Critically ill patients have high basal growth hormone levels with attenuated oscillatory activity associated with low levels of insulin-like growth factor-I

Abstract

Objective: The aim was to study the relationship between growth hormone (GH) and insulin-like growth factor-I (IGF-I) in critically ill patients.

Design: Case-control study of critically ill patients admitted to the intensive care unit was carried out.

Patients: Six critically ill patients (51-78 years) who required ventilation and parenteral nutrition and six age, weight, height, and sex-matched healthy adults were studied.

Measurements: The patients and controls were studied for two 24-hour periods; the patients before and after starting parenteral feeding, and the controls during a 36-hour fast and when taking meals equivalent in calories and protein to the patients' parenteral feed. Serum GH was measured at 20-minute intervals and analysed by a pulse detection algorithm (Pulsar) and Fourier transformation. IGF-I was measured at 0, 12, and 24 hours.

Results: Patients had low serum IGF-I levels compared with controls, whether fasted or fed, despite having mean GH levels similar to fasted controls. For fasted patients vs fasted controls the mean (+/- 1 SD) GH levels were 4.5 +/- 2.0 vs 4.0 +/- 2.4 mU/l respectively, and IGF-I levels at the end of the fast were 0.17 +/- 0.11 vs 0.78 +/- 0.29 U/ml (P = 0.003). Patients showed elevated baseline GH levels compared with controls when fasted and during parenteral feeding (patients vs controls fasted 3.1 +/- 1.9 vs 0.8 +/- 0.5 mU/l, P = 0.01; patients vs controls fed 4.2 +/- 4.5 vs 0.5 +/- 0.04 mU/l, P = 0.028). Fourier transformation confirmed oscillatory GH levels in the controls, fasted or fed, but this activity was attenuated in the patients. Parenteral feeding had no effect on the GH profiles or IGF-I levels of patients, but controls showed greater mean GH levels during their fast than when fed.

Conclusions: We have demonstrated that critically ill patients have low IGF-I levels associated with augmented baseline GH levels which show reduced oscillatory activity. The results would be compatible with the hypothesis that there is an adaptive change in critically ill patients away from the indirect effects of GH (stimulation of IGF-I production and anabolism) and toward the direct effects (lipolysis and insulin antagonism) which increase the availability of energy substrates. The pattern of GH levels seen in our patients may be important in this adaptation.