Low birth weight (BW) is associated with an increased risk of the metabolic syndrome and cardiovascular disease in adulthood. Programmed hypersecretion of glucocorticoids or reduced secretion of GH has been postulated as mechanisms for this effect. However, other variables such as premature birth may confound the association of birth size with later endocrine function. To separate the effect of BW from other variables, we examined basal and dynamic function of the hypothalamus-pituitary-adrenal axis and GH-IGF axis in twin siblings with differing BW. Twenty pairs of same-sex healthy adult twins underwent measurement of serum cortisol before and after low-dose (1 mug) Synacthen stimulation, and plasma GH during glucose suppression and exercise stimulation. In paired statistical analysis, the lower BW twins had significantly lower morning serum cortisol than their heavier BW siblings (mean intrapair difference 60 nmol/liter, 95% confidence interval 5-114, P < 0.03) but no difference in peak cortisol level after ACTH. Lower BW was associated with a trend to lower baseline plasma GH and a significantly lower peak GH concentration after exercise (difference 7.6 mU/liter, 95% confidence interval 1.7-13.5, P = 0.01). Intrapair differences in basal and stimulated cortisol and basal GH also correlated significantly with the intrapair difference in BW, demonstrating a dose-response effect of BW on hypothalamus-pituitary-adrenal axis function and basal GH secretion. In a twin model that isolates BW from other confounding variables, our data suggest that low BW programs individuals for reduced GH secretion and reduced basal cortisol secretion but preservation of a cortisol secretory response to ACTH.