Supporting authors' views would be unwise

Editor—Gordon R B Skinner and colleagues advocated that a diagnosis of hypothyroidism in patients with suggestive symptoms should not be excluded on the basis of “normal” hormone concentrations.1 Furthermore, they proposed that an incremental trial of thyroxine for three months is not unreasonable in these patients.

The consensus is that patients with subclinical hypothyroidism in whom serum thyroid stimulating hormone concentrations are consistently above the usually quoted upper limit of the reference range of 5 mU/l should be treated with thyroxine; this is particularly so if the patients have antibodies to thyroid peroxidase, a history of treatment of thyrotoxicosis, or a goitre.2 Two double blind trials in patients with subclinical hypothyroidism have shown that, after treatment with thyroxine, target organ function may improve and there may be a greater sense of wellbeing in some patients, though by no means all.3 The most cogent reason for treatment, however, is the knowledge that a considerable proportion of patients will develop overt hypothyroidism in future years,4 and it makes sense for the disorder to be “nipped in …