Publications

Abstract

OBJECTIVE: Early studies of acromegaly undertaken before the general availability of insulin-like growth factor I (IGF-I) assays have used arbitrary and varying growth hormone (GH) threshold levels for diagnosing and assessing outcome of treatment for this disease. We have undertaken a detailed study of GH secretion and its relationship to IGF-I levels to assess the usefulness of GH and IGF-I measurements in the assessment of acromegaly. PATIENTS: Thirty acromegalic subjects (12 untreated and 16 previously treated) and 30 age and sex-matched normal subjects were studied. MEASUREMENTS: Twenty-four-hour GH secretion was obtained from 20-minute sampling and serum IGF-I was measured. Comparisons were made of IGF-I, mean 24-hour GH concentration, and of the pulsatile and diurnal characteristics of GH secretion between the two groups. RESULTS: IGF-I levels in untreated acromegaly were elevated and clearly separated from the normal range. Mean 24-hour GH concentrations in untreated and treated acromegalic subjects with elevated IGF-I (> 40 nmol/l) were greater than (P < 0.01), and showed good separation from, those of normal subjects only after age-matching. From the 24-hour profiles, nadir GH levels in normal subjects fell below the level of detectability while those in untreated acromegalic subjects did not. Pulse amplitude (P < 0.01), ratio of pulsatile to non-pulsatile GH release and the night to daytime GH ratio (P < 0.05) were significantly reduced in acromegaly. In the six patients who attained normal IGF-I levels after surgery, pulsatile characteristics remained abnormal in four. Mean 24-hour GH was significant related (r = 0.57) to IGF-I. A random GH concentration > 2.5 micrograms/l (5 mIU/l) has a sensitivity of 77% and specificity of 95% in identifying acromegalic patients who have biochemically active disease (elevated IGF-I) after treatment. CONCLUSIONS: Patients with active acromegaly secrete more GH than age-matched normal controls. GH secretion in acromegaly is characterized by marked blunting of pulsatile secretion and, in contrast to normal subjects, the failure of GH to fall to undetectable levels at any time during the 24-hour day. IGF-I measurement is a more practical alternative in the diagnosis of acromegaly and in the assessment of therapeutic outcome. Since abnormalities of GH regulation may persist despite normalization of IGF-I, a distinction between remission and cure should be made. Detailed post-treatment evaluation of GH secretion is necessary to define the nature of underlying GH regulation and to evaluate the risk of disease recurrence.