1. We studied β-cell function in 40 hypopituitary adults and in 36 matched control subjects. Hypopituitary patients were studied again at 1, 3 and 6 months during a double-blind placebo-controlled trial of growth replacement lasting for 6 months. Biosynthetic human growth hormone was given subcutaneously in a daily dose of 0.02–0.05 i.u./kg at bed time. Fasting insulin, intact proinsulin and 32–33 split proinsulin were measured by two-site immunoradiometric assays.
2. Hypopituitary patients were aged 19–67 years and had a body mass index of 27.7 (18.0–41.1) kg/m2. They were receiving replacement thyroxine, adrenal steroids and sex hormones and they were growth hormone deficient. Control subjects were matched for age, sex and body mass index. Hypopituitary patients with normal glucose tolerance and with impaired glucose tolerance were compared separately with subgroups of control subjects matched for age and body mass index.
3. Twenty-six hypopituitary patients had normal glucose tolerance and 14 had impaired glucose tolerance. All control subjects had normal glucose tolerance by World Health Organization criteria. Patients with impaired glucose tolerance were significantly older than those with normal glucose tolerance (P < 0.03). Hypopituitary patients with normal glucose tolerance compared with normal control subjects had a significantly lower fasting plasma glucose concentration (P < 0.01), a lower fasting insulin concentration (P < 0.006), a lower insulin—glucose ratio (P < 0.02) and a lower percentage of insulin to total insulin-like molecules [hypopituitary patients, 90% (81–96%); control subjects, 93% (78–97%); P < 0.02]. Hypopituitary patients with impaired glucose tolerance had similar glucose and insulin concentrations and insulin—glucose ratios as matched control subjects. They had significantly higher split proinsulin concentrations than control subjects [hypopituitary patients, 5.5 (1.2–20.7) pmol/l; control subjects, 1.6 (0.4–23.7) pmol/l; P = 0.02]. They also had a lower percentage of insulin to total insulin-like molecules [hypopituitary patients, 87% (78–92%); control subjects, 93% (78–98%); P < 0.001] and a higher percentage of proinsulin-like molecules [hypopituitary patients, 13.5% (7.7–21.8%); control subjects, 7.0% (2.4–22.0%); P = 0.0005]. In the hypopituitary group as a whole, insulin, proinsulin and split proinsulin correlated significantly with body weight (r = 0.71, 0.41 and 0.59 respectively), body mass index (r = 0.62, 0.64 and 0.60 respectively), waist circumference (r = 0.59, 0.45 and 0.60 respectively) and body fat mass (r = 0.72, 0.50 and 0.61 respectively).
4. After 6 months of growth hormone treatment, there was no change in fasting plasma glucose on either growth hormone or placebo. Fasting plasma insulin and proinsulin and split proinsulin increased significantly on growth hormone therapy but not on placebo. Changes in patients with initial normal and impaired glucose tolerance were considered separately. The changes in fasting plasma glucose, insulin, proinsulin and split proinsulin on growth hormone therapy were more marked in those with initial normal glucose tolerance than in those with impaired glucose tolerance.
5. In conclusion, these data demonstrate that glucose-intolerant hypopituitary patients have decreased basal insulin secretion and hyperproinsulinaemia. Growth hormone treatment caused increases in fasting plasma insulin and intact and split proinsulin concentrations.