THE ORDERLINESS OF THE GROWTH-HORMONE (GH) RELEASE PROCESS AND THE MEAN MASS OF GH SECRETED PER BURST ARE HIGHLY CONSERVED IN INDIVIDUAL MEN ON SUCCESSIVE DAYS

Endocrine glands signal their remote target tissues via physiologicall y pulsatile release of regulatory molecules. A cardinal assumption of most pathophysiological experiments is that discrete attributes of pul satile hormone secretion are stable over successive untreated observat ion intervals; i.e. repeated measurements show serial within-subject r eproducibility. To test this hypothesis in the GH axis, we sampled blo od every 10 min for 48 h in 14 healthy men (age range, 29-77 yr; body mass index, 21-51 kg/m(2)). The 2 consecutive 24-h serum profiles were subjected to ultrasensitive GH chemiluminescence assay (sensitivity, 0.002 mu g/L) with a new dose-dependent variance model to estimate wit hin-assay precision. We then applied deconvolution analysis to estimat e the number, mass, amplitude, and duration of underlying GH secretory bursts as well as simultaneously calculate the apparent GH half-life and any concurrent basal hormone secretion. Test-retest consistency wa s assessed by the Pearson correlation coefficient, and differences wer e determined by paired nonparametric (Wilcoxon) testing. Comparing suc cessive 24-h profiles, no significant differences existed in any of th e foregoing secretion or half-life measures or in a novel estimate of the relative disorderliness of hormone release, namely approximate ent ropy, Correlation was minimal for secretory burst amplitude and half-d uration. In contrast, the calculated mean mass of GH secreted per burs t was highly conserved across sessions within subjects, with an r valu e of 0.932 (P < 10(-6)). This correlation equaled or exceeded that of mean and integrated serum GH concentrations on consecutive days (r = 0 .920; P = 0.00003). The calculated daily GH production rate was also s trongly reproduced (r = 0.784; P = 0.0009). Moreover, the within-subje ct GH half-life and GH secretory burst frequency estimates were well c orrelated on successive days (P = 0.034-0.004; r = 0.568-0.711). Appro ximate entropy values were consistent at r = 0.837 (P = 0.0019). In ad dition, basal GH secretion rates correlated at r = 0.622 (P = 0.0176). We conclude that homeostatic control mechanisms within the GH-insulin -like growth factor I axis strongly preserve the day to day mean mass of GH secreted per burst and the serial orderliness of the GH release process in individual healthy men across a wide span of ages and body compositions.