In 1987 a multicentre trial of recombinant human growth hormone (GH) was started in girls with Turner syndrome. Fifty-four patients were randomly assigned to receive GH, 8 IU/m2 3 times/week (group 1), or 4 IU/m2 6 times/week (group 2). In addition, the 35 patients older than 12 years received ethinyloestradiol, 100 ng/kg body weight/day, and after 2 years GH therapy was increased to 6 IU/m2 6 times/week. Recombinant human GH treatment was stopped when the height increment during the previous 6 months of treatment was less than 0.5 cm. Treatment has so far been stopped in 48 patients: treatment was stopped early in 2 patients due to lack of motivation, 1 patient died suddenly and the treatment protocol was completed in 45 patients. The last height measurement obtained, which was considered as (near) final height, was 152.3 +/- 5.3 cm (mean +/- SD) in these patients, which is higher (p < 0.001) than the adult height of 147.0 +/- 6.3 cm reported in 63 untreated adult Dutch patients with Turner syndrome. No differences in outcome were found between the two dose regimens.