Although many studies have discerned higher serum GH concentrations in women than in men, based on measurements of single random blood samples or integrated 24-h means, the neuroendocrine mechanisms that underlie such gender differences have not been defined. Such mechanisms might entail an increase in GH-secretory burst frequency, amplitude, or duration, heightened interpulse basal GH release, or a prolonged half-life of GH. These mechanisms can be distinguished by deconvolution analysis of appropriate GH time series. Earlier studies employed RIA or IRMA with sensitivities of 0.1-0.5 microgram/L, which result in frequently undetectable serum GH concentrations. To address these limitations, we undertook blood sampling at 10-min intervals for 24 h and applied a high-sensitivity immunofluorometric assay of GH (sensitivity 0.0115 microgram/L). Multiparameter deconvolution analysis was used to estimate specific features of GH secretion, while simultaneously calculating the half-life of endogenous GH. Eleven men and 11 premenopausal women from the same community were studied. Discrete peak detection by Cluster was employed as a complementary half-life-independent technique to assign variations in serum GH into pulsatile and basal fractions over 24 h. Cluster revealed significantly higher mean serum GH concentrations over 24 h in women (0.78 +/- 0.08 microgram/L) compared with in men (0.27 +/- 0.03 microgram/L, P < 0.00005). Women exhibited significantly higher maximal serum GH concentration peak values than men (2.08 +/- 0.34 microgram/L in women, 0.67 +/- 0.11 microgram/L in men, P = 0.0008), which could be, in turn, attributed to a significantly increased incremental serum GH peak amplitude (1.85 +/- 0.33 microgram/L in women vs. 0.60 +/- 0.10 microgram/L in men, P = 0.0021) and a longer peak duration (114 +/- 8 min in women, 86 +/- 4 min in men, P = 0.008). The mean area under the serum GH concentration peak was significantly (3-fold) higher in women than in men (98 +/- 17 micrograms/L.min in women, 34 +/- 8 micrograms/L.min in men, P = 0.0046). Serum GH peak frequency was similar in women (9.7 +/- 0.8 peak/24 h) and men (10.7 +/- 1.1 peak/24 h, P = NS). The mechanisms underlying the increase in serum GH concentration pulse amplitude, duration, and area were investigated further by deconvolution analysis. Deconvolution analysis disclosed equivalent secretory pulse frequencies in women and men (13 +/- 0.9 bursts/day in women, 10.5 +/- 1.3 bursts/day in men,P = NS), and statistically indistinguishable mean interburst intervals of 106 +/- 8 min in women and 150 +/- 26 min in men (P = NS). GH-secretory burst mass was significantly higher in women by approximately 2.4-fold (P = 0.0013) compared with in men, which was attributed to a greater burst amplitude. Only low levels of basal GH release were inferred in women (5%) and men (9%), which did not differ significantly between genders. Moreover, the calculated half-life of endogenous GH was no different in women compared with in men: 15.8 +/- 0.7 min vs. 17.1 +/- 0.8 min, respectively (P = NS). The calculated daily secretion rate was 3-fold higher in women (47 +/- 4.8 micrograms/L.24 h) than in men (15 +/- 1.8 micrograms/L.24 h) (P < 0.001). In summary, discrete peak-detection analysis of serum GH concentration profiles collected at 10-min intervals over 24 h in men and premenopausal women discloses significantly different mean serum GH concentrations that are accounted for by higher maximal and incremental serum GH peak amplitudes. Deconvolution analysis demonstrated that the mechanism supporting the amplitude-specific difference in women was an augmentation of the GH-secretory burst mass caused by a higher GH-secretory burst amplitude. These gender differences were highly specific because the frequency of detectable GH-secretory bursts, the calculated endogenous half-life, and the estimated basal GH release were no different in women than in men.