Background: Hormone administration to elderly individuals can increase lean body mass (LBM) and decrease fat, but interactive effects of growth hormone (GH) and sex corticosteroids and their influence on strength and endurance are unknown. Objective: The purpose of this study was to evaluate the effects of recombinant human GH or sex corticosteroids on body composition, strength, endurance, and adverse outcomes in aged persons. Design, setting, and participants: This was a 26âweek randomized, doubleâblind, placeboâcontrolled parallelgroup trial in healthy, ambulatory, communityâdwelling US women (n = 57) and men (n = 74) aged 65 to 88 years recruited between June 1992 and July 1998. Interventions: Participants were randomized to receive GH (starting dose, 30 ÎŒg/kg, reduced to 20 ÎŒg/kg, subcutaneously 3 times/wk) + sex corticosteroids (women: transdermal estradiol, 100 ÎŒg/day, plus oral medroxyprogesterone acetate, 10 mg/day, during the last 10 days of each 28âday cycle [HRT]; men: testosterone enanthate, biweekly intramuscular injections of 100 mg) (n = 35); GH + placebo sex corticosteroid (n = 30); sex corticosteroid + placebo GH (n = 35); or placebo GH + placebo sex corticosteroid (n = 31) in a 2 Ă 2 factorial design. Main outcome measures: Lean body mass, fat mass, muscle strength, maximum oxygen uptake (VO2 max) during treadmill test, and adverse effects. Results: In women, LBM increased by 0.4 kg with placebo, 1.2 kg with HRT (p = .09), 1.0 kg with GH (p = .001), and 2.1 kg with GH + HRT (p < .001). Fat mass decreased significantly in the GH and GH + HRT groups. In men, LBM increased by 0.1 kg with placebo, 1.4 kg with testosterone (p = .06), 3.1 kg with GH (p < .001), and 4.3 kg with GH + testosterone (p < .001). Fat mass decreased significantly with GH and GH + testosterone. Women's strength decreased in the placebo group and increased nonsignificantly with HRT (p = .09), GH (p = .29), and GH + HRT (p = .14). Men's strength also did not increase significantly, except for a marginally significant increase of 13.5 kg with GH + testosterone (p = .05). Women's VO2 max declined by 0.4 mL/min/kg in the placebo and HRT groups but increased with GH (p = .07) and GH + HRT (p = .06). Men's VO2 max declined by 1.2 mL/min/kg with placebo and by 0.4 mL/min/kg with testosterone (p = .49) but increased with GH (p = .11) and with GH + testosterone (p < .001). Changes in strength (r = .355; p < .001) and in VO2 max (r = .320; p = .002) were directly related to changes in LBM. Edema was significantly more common in women taking GH (39% versus 0%) and GH + HRT (38% versus 0%). Carpal tunnel symptoms were more common in men taking GH + testosterone (32% versus 0%) and arthralgias were more common in men taking GH (41% versus 0%). Diabetes or glucose intolerance occurred in 18 GHâtreated men versus 7 not receiving GH (p = .006). Conclusions: In this study, GH with or without sex corticosteroids in healthy, aged women and men increased LBM and decreased fat mass. Sex corticosteroid + GH increased muscle strength marginally and VO2 max in men, but women had no significant change in strength or cardiovascular endurance. Because adverse effects were frequent (importantly, diabetes and glucose intolerance), GH interventions in the elderly should be confined to controlled studies.