The low dose aspirin controversy solved at last?

though thyroid hormone values are normal, albeit usually in the upper part of the reference ranges.' Dr A B Davies and colleagues (21 September, p 773) have challenged this view by showing that the combination of extreme age and non-thyroidal illness may be associated with temporary TSH unresponsiveness to TRH whether atrial fibrillation is present or not. In this circumstance and in the absence of other evidence of hyperthyroidism, few clinicians, ourselves included, would want to prescribe antithyroid therapy, preferring to reassess thyroid state at a later date. The difficulty arises if TSH unresponsiveness to TRH persists despite apparent recovery from non-thyroidal illness. Age alone does not cause an absent TSH response to TRH.23 Furthermore, it may be inappropriate to have used standard reference ranges for free triiodothyronine and free thyroxine in the population studied. The decision to treat must ultimately depend on clinical judgment, but, given the presence of goitre, intermittent or sustained atrial fibrillation, and a persistently absent TSH response to TRH (cases 5 and 7), we believe that a trial of antithyroid therapy would be reasonable and, indeed, a firm recommendation in younger patients. In the light of the results from the Cardiff group our view of the value of the TRH test in unexplained atrial fibrillation will be modified to take account of the influence of extreme age and non-thyroidal illness. It would be unfortunate, however, if their findings in such a singular group of patients were to be overinterpreted and, as suggested by the authors, used as a blanket dismissal of the value of the TRH test (or basal TSH measured by sensitive immunoradiometric assay) as the basis for antithyroid treatment.