LOW BASAL AND PERSISTENT PULSATILE GROWTH-HORMONE SECRETION ARE REVEALED IN NORMAL AND HYPOSOMATOTROPIC MEN STUDIED WITH A NEW ULTRASENSITIVE CHEMILUMINESCENCE ASSAY

Serum GH concentrations assessed by conventional RIA and immunoradiometric assay (IRMA) are often undetectable in healthy fed and awake humans. Serum GH concentrations are also low in obese, hypothyroid, middle-aged, and older individuals. Accordingly, to better investigate the pathophysiology of GH release in relatively hyposomatotropic states, we compared a new and putatively ultrasensitive GH chemiluminescence (CL) assay with a conventional IRMA. We report that the sensitivity of the CL assay is approximately 0.005 mu g/L, with a median intraassay coefficient of variation over more than 3 logarithms of serum GH concentrations of 5.2% (range, 3.6-8.3%). In each of 2 normal young men and 4 categories of hyposomatotropic adults studied (2 obese, 2 hypothyroid, 2 middle-aged, and 3 older men), we observed detectable serum GH concentrations by CL assay of all samples collected at 10-min intervals for 24 h. In the same sera analyzed by IRMA, 22-98% of the daytime sample values fell below IRMA sensitivity (0.10 mu g/L). Within the IRMA-detectable range, the correlation between the 2 assays ranged between r = 0.893 to 0.989 (P < 0.001 in each subject). In the CL assay, serum GH concentrations declined to 0.018-0.030 mu g/L, but were never undetectable during the awake fed state. Deconvolution analysis of 11 GH series assayed by CL disclosed basal GH release and superimposed GH secretory bursts. In the low GH series, there was a mean frequency of 12 +/- 1.4 secretory events/24 h, a mass of GH secreted per burst of 1.0 +/- 0.35 mu g/L, a secretory burst half-duration of 21 +/- 3.3 min, and an apparent endogenous GH half-life of 14 +/- 0.85 min. Compared to the GH IRMA, the CL assay detected approximately 50% more GH secretory events, and GH secretory burst frequency was increased significantly at night vs. the daytime. Cosinor analysis revealed significant 24-h serum GH concentration rhythms in both assays, but an earlier mean acrophase (time of maximum) and lower mesor (mean values) were estimated in the CL assay than the IRMA. We conclude that within the IRMA-detectable range, the correlation between serum GH concentrations measured by CL and IRMA techniques is high. However, unlike the IRMA, the GH CL assay can detect serum GH concentrations consistently in the awake and fed state not only in young individuals but also in subjects with obesity, hypothyroidism, and middle and older age. In both normal men and those with relative hyposomatotropism, the ultrasensitive CL assay revealed for the first time a low rate of basal GH secretion that was mixed with pulsatile and nyctohemeral variations. GH secretory burst frequency increased at night. These results indicate that undetectable serum GH concentrations in conventional RIAs or IRMAs conceal persistent basal and pulsatile GH secretion that can be unmasked by a more sensitive CL assay. This technology should offer new insights into the regulation of the somatotropic axis.