Thyroid testing is increasingly used as a tool for identifying cases of thyroid disease in both primary and secondary care even in the absence of a strong clinical suspicion of disease. Patients with unsuspected thyroid disease are, therefore, likely to be identified, and among those identified are a few patients who have pituitary tumours or hypopituitarism. In the United Kingdom there is variation in which tests are offered for first line thyroid testing. In a 1994 survey of endocrinological testing in clinical biochemistry laboratories in the United Kingdom, Barth et al found that 30 different combinations of first line and second line profiles were being used.1 In a total of 186 replies 34% of laboratories reported offering testing for free thyroxine concentrations and thyroid stimulating hormone, 32% offered testing only for thyroid stimulating hormone, and 18% offered testing for total thyroxine concentration. The cases of the six patients described here highlight the fact that offering testing only for thyroid stimulating hormone (TSH) may be inappropriate.

The first five patients discussed had concentrations of free thyroxine hormone and thyroid stimulating hormone measured using a highly sensitive, third generation method (Amerlite hTSH, Kodak Clinical Diagnostics, Amersham). In the sixth patient these were measured by a two step, second generation method (Beckman Instruments, High Wycombe). Prolactin, luteinising hormone, and follicle stimulating hormone were measured using the Abbott IMX fluoroimmunoassay (Abbott Diagnostics, Maidenhead). Growth hormone was measured using the Nichols Diagnostics immunoassay (Nichols Institute Diagnostika, Bad-Nauheim, Germany). Testosterone was measured by radioimmunoassay. Anterior pituitary function was investigated in five patients using glucagon, thyrotrophin releasing hormone, and gonadotrophin releasing hormone challenge. Insulin was used instead of glucagon in one instance.