Since 1985 molecular biology techniques have made possible the synthetic synthesis of pure human growth hormone in potentially unlimited amounts. With this increased availability, its use in patients other than children with growth hormone deficiency has been associated with clinical and ethical questions. This statement presents an analysis of the current status of the use of human growth hormone in children.
Since the first description of the effectiveness of a crude human pituitary extract in accelerating the growth of a patient with hypopituitarism,1 thousands of patients have received human growth hormone (GH) therapy. This extract was made available to investigators by the National Institutes of Health, Bethesda, MD, to treat patients considered to be GH-deficient according to the strict criteria of below-normal growth velocity for age, decreased bone age, and lack of response to two GH stimulation tests. Supplies were limited, but several conclusions emerged from these therapeutic trials:
GH therapy increased growth in GH-deficient individuals,2
growth monitored during a 5-year period shows that rates of upper and lower body segment growth have been equal.7
Various lots of extract were found to differ in potency, and contamination with infectious agents was a problem. Several of the extract recipients developed Creutzfeldt-Jakob disease through prior contamination of the pituitary extract, and its use was discontinued in April 1985.8,9
In early 1985, GH was produced by recombinant DNA technology. Suddenly, clinicians found themselves with an almost inexhaustible, though costly, supply of human GH. This dramatic change in hormone availability has raised a number of diagnostic, therapeutic, and ethical questions concerning the rational, safe, and cost-effective prescribing of human GH.
Recombinant Human GH Products
Naturally occurring GH is a polypeptide containing 191 amino acids. The biosynthetic process involves a chemical synthesis of the DNA fragment encoding the first 24 amino acids. The remaining amino acid residues are obtained by making complementary DNA copies of messenger RNA prepared from human pituitary cells. The entire DNA sequence is introduced into a bacterium, Escherichia coli, which is then able to synthesize human GH.
Clinical trials using synthetic human GH were begun in the early 1980s. In 1985, recombinant human GH became available commercially. The products currently available in the United States are somatrem (Protropin, Genentech, Inc, South San Francisco, CA) and somatropin (Nutropin, Genentech, Inc, South San Francisco, CA and Humatrope, Eli Lilly & Co, Indianapolis, IN). They differ in that somatrem contains an additional methionine group that is added to the 24-amino acid initiation sequence.
Optimal dosing strategies have not been developed fully for any product. The currently recommended starting dose is .1 mg/kg (0.26 IU/kg) administered either intramuscularly or subcutaneously three times per week.10 Most pediatric endocrinologists prescribe GH at a dose of .18 to .30 mg/kg per week given in either six or seven injections subcutaneously. In general, patients who respond will respond to the initial dosing regimen. Dosage escalation strategies vary. While increased growth may result from a dosage increase, especially in those children with GH resistance (eg, children with chronic renal insufficiency), the increase in growth response is not proportional to the increase in dose. The Lawson Wilkins Pediatric Endocrine Society has recently published their guidelines on the use of growth hormone in children with short stature.11
Problems in the Diagnosis of GH Deficiency
Before the advent of the radioimmunoassay for determination of GH concentrations, clinicians took advantage of the fact that GH is an insulin counterregulatory hormone. The diagnosis of hypopituitarism was made based on the serum glucose level response to intravenously administered insulin.12,13 With the introduction of the radioimmunoassay, the determination of serum GH concentrations became part of the endocrinologic evaluation of short stature.14
Increasingly sensitive radioimmunoassays for GH have revealed previously unappreciated subtleties in GH secretion. Because fasting serum GH levels do not differ between GH-deficient and non-GH-deficient patients, other physiologic and pharmacologic tests have been developed in an effort to identify GH-deficient children.12 Sleep and exercise are used as two of the physiologic releasers. Many of the pharmacologic tests involve the use of neuromodulators, such as levodopa, or agents such as arginine, insulin, or propranolol, which are known to stimulate the release of GH directly or indirectly. While the details of these tests are not pertinent to this statement, classic severe GH deficiency could be diagnosed if the peak stimulated value on two tests was 10 μg/L or lower in association with a delayed bone age and slow growth rate.
The diagnosis of partial GH deficiency is more controversial, and there is no generally accepted definition. Moreover, this diagnosis should not be confused with GH resistance, which represents a problem with responsiveness rather than production. Testing for GH production is costly, and both fees and assays vary among laboratories. Moreover, a number of physicians believe that because testing may be traumatic and costly and because the diagnosis based on serum GH concentration determinations is uncertain, such testing should not be a prerequisite for human GH therapy.15-17 As an alternative, they advocate making the decision to treat purely on clinical grounds.18
Goals for GH Administration to Children With Short Stature
Much of the controversy surrounding the use of human GH is related to the absence of well-defined goals for therapy, the paucity of data on the long-term response to therapy, and screening efforts to identify candidates for therapy. Inherent in the prescription of human GH therapy is the desire to make children grow; but how much, over how long a period of time, and with what end point? Some clinicians maintain that recombinant human GH therapy should be reserved for those patients with “classical” severe GH deficiency and other disorders for which efficacy and at least short-term risk-benefit analysis has been performed. In the group of patients with “classical” severe GH deficiency, the goal is to augment growth so that at the time of epiphyseal closure, a normal or normally expected adult height is attained.19 In patients with chronic renal insufficiency, the goal of GH treatment is to attempt to maintain age-appropriate growth so that with the reestablishment of normal GH responsiveness after transplantation, patients might attain a final adult height that is more consistent with their genetic potential.20
In contrast, other clinicians argue that short stature regardless of cause may constitute both physical and psychosocial handicaps. For these reasons, they believe that human GH should be made available to all children with short stature with the hope that the rate of growth will be increased and that full adult growth potential will be realized.18,20-22 However, there is accumulating evidence to show that such a goal may be unattainable.23,24
Unfortunately, just as it is difficult to define GH deficiency biochemically, in the era of recombinant human GH therapy, the medical community has no universally accepted standard definition for short stature. Clearly, both sex and age must be considered. Some physicians label patients short for age and sex if they are below a certain percentile on a standard longitudinal height chart, whereas others believe that it is actually height velocity for age and sex that is the more appropriate indicator.
Thus, latitude exists in identifying patients who might benefit from recombinant human GH therapy. Some increment in growth velocity has been demonstrated in most patients who have received human GH.25-27 However, it is not clear how many of these patients benefit psychologically or that increased adult height is achievable.28
In addition to debate over proper criteria for therapy, controversy surrounds screening efforts to identify candidates. School-based screening for children of short stature has raised concerns about overidentification of candidates and stigmatization of short children. Those concerns are particularly serious when companies that stand to profit from the marketing or manufacture of recombinant human GH are involved in the screening.
Labeled Indications for Recombinant Human GH Therapy
All recombinant human GH products are currently labeled for use in “children who have growth failure due to an inadequate secretion of normal endogenous growth hormone.”29 Recently, the Nutropin brand of somatropin has also been labeled for use in children with chronic renal failure before transplantation.20,30 In the case of classic GH deficiency, diagnosis is not a problem. However, in the absence of generally accepted criteria for diagnosing “inadequate secretion,” partial GH deficiency, or “GH dysregulation,” the indications for these products have become extremely subjective.
The findings of a recent American Academy of Pediatrics Periodic Survey of Fellows on the use of GH in clinical practice revealed that provocative testing was the most frequently named criterion used to support the use of GH; the next most frequent criterion was the degree of short stature or depression of growth velocity. The overwhelming majority of respondents also indicated that GH should be prescribed only by pediatric endocrinologists and that the goal of therapy is to increase current and adult stature.31
Off-Label Uses of Recombinant Human GH Therapy
Indications for the off-label use of human GH are not well defined, and there are many causes of short stature other than GH deficiency (Table1) that bear consideration; these conditions are not a reason to consider growth hormone therapy.32 Interest in the use of human GH in a variety of disorders (Table 2) has been stimulated by both the success of human GH therapy in GH-deficient individuals of short stature and the ready availability of synthetic GH products. The key issue then becomes whether GH treatment increases height among children with these disorders regardless of the patient's GH status. In some instances, the answer is yes, especially in disorders such as chronic renal insufficiency and Turner's syndrome in which a resistance to physiologic concentrations of GH has been documented.20,33 In many other instances, the use of GH has been justified on the grounds that persons with short stature (defined as more than 2 SDs below the mean for age and sex) experience stigma in an affluent society. These children are often teased in school about their short stature; moreover, empiric evidence indicates that numerous social benefits are linked to tall stature. In some children, short stature may be part of an acquired or inherited disorder. For these children, growth augmentation is viewed as an avenue to normalcy. Despite these concerns and the fairly extensive use of recombinant human GH in these patient groups, no objective current data demonstrate the psychosocial benefits of hormonal therapy in this group of children and few physiologic data demonstrate an effect on final adult height. The above considerations have led some to question whether research on the use of human GH to attempt to increase the final adult height of non-GH-deficient children is warranted.
Risks Associated With Recombinant Human GH Therapy
Short-term treatment with recombinant human GH has been associated with few, if any, side effects. The long-term risks are unknown. The available GH preparations result in some degree of antibody formation, but the overall level is 10% or lower. The antibodies formed seem to have no clinical effects and do not form immune complexes.43
Risks that have been attributed to human GH therapy include those associated with the physiological and psychological trauma related to years of regular injections. Certainly, this is one of the major stressors reported by diabetic patients related to their medical treatment. In addition, the repeated injections may actually serve to enhance the negative self-image and stigmatization that already exist for many of these children.44
Additional concern exists about metabolic effects that have been observed in GH-treated patients.45 Children may become unusually lean and inappropriately muscular due to increased cellular metabolism.46
Some authors have expressed the theoretical concern that GH therapy might also be associated with an increased risk of malignancy. It causes an increase in the synthesis of insulin-like growth factor I by the liver, and it has been suggested that insulin-like growth factor I may be associated with a greater risk of cancer or a poorer prognosis if cancer develops. Reports of a possible increased incidence of leukemia, lymphoma, and tumors in patients receiving GH along with chemotherapy and radiation therapy continue to be investigated. In addition, 23 patients treated with human GH or insulin-like growth factor I have developed benign intracranial hypertension (pseudotumor cerebri) first manifested by papilledema. Signs and symptoms resolved with a cessation of therapy but recurred promptly in three patients when they were re-treated.47
Proper Uses of GH Therapy
In light of the above discussion, which children are appropriate candidates for GH therapy? Certain dimensions of the clinical context make this an ethically complex question. First, the proper diagnostic criteria for GH deficiency are currently being disputed.22,48 Although the criteria for classic GH deficiency are widely accepted, there is disagreement over whether these criteria encompass the full scope of GH deficiency. To some extent, any resolution of this debate depends on how one defines “sufficient” and “insufficient” concentrations of GH.49,50 Moreover, some children who are not GH-deficient, including girls with Turner's syndrome and children with idiopathic short stature, respond to GH therapy with accelerated growth velocity.33-35 Currently available data indicate that final height (versus predicted height) is improved with GH treatment in patients with Turner's syndrome.33
Pediatricians must address directly whether GH therapy is acceptable for any children who do not fulfill the “classic” description of GH deficiency, and, if so, for which ones. This requires analysis of and agreement on the appropriate goals of GH therapy.
A second important clinical consideration is the present lack of an established risk-benefit ratio for GH therapy. This therapy is undeniably effective in increasing the final adult height of children with “classic” GH deficiency when compared with estimates made on the basis of growth velocity before therapy.22 However, benefits for other children are as yet uncertain. Although GH therapy seems to accelerate their growth, it is not clear that such children have a greater final adult height.51 It is also unclear whether GH therapy reduces the psychosocial problems that very short children may experience.52-54 Indeed, there is evidence that GH therapy exacerbates these problems in some children owing to unrealistic expectations concerning the therapeutic outcome and enhanced feelings that something is “wrong” with them.55,56
There is also the question of how to define treatment “success.” Short stature is a characteristic that must be defined relative to the general population in which people will always be of different heights. Thus, even if GH therapy were available to and effective in all “short” stature children, a population of short children will still exist—they will simply be a few inches taller than those in the former population.56,57
Even though the risks associated with human GH therapy are largely unknown, the burdens of undergoing treatment for non-GH-deficient children are plain. Ideally, children should receive GH injections daily and at a minimum three times a week.58 Treatment usually continues for 4 to 5 years and costs between $10 000 and $50 000 annually, depending on the child's body weight.59,60
Is Short Stature a “Disease” or “Disability”?
Being tall is indisputably viewed as a benefit in our culture. At least parents would wish their children to be what they perceive to be in a normal range of height. Considerable evidence demonstrates the career and other social advantages of being tall, including higher income, academic achievement, self-esteem, and social status.61-67 Accordingly, it could be argued that GH therapy is warranted as treatment for the disease or disability of short stature that impairs “normal social function,” similar perhaps to drug therapy for attention-deficit disorder or to orthodontia.22,68
The concepts of normality and abnormality are notoriously difficult to define; inevitably, they incorporate many sociocultural variables. The line between the two categories may be drawn at any number of points, depending on how much of a deviation is deemed a severe enough obstacle to functioning. If short stature is indeed an appropriate target for medical intervention, at what point is a child “short enough?” Moreover, in this context it is arguably society that needs “treatment,” not the people who are now at a disadvantage. Discrimination against short people is as unjustified as race, sex, and other forms of discrimination now condemned by our society. It would be better to eradicate the bias against short individuals than to attempt to eradicate the condition of being short, which, in any event, is doomed to failure. In this view, the appropriate “treatment” for short stature is removal of the physical and social barriers to normal functioning that people with this characteristic now face.56,69
On the other hand, the prior view arguably is unreasonable and unfair to the individuals who must bear the burdens of discrimination. Moreover, it is not easy to dismiss the profound and understandable distress some children experience because they are short. Such children undeniably are disadvantaged by their condition. The relevant issue, however, is whether GH therapy is an appropriate response to their problems. In extreme cases, it could be, because physical barriers can exclude very short individuals from basic activities of daily living, such as driving and food preparation. In these cases, GH could be justified to avoid serious dysfunction.49 A more defensible alternative in other cases may be to assist children and their parents in achieving pride and fulfillment on their own terms, rather than those society has dictated to them. Counseling and self-help organizations also may be helpful.
Social Justice Considerations
The economic implications of widespread GH therapy are significant. The direct medical costs entailed to provide GH therapy for 4 years to children below the present first percentile of height for age were recently estimated to be $1.5 to $3 billion annually.59 In contrast, the cost of providing GH therapy to children with classic GH deficiency is approximately $135 million.22 The price of GH may decrease in the near future, after synthetic GH loses its current orphan drug status and competition increases among the producers (New York Times Magazine. 1991;47:14–17, 28–29). The potential magnitude of this reduction is unknown.
In light of the likely expense of GH therapy, the question of insurance coverage must be addressed. Currently, GH therapy for children with “classic” GH deficiency is covered by Medicaid and some private insurers. The manufacturer covers the costs of GH for some other affected children. Coverage for children who are not “classically” GH-deficient varies and is difficult to ascertain. Some children participating in GH therapy research receive therapy at no cost. In addition, certain insurance companies apparently reimburse for such therapy, with the remainder paid for by the parents themselves.70
The availability of GH therapy raises serious social justice questions. Some children who are not classically GH-deficient now receive GH therapy. These children are likely to come from the more financially well-off sectors of society. They and their parents have access to the most advanced health care available through extensive health insurance coverage or their own financial resources. Consequently, the children who now obtain the economic and social benefits of growing taller are those who already have numerous advantages over the children who now lack access to GH therapy.68,71 Moreover, there is evidence that treatment is now concentrated in white males, with females and African-American males underrepresented among the treatment population.72 If differential access continues, GH therapy will serve to intensify the social and economic disadvantages faced by members of less well-off groups.
If future data establish the effectiveness of therapy for increasing the final height of children who do not have classic GH deficiency, the social justice implications of the current inequality in access could be significantly exacerbated. Society will face the difficult choice of whether to provide GH therapy at great cost to all children whose final height may be increased or to permit the present patchwork reimbursement approach to continue with all its inequities. Moreover, even a decision to cover treatment of all children who could be made taller could have serious health repercussions for disadvantaged children. Health care resources are limited; thus, the decision to guarantee access to one therapy may result in the denial of other health benefits to children in need.59 With so many children presently lacking access to the most basic health care, it would be ethically inappropriate to expend scarce monetary resources to provide GH therapy to anyone who is not either classically GH-deficient or truly GH-resistant.
Conflict of Interest
Screening for candidates and prescription of GH therapy has provoked conflict-of-interest concerns when companies that manufacture or market therapeutic agents are involved in screening or rewarding physicians for GH use. Because controversy surrounding GH therapy forces pediatricians to exercise considerable discretion in evaluating indications for therapy and advising as to its risks and benefits, special care must be taken to avoid financial relationships that either compromise or appear to compromise the physician's commitment to serving the patient's best interests.
CONCLUSIONS AND RECOMMENDATIONS
In light of previous considerations, the American Academy of Pediatrics offers the following recommendations:
Therapy with GH is medically and ethically acceptable for children (1) with classic GH deficiency, (2) children with chronic renal insufficiency who are awaiting kidney transplantation, (3) girls with Turner's syndrome, and (4) children whose extreme short stature keeps them from participating in basic activities of daily living and who have a condition for which the efficacy of GH therapy has been demonstrated. The diagnosis of GH deficiency should be made in consultation with a pediatric endocrinologist. Therapy for children with GH deficiency and for those with extreme short stature should be supervised by the consulting pediatric endocrinologist. In children with chronic renal insufficiency, GH therapy should be supervised by a pediatric nephrologist with expertise in GH therapy or by a pediatric nephrologist in conjunction with a pediatric endocrinologist as part of the comprehensive management of this chronic disease.
Numerous considerations argue against widespread administration of GH therapy to other short children. First, the therapy's risk-benefit ratio in this population is not established. There could be unknown long-term risks, and the treatment could result in either no increase or only an insignificant increase in final adult height. Research being conducted in this population will eventually determine the efficacy of GH therapy for this group. Even if the clinical data show a positive risk-benefit ratio, however, the benefits of GH therapy will inevitably remain somewhat elusive. Individual children may escape the stigma of being very short, but a group of very short children will always exist. On a broader scale, the best “therapy” for these children would be a campaign against the current prejudice against short people instead of an implicit medical reinforcement of such prejudice. Last, unrestricted administration of GH therapy would add to the burdens already borne by low-income and other disadvantaged children in our society. Either such children would increasingly be among the very shortest children, or scarce health care resources would be devoted to GH therapy rather than the primary health care so many are now denied.
For the above reasons, pediatricians need to exercise caution in administering GH therapy. Therapy may be justified for children whose height could prevent them from participating in the basic activities of daily living. Additional research is necessary to determine which of these children actually have increased final adult height, whether they actually experience a psychosocial benefit from any such increase, and whether GH therapy creates significant risks or other negative effects. The results of this research should provide clearer guidance on the future appropriate use of GH therapy.
Pediatricians should also be alert to commercial efforts to stimulate parental interest in GH therapy as the avenue to improved athletic ability and other forms of social “success” for their children (New York Times Magazine. 1991;47:14–17, 28–29). In this difficult area, it is especially important for pediatricians to fulfill their traditional role of protecting their patients from harm, which can include the refusal to administer burdensome and nonbeneficial treatment. Finally, pediatricians should exert their best efforts to counsel parents and children concerned about short stature that there often are better alternatives than GH therapy to address their concerns.
Committee on Drugs, 1995 to 1996
Cheston M. Berlin, Jr, MD, Chairperson
D. Gail McCarver-May, MD
Daniel A. Notterman, MD
Robert M. Ward, MD
Douglas N. Weismann, MD
Geraldine S. Wilson, MD
John T. Wilson, MD
Elizabeth B. Weller, MD
American Academy of Child and Adolescent Psychiatry
Donald R. Bennett MD, PhD
American Medical Association/United States Pharmacopeia
Joseph Mulinare, MD, MSPH
Centers for Disease Control and Prevention
Iffath Abbasi Hoskins, MD
American College of Obstetricians and Gynecologists
Paul Kaufman, MD
Pharmaceutical Research and Manufacturers of America
Carmen Krough, BSc
Health Protection Branch, Canada
Michael J. Rieder, MD
Canadian Paediatric Society
Gloria Troendle, MD
Food and Drug Administration
Sumner J. Yaffe, MD
National Institutes of Health
AAP Section Liaison
Charles J. Coté, MD
Section on Anesthesiology
Stanley J. Szefler, MD
Section on Allergy and Immunology
Jeffrey L. Blumer, MD, PhD
Committee on Bioethics, 1995 to 1996
Joel E. Frader, MD, Chairperson
Lucy S. Crain, MD
Kathryn L. Moseley, MD
Robert M. Nelson, MD
Ian H. Porter, MD
Felipe E. Vizcarrondo, MD
Watson A. Bowes, MD
American College of Obstetricians and Gynecologists
Allessandra Kazura, MD
American Academy of Child and Adolescent Psychiatry
Ernest F. Krug, III, MD
American Board of Pediatrics
Donna A. Caniano, MD
Section on Surgery
Rebecca Dresser, JD Nancy M. P. King, JD
The recommendations in this statement do not indicate an exclusive course of treatment or serve as a standard of medical care. Variations, taking into account individual circumstances, may be appropriate.
- GH =
- human growth hormone
- Tanner JM,
- Whitehouse RH
- ↵Laron Z, Bauman B, Josefsberg Z. Importance of age at initiation of hGH therapy. In: Laron Z, Butenandt O, Raiti S, eds. Clinical Use of Growth Hormone: Present and Future Aspects. New York, NY: S Karger AG; 1987;16:48–60
- ↵Guyda H, Dean H. Intermittent vs continuous GH administration in growth hormone deficiency. In: Laron Z, Butenandt O, Raiti S, eds. Clinical Use of Growth Hormone: Present and Future Aspects. New York, NY: S Karger AG; 1987;16:61–71
- Smith PJ,
- Hindmarsh PC,
- Brook CG
- Bundak R,
- Hindmarsh PC,
- Brook CG
- Brown P
- Frasier SD
- ↵Alsever RN, Gotlin RW, eds. Handbook of Endocrine Tests in Adults and Children. 2nd ed. Chicago, IL: Year Book Medical Publishers; 1978
- Fine RN,
- Kohaut EC,
- Brown D,
- Perlman AJ
- Brook CG,
- Hindmarsh PC
- Genentech Collaborative Study Group
- ↵Physician's Desk Reference. 49th ed. Montvale, NJ: Medical Economics Data Production Co; 1995:1087–1088, 1332–1333
- ↵American Academy of Pediatrics. Findings from Periodic Survey of Fellows #14. Use of Growth Hormone in Pediatric Practice. Elk Grove Village, IL: American Academy of Pediatrics; 1991
- ↵Schwartz ID, Bercu BB. Normal growth and development. In: Hung W, ed. Clinical Pediatric Endocrinology. St Louis, MO: Mosby-Year Book; 1992:13–41
- Hindmarsh PC,
- Brook CG
- Laron Z, Bauman B, Josefsberg Z. Discussion to Job and Preece. In: Laron Z, Butenandt O, Raiti S, eds. Clinical Use of Growth Hormone: Present and Future Aspects. New York, NY: S Karger AG; 1987;16:116–121
- Anneren G,
- Sara VR,
- Hall K,
- Tuvemo T
- Cowell CT,
- Quigley CA,
- Moore B,
- et al.
- Stanhope R
- Frasier SD
- Law CM
- Grew RS,
- Stabler B,
- Williams RW,
- Underwood LE
- Lantos J
- Baily MA
- Abbott D,
- Rotnem D,
- Genel M,
- Cohen DJ
- ↵Cuttler L, Michon R, Reinecke M, Stocking C. Attitudinal and behavioral concomitants of short stature. Adolesc Med. 1990:4A
- Benjamin M,
- Muyskens J,
- Saenger P
- Daniels N
- Kass NE
- Allen DB
- Lippe BM
- Copyright © 1997 American Academy of Pediatrics