GH treatment of adult patients with GH deficiency (GHD) or healthy volunteers results in weight gain and fluid and sodium retention. In the present study we have challenged normal short children with dehydration and water load, and evaluated their water and sodium clearance, plasma renin activity (PRA) and aldosterone over 4 weeks of GH administration. Eleven prepubertal short normal children, aged 4-9 years, were the subjects of this study. Recombinant GH was administered daily at a dose of 2.5 units/m2. Dehydration and water load experiments were conducted before GH (day 0) and on days 3, 7 or 28 of GH. An initial 15-hour fast ended with a 3-hour urine collection. A tap-water load of 800 ml/m2 was given orally. Urinary volumes were followed hourly for 3 h, as were urinary and serum creatinine, Na, K, aldosterone, plasma osmolality and PRA. Before GH therapy the subjects excreted within 2 h a mean 42% and within 3 h a mean 65% of the load. After 3 days of GH therapy the same children retained water significantly and excreted only 22 (p < 0.02) and 45% (p < 0.05) of the load volume, respectively. Calculating the free water clearance revealed no effect of GH therapy, whereas the fractional excretion of sodium and potassium decreased significantly by day 3 of GH administration, along with an increase in PRA and serum aldosterone. All these changes normalized by days 7 and 28 of GH therapy. It is concluded that short-term administration of GH to short normal children results in a transitory mild retention of sodium and a secondary water retention, and suggests that the primary event leading to sodium retention during the early phase of GH therapy of short normal children is an inappropriate increase in PRA.