Clinical trials of recombinant human GH therapy in Turner syndrome that began more than a decade ago show that GH accelerates the linear growth rate. Several studies indicate that final height is also improved, although the magnitude of the increase has been debated. The age at which feminization is induced could be an important factor in determining the patient's ultimate growth response. To test this, 60 patients from a large (n = 117), previously unreported, clinical trial of GH treatment were randomly assigned to begin conjugated estrogens at either 12 or 15 yr of age. The 60 patients were all less than 11 yr of age at entry (mean, 9.5 yr) and received 0.375 mg/kg x week of GH for nearly 6 yr on a daily or three times weekly regimen. Height gain was calculated by comparing the study patients' final or near final heights to their pretreatment projected heights as well as to those of a separate set of age-matched, historical control patients. Patients in whom estrogen treatment was delayed until age 15 yr gained an average of 8.4 +/- 4.3 cm over their projected height, whereas those starting estrogen at 12 yr gained only 5.1 +/- 3.6 cm, on the average (P < 0.01). Analysis of the interval data showed that growth was stimulated for approximately 2 yr after estrogen initiation, but then declined in association with bone age advancement. Patients who were older than 11 yr at entry (n = 57) all initiated estrogen 1 yr after beginning GH and showed a mean gain in adult height of 4.7 cm, similar to those given estrogen at age 12 yr. Multivariate analysis revealed that the number of years of GH therapy before estrogen treatment was a strong factor in predicting height gained, indicating that the timing of estrogen introduction is an important determinant of final height in this cohort of GH-treated patients with Turner syndrome matched for baseline age and other characteristics.