overviews from the Cochrane Stroke Review Group. Among these, four trials of anti platelet agents showed no significant advantage in rates for death, deep vein thrombosis or intracerebral hemorrhage when compared to other methods of treatment (Stroke Module, Cochrane database of systematic reviews, Updated 9 March 1995).

!APat Thomas