The 'hook effect' on serum prolactin estimation in a patient with macroprolactinoma.

Large quantities of antigen in an immunoassay system impair antigen-antibody binding, resulting in low antigen determination. This is called the 'high dose hook effect'. We report this phenomenon in a patient with a large macroprolactinoma. In this patient, the correct estimate of serum prolactin (PRL) was obtained only after appropriate dilution of serum. We suggest that in order to avoid the high dose hook effect, the serum PRL be estimated in appropriate dilution in all patients with large pituitary tumours. This is particularly important when the clinical suspicion of high PRL is strong, as in women with amenorrhoea-galactorrhoea and men with long standing hypogonadism.

Thyroid stimulating hormone receptor antibody in thyroid diseases.

Thyroid stimulating hormone receptor antibody (TRA) was estimated as a measure of TSH binding inhibitory immunoglobulin (TBII) in 48 persons. These included (i) 14 controls; (ii) 23 patients with Graves' disease who were tested for TRA within 3 months of commencing treatment with carbimazole of which 13 were studied serially; (iii) 5 patients with toxic nodular goitre; (iv) 4 with euthyroid exophthalmos; and (v) 2 neonates of thyrotoxic mothers. TRA was measured with an RIA system, while total thyroxine (T4), free thyroxine concentration (FTC) and TSH were also estimated along with TRA. All controls showed undetectable TRA levels; 87 per cent of patients with Graves' disease were TRA positive within 3 months of starting carbimazole therapy. In the serial study, 5 patients with Graves' disease who had undetectable TRA initially remained so while on treatment. Seven out of the remaining 8 patients showed a decline of TRA levels to normal over 3 to 18 months. This decline coincided with clinical and biochemical recovery.