Growth Hormone Effects on Adult Height in Idiopathic Short Stature

Source: Leschek EW, Rose SR, Yanovski JA, et al. Effect of growth hormone treatment on adult height in peripubertal children with idiopathic short stature: a randomized, doubleblind, placebo-controlled trial. J Clin Endocrinol Metab. 2004;89:3140–3148.The authors, acting on behalf of the National Institute of Child Health and Human Development and Eli Lilly Company Growth Hormone Collaborative Group, performed a randomized, double-blind, placebo-controlled trial to determine the effect of growth hormone (GH) on final adult height when used in peripubertal children with idiopathic short stature. Subjects included 68 children (53 males) 9–16 years of age with marked idiopathic short stature, sexual maturation not greater than Tanner stage 2, and a bone age of 13 years or less in boys and 11 years or less in girls. Marked idiopathic short stature was defined by a height SD score (SDS) or predicted adult height SDS of −2.5 or less within the 12 months before study initiation, except before 1993 when a cutoff of −2.25 was used (6 such patients between −2.25 and −2.5 were included). Children were excluded if they had a chronic illness, known genetic syndrome, or had ever received growth or sex hormonal therapy, or were currently receiving medications with the potential to affect growth. Children were randomly assigned to receive either GH or placebo subcutaneously 3 times per week. The study drug was continued until growth rate, measured over 1 year, decreased to less than 1.5 cm/year, indicating near-adult height.Adult height measurements were available for 33 children after a mean treatment duration of 4.4 years. Mean height velocity was significantly greater in the GH group compared with the placebo group. Subjects in the GH group achieved a mean height 3.7 cm greater than the placebo-controlled group (P=.02). The bone age progression and progression of puberty were similar between controls and GH-treated children.Because the study was terminated before many subjects achieved final adult height, 2 modified intent-to-treat analyses were performed for children who had been treated for at least 6 months. This analysis suggested that the GH-treated children would have a significantly greater increase in final adult height than the placebo group. There were no significant differences between the 2 groups with regard to safety and adverse events. The authors conclude that GH treatment can increase adult height in peripubertal children with marked idiopathic short stature.Financial Disclosure: Dr. Varma has disclosed that he has no relationships relevant to this commentary.A Lawson-Wilkins Pediatric Endocrine Society survey revealed that 94% of pediatric endocrinologists admit that they would recommend GH therapy for children without GH deficiency whose height falls below 2.25 standard deviations below the mean.1 The above study is the first long-term, randomized, double-blind, placebo-control trial to determine the effect of GH on adult height when administered to peripubertal children. Because the study was terminated by the Data Safety Monitoring Board for slow accrual, the study had mean duration of 4.4 years for only about 50% of the enrolled children. Data on the remaining enrolled children had to be extrapolated from the first 6 months of treatment. In administering GH 3 times a week, the study protocol differed from current treatment recommendations to administer GH 6–7 times a week. Using this protocol, children receiving GH appeared to achieve some increase in final adult height without any premature advancement of bone age. It is difficult to know whether greater increases in adult height could have been achieved by administering GH 6–7 days per week. A metaanalysis of 4 controlled studies reported a 5–6 cm difference in adult height between GH treatment and control groups without any serious side effects related to GH treatment.2Most pediatricians will care for children with idiopathic short stature. While the results of this study may offer hope to these children, we must also recognize that GH may not be the solution to social and physical problems encountered by some children with short stature. Certainly, treatment with GH is not for children whose height is at or greater than the 5th percentile, and it should be used with caution in children with idiopathic short stature. It is essential that other causes of growth failure be ruled out before GH therapy is started. Recently, the Food and Drug Administration has approved the use of GH in children with idiopathic short stature (see AAP Grand Rounds, August 2004;12:14–15).3Can GH therapy increase the final adult height of children with idiopathic short stature? The answer from this study appears to be yes, an average of 1.5 inches. But to stop there is to stop short of asking a more important question: should children with idiopathic short stature be treated with an expensive medication that requires multiple injections per week for several years? These are not children with a demonstrable medical disorder, but children who happen to be considerably shorter than most of their peers. The literature suggests that most short children have normal psychosocial functioning (see AAP Grand Rounds, April 2004;11:39–40),4 and that extremes of stature have minimal impact on peer perceptions of social behavior, friendship or acceptance.5 There is no data to suggest that treatment with GH improves psychosocial functioning, and while treatment with GH will potentially add a few inches to a child’s final adult height, that person will remain shorter than most of his or her peers. The cost of treatment for a child with short stature can exceed $35,000 per inch gained.2 Add to that the “costs” to the child of regular injections over several years to administer the medication, and it seems clear that the decision to treat a child with short stature must be made very carefully and with a complete consideration of the costs and benefits of treating what amounts to a problem of social perception. In addition, although 2 recent studies have concluded that GH does not increase the risk of cancer (See AAP Grand Rounds, November 2002;8:49–50),6,7 there is still some biological basis for this association since GH treatment increases insulin-like growth factor, a substance that has been implicated in cancer risk.8 We would do well to remember that treatment with GH is not the only option for children with short stature. It may be far preferable to coach parents to help their short children accept and master their shortness.