Somatostatin infusion withdrawal: studies in the acute and recovery phase of anorexia nervosa, and in obesity

Abstract

OBJECTIVE: Changes in GH/IGF-I axis activity occur in both anorexia nervosa (AN) and obesity (OB). A GH hypersecretory state with very low plasma IGF-I levels is present in AN, whereas in morbid OB, GH secretion is dull and plasma IGF-I levels are generally preserved. Endogenous GHRH activity in AN and OB has never been directly studied, although indirect evidence would indicate that GHRH function is altered in either condition, possibly enhanced and reduced respectively. Somatostatin (SS) infusion withdrawal (SSIW) is followed by a rebound rise of plasma GH in animals and humans, an event which, allegedly, is mediated by endogenous GHRH release. METHODS: In the present study, 28 young women, eight with active AN (A-AN), six with AN in the recovery phase (R-AN), eight with morbid OB, and six healthy age-matched normal weight subjects (NW), were studied. All subjects underwent, on different occasions, the following two tests: (i) acute GHRH injection (1 microg/kg, i.v.); (ii) infusion of SS (9 microg/kg per h i.v. over 60 min), with blood samples drawn prior to and at different intervals after drug injections. Plasma GH levels were measured at each time interval in all sessions, and, in addition, baseline plasma estradiol, free triiodothyronine, TSH, IGF-I and insulin were measured at -30 min. RESULTS: Baseline plasma GH concentrations were significantly higher in A-AN than in NW (4.7+/-0.7 vs 2.1+/-0.6 microg/l, P<0.01). Baseline GH levels in R-AN were also higher than in NW, but the difference did not reach statistical significance (5.6+/-1.7 microg/l, not significant (NS)). Baseline plasma GH concentrations were significantly lower in OB than in NW (0.3+/-0.1 microg/l, P<0.01). GHRH-stimulated GH release was significantly higher in A-AN than in NW (mean change in area under the curve (DeltaAUC) 1904.9+/-626.1 vs 613.9+/-75.9 microg/l per min, P<0.01), whereas no statistically significant difference was present between R-AN and NW (mean DeltaAUC 638.2+/-293.0 microg/l per min, NS); in OB, GHRH failed to evoke a plasma GH rise (mean DeltaAUC 239.8+/-89.9 microg/l per min vs A-AN, R-AN, and NW, P<0.01). SS infusion markedly reduced plasma GH concentrations in both A-AN and R-AN and, to a lesser extent, in NW, but failed to do so in OB. In A-AN, SSIW was followed by a plasma GH rise markedly higher than that present in NW (mean DeltaAUC 193.0+/-42.3 vs 60.1+/-11.4 microg/l per min, P<0.01), whereas in R-AN the GH response after SSIW was nearly superimposable on that registered in NW (mean DeltaAUC 72.9+/-22.8 microg/l per min, NS). There were no changes in plasma GH levels after SSIW in OB (mean DeltaAUC 22.8+/-9.7 microg/l per min). In all groups, DeltaAUCs of the GH response to GHRH and after SSIW were highly positively correlated (r=0.7, P<0.01). CONCLUSIONS: These data support the view that a high endogenous GHRH tone, which subsides in the recovery phase of the disease, is present in AN, whereas GHRH hypofunction, possibly associated with pituitary impairment, might indicate OB.