Seventy girls with Turner syndrome, 4 to 12 years of age, participated in a prospective, randomized study to determine the effects on growth of methionyl human growth hormone (met-hGH) or oxandrolone. Subjects were randomly assigned to receive either no treatment (control) or met-hGH (0.125 mg/kg three times per week), oxandrolone (0.125 mg/kg/day), or combination met-hGH plus oxandrolone. At the end of an initial period of 12 to 20 months, patients in the original control and oxandrolone groups were given combination met-hGH plus oxandrolone. At that time the dosage of oxandrolone was lowered to 0.0625 mg/kg/day. Sixty-five subjects have now completed the first 3 years of the study. Compared with the control growth rate for year 1 (3.8 cm/yr), significant increases in growth rate were seen in all 3 years of combination therapy (9.8, 7.4, and 6.1 cm/yr, respectively) and in the first 2 years of treatment with met-hGH alone (6.6, 5.4, and 4.6 cm/yr). When growth velocity was expressed as standard deviation for age in girls with Turner syndrome, significant increases relative to the control group for year 1 (-0.1 SD) were seen in all three years of both combination therapy and met-hGH alone (combination, +6.6, +4.3, +3.0 SD; met-hGH, +3.1, +2.0, +1.4 SD). After 3 years of treatment, predicted adult height by the method of Bayley-Pinneau increased 4.5 cm in the met-hGH group and 8.2 cm in the combination group.