Fasting or severe caloric restriction in the human or experimental animal suppresses serum LH and sex steroid concentrations. In healthy men undergoing prolonged (5-day) nutrient deprivation, the daily LH secretion rate, the mass of LH secreted per burst, and the serum testosterone concentration fall markedly, with no decrease in responsiveness to a single bolus of GnRH. Here we test the hypothesis that the hypogonadotropic hypoandrogenemia accompanying fasting reflects decreased endogenous GnRH release. To this end, six healthy young men were studied on a fed day and during two 83-h fasting sessions with concurrent saline or pulsatile GnRH administration (100 ng/kg, iv, every 90 min for 24 h) followed by a single bolus of 10μ g GnRH, iv, to evaluate pituitary responsiveness. We employed a highly sensitive LH immunoradiometric assay, which correlates well with an in vitro Leydig cell bioassay, and deconvolution analysis to calculate in vivo LH secretory burst frequency, amplitude, duration, mass, and LH half-life. Fasting resulted in 30–50% declines in serum total and free testosterone and LH concentrations, and a 3-fold decrease in the calculated 24-h LH secretion rate (fed, 42 ± 12; fasting, 14 ± 1.9 U/L distribution volume·day; mean ± sem; P < 0.05, by ANOVA). Reduced LH secretion was accounted for by dual mechanisms, viz. a fall in both the apparent number of computer-resolved LH secretory bursts per 24 h (fed, 16 ± 1.1; fasting, 10 ± 1.2; P < 0.01) and the mass of LH secreted per burst (fed, 2.5 ± 0.5; fasting, 1.5 ± 0.1 U/L; P < 0.05). Fasting also decreased the mean value of the 24-h (nyctohemeral) rhythm in serum LH concentrations and reduced the approximate entropy (disorderliness) of LH release. Exogenous pulsatile GnRH injections prevented both the reduction in the calculated daily LH secretion rate (fed, 42 ± 12; fasting plus GnRH, 64 ± 16 IU/L; P = NS) and the decline in serum testosterone concentrations (fed, 556 ± 71 ng/dL; fasting, 391 ± 41; fasting plus GnRH, 859 ± 65). Pulsatile GnRH treatment also restored the nyctohemeral mesor of serum LH concentrations and the approximate entropy value to baseline. Administration of a submaximal dose of exogenous GnRH (10 μg, iv) at the end of the fasting interval revealed statistically identical LH release in the three study groups, suggesting that pituitary responsiveness to GnRH was unchanged in this paradigm.

In summary, a pulsatile iv GnRH infusion in young men averts completely the fasting-induced decline in LH secretory burst mass/amplitude and frequency, reinstates serum total and free testosterone concentrations, and restores the mesor of LH’s nyctohemeral rhythmicity and the approximate entropy of LH release. Rescue of hypogonadism by pulsatile GnRH stimuli supports the thesis that nutrient withdrawal decreases the output of the human hypothalamic GnRH burst generator.