In our article on agranulocytosis (July 1997) we argued that “routine monitoring...seems sensible for patients beginning treatment with an antithyroid drug” and suggested fortnightly monitoring of the full blood count during the first 3 months of treatment.1 These recommendations have been criticised by several endocrinologists. The substance of their comments is that agranulocytosis induced by antithyroid drugs is extremely rare in Britain, that when it occurs it develops very rapidly and that there is no evidence that monitoring would benefit patients. They also point out that monitoring would complicate and increase the cost of management and would be inconvenient both for patients and their medical carers.

Our recommendations were largely based on the results …