Growth hormone replacement inhibits renal and hepatic 11β- hydroxysteroid dehydrogenases in ACTH-deficient patients

Abstract

OBJECTIVES: The commonest side-effects of GH replacement therapy relate to sodium retention but its mechanism is unclear. In rats, GH inhibits renal and hepatic 11β-hydroxysteroid dehydrogenase (11βHSD) activities. In man, this action might impair inactivation of cortisol to cortisone in the distal nephron thereby allowing cortisol to activate mineralocorticoid receptors. In this study, we examined the effects of GH replacement on cortisol metabolism. DESIGN: A randomized double-blind study comparing 6 months GH replacement with placebo in adults with GH deficiency due to pituitary or hypothalamic disease. PATIENTS: 6 men and 8 women received placebo and 8 men and 9 women received GH. MEASUREMENTS: Cortisol and its metabolites were measured in overnight urine samples and in a fasting morning plasma sample at baseline and at 6 months. RESULTS: Five patients withdrew from the study, 4 because of adverse effects of GH. Amongst those who completed the study, the effects of GH on cortisol metabolism differed between patients with and without ACTH deficiency. Amongst those receiving hydrocortisone replacement (n = 18), GH had no effect on plasma cortisol/cortisone ratio, or urine tetrahydrocortisols/tetrahydrocortisone ratio, but produced a substantial reduction in total urinary cortisol metabolites (from 1326 ± 191 to 777 ± 229 μg/mmol creatinine; P< 0.01) and an increase in urinary free cortisol/cortisone ratio (from 0.88 ± 0.10 to 2.57 ± 0.74; P<0.02). By contrast, GH had none of these effects in patients with normal ACTH secretion. There were no changes in blood pressure or plasma electrolytes, aldosterone, or renin activity with GH, and no changes in any variables with placebo. CONCLUSIONS: The increase in cortisol/cortisone ratio in urine but not in plasma, and the lack of effect on hepatic cortisol metabolites, suggests that GH inhibits both the conversion of cortisol to cortisone by renal 11β-HSD2 and the conversion of cortisone to cortisol by hepatic 11β- HSD1. It is unclear why this action of GH only affects patients with ACTH deficiency. The reduction in total cortisol metabolite excretion suggests that GH also affects bioavailability of hydrocortisone tablets. These observations suggest that glucocorticoid replacement therapy may need to be adjusted in hypopituitary patients who are commenced on GH replacement.