How safe is safe? Is a drug safe, for instance, if it undergoes trials involving 20,000 people, and no serious adverse reactions are reported?

Certainly, that seems to be good enough for heart consultants who have embraced the new antistroke drug clopidogrel. In the US alone, specialists have prescribed it in the two years since its approval to 3 million patients who are at risk of heart attack or stroke, or who have undergone heart surgery.

Its success is pretty much down to the results of a clinical trial involving 20,000 patients which has given the drug a clean bill of health. This was music to the ears of consultants whose first choice was ticlopidine, another antiplatelet agent. But ticlopidine was thought to trigger the usually fatal condition thrombotic thrombocytopenic purpura in a small minority of cases.

Then, word started getting around that doctors were beginning to see just such a link with clopidogrel. With the backing of special grants, a team of researchers from Northwestern University in Chicago decided to investigate further. They looked at the records of 11 patients who developed thrombotic thromocytopenic purpura, and found that 10 of them had been taking clopidogrel at the time. One later died.

Unlike with ticlopidine, the condition started early in the treatment, usually within 14 days, tended to recur and required up to 30 plasma transfusions before the patient started to improve

(N Engl J Med 2000; 342: 1773-7).

Researchers point out that ticlopidine had a clean bill of health for seven years before the link with the thrombosis was discovered.

So, as we say, just how safe is safe. . . ?