Background: In clinical practice, height velocity and rate of bone maturation during each follow-up interval are among the parameters used to modify glucocorticoid treatment in children with 21-hydroxylase deficiency. There is controversy on the relative importance of treatment and disease variables for short-term growth and final height.
Patients and methods: Short- and long-term growth data of 87 patients (n = 35 salt-wasting form; n = 52 simple virilizing form) were analysed retrospectively by stepwise multiple regression.
Results: Height SDS for chronological age (HtSDSCA) was elevated prior to treatment in boys (m = 3.22, p < 0.0001) and decreased when compared to HtSDSCA > or = 4 years after the start of treatment in both boys (m = -0.64, p < 0.0001) and girls (m = -1.24, p < 0.0023). When the simultaneous effects of hydrocortisone dose, salt-wasting status, degree of hormonal control, and patient sex on short-term growth between follow-up visits were analysed, only a minor effect of hydrocortisone dose (partial coefficient of determination [Pr2] = 0.04, p < 0.011) and salt-wasting status (Pr2 = 0.03, p < 0.024) in the prepubertal group, and a small effect of the treatment quality rating in the pubertal group (Pr2 = 0.07, p < 0.0042) on height velocity SDS for chronological age (HVSDSCA) were noted. Height velocity SDS for bone age (HVSDSBA) was influenced to a similar degree by treatment quality (Pr2 = 0.098, p < 0.0017) in the prepubertal, and by hydrocortisone dose (Pr2 = 0.063, p < 0.021) in the pubertal group. Mean daily hydrocortisone doses used in normally growing patients ranged between 17.9 and 21.8 mg/m2/d if analysed separately for sex and salt-wasting status. 34 patients had reached final height which with the exception of one boy was below the population mean and ranged from 82.3% to 100.1% of target height in the total group. 41.2% of the patients had adult short stature (HtSDS < -2). Treatment with synthetic glucocorticoids for more than 1 year started before the age of 1.5 years was associated with the most severely compromised final height SDS (m = -3.73, p < 0.029).
Conclusions: The proportion of the short-term height velocity SDS that can be explained by the tested treatment and disease variables is low (< or = 9.8%). Conclusions drawn from observed changes in height velocity during single short follow-up intervals on treatment modalities must therefore be viewed with caution. In the long run, however, use of daily hydrocortisone doses > 25 mg/m2/day and of synthetic glucocorticoids started early in the course of the disease does not only lead to a transient deceleration of height velocity in growing children with 21-hydroxylase deficiency, but carries a definite risk for decreased final height.