Relationships between insulin-like growth factor-1 levels and growth hormone concentrations during diurnal profiles and following oral glucose in acromegaly.

OBJECTIVE

The aim of this study was to refine the biochemical definition of disease activity in acromegaly by comparing serum growth hormone (GH) measurements during a 10-hour day profile with serum GH values during an oral glucose tolerance test.

DESIGN

Using plasma insulin-like growth factor-1 (IGF-1) levels as a measure of disease activity, serum GH data from a day profile and from an oral glucose tolerance test were compared.

PATIENTS

Thirty-five acromegalic patients were studied, 13 of whom had serum GH measured during a day profile and 22 during an oral glucose tolerance test. In addition, basal plasma IGF-1 levels were estimated in all acromegalic patients, and in 24 normal subjects.

MEASUREMENTS

Following acid-ethanol extraction of the plasma samples, IGF-1 levels were measured by radioimmunoassay using a polyclonal antibody. In a day profile, six to eight blood samples for serum GH estimation were taken at hourly intervals during the day; during an oral glucose tolerance test samples for serum GH estimation were taken in the fasting state and every 30 minutes for 2 hours and measured by a two-site IRMA for GH.

RESULTS

Ninety-four per cent of acromegalic patients with raised plasma IGF-1 levels had serum GH concentrations > 10 mU/l whilst 98% of acromegalic patients with plasma IGF-1 levels in the normal range had serum GH concentrations < 6 mU/l. A highly significant positive correlation was found between the mean serum GH concentrations (r = 0.67), the minimum serum GH concentration (r = 0.65) and the area under the GH curve (r = 0.66) estimated during an oral glucose tolerance test and plasma IGF-1 concentrations. The relations between identical indices of serum GH concentration measured during a day profile and plasma IGF-1 levels, although significant, show a less powerful correlation. The relation between serum GH and plasma IGF-1 levels describes a curvilinear model, plasma IGF-1 levels exhibiting a plateau at serum GH concentrations > 40 mU/l but maintaining a linear relationship with serum GH levels < 20 mU/l.

CONCLUSIONS

A highly significant correlation exists between plasma IGF-1 levels and various parameters of serum GH levels in acromegalic patients. Hormonal assessment of disease activity in acromegaly is more accurately reflected by the serum GH concentration during an oral glucose tolerance test rather than by the serum GH level during a day profile. Normalization of plasma IGF-1 levels is rarely achieved unless the mean serum GH level is reduced to < 6 mU/l.