91 girls with Turner syndrome (TS) with a mean chronological age (CA) and bone age (BA) of 10.3 ± 2.3 and 8.9 ± 1.9 years, respectively, were randomly assigned to subcutaneous treatment with recombinant human growth hormone (rhGH) alone (n = 47), 2.6 IU/m<sup>2</sup> body surface area daily or combination treatment (n = 44) with the same dose of rhGH and oxandrolone 0.1 mg/kg body weight orally, for the first 12 months of this study. During the 1 st year of therapy, there was a striking increase in height velocity (HV) in both groups, from 4.0 ± 0.8 to 6.3 ± 1.3 cm/year [HV standard (standards of untreated Turner patients) deviation score (SDS) for CA from 0.0 ± 0.7 to 2.9 ± 1.3] in the rhGH group and from 4.2 ± 1.2 to 8.5 ± 1.7 cm/year (HV SDS-CA from +0.3 ± 1.0 to+5.6 ± 1.6) in the combination group. The difference between the groups was statistically significant (p < 0.01). During the 2nd year of treatment, the rhGH dose was increased to 3.4 IU/m<sup>2</sup> daily for the rhGH-alone group, whereas in the combination treatment group the oxandrolone dose was reduced to 0.05 mg/kg daily. HV was maintained at significantly higher levels than those prior to treatment, at 5.3 ± 1.1 cm/year (HV SDS-CA:+2.1 ± 1.3) and 6.2 ± 1.5 cm/year (HV SDS-CA:+3.6 ± 1.4) in the rhGH-alone and the combination group, respectively (p < 0.001). After 2 years of treatment, the mean predictable adult height had increased by +3.6 ± 2.4 cm in the rhGH-alone group and by +5.3 ± 3.6 cm in the combination group. Both treatment regimens were very well tolerated although clitoral enlargement was seen in 15 of the 44 girls of the combination group who had been treated with 0.1 mg oxandrolone daily for the 1 st year. Impaired glucose tolerance was infrequently seen in girls of either group and the higher oxandrolone dose of the 1st year affected serum lipoprotein levels. Our data suggest that rhGH may improve adult height in girls with TS. This beneficial effect may be further increased by the addition of a low dose of oxandrolone (0.05 mg/kg/day).