| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Pediatrics, Division of Pediatric Endocrinology, Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, New York 10467; and Genentech, Inc., South San Francisco, California 94080
Address all correspondence and requests for reprints to: Paul Saenger, M.D., Division of Pediatric Endocrinology, Montefiore Medical Center, 111 East 210th Street, Bronx, New York 10467.
| Abstract |
|---|
|
|
|---|
In summary, careful monitoring has not revealed any currently discernible metabolic side-effects of clinical significance after GH therapy in this 5-yr study of children with idiopathic short stature.
| Introduction |
|---|
|
|
|---|
Therapy with recombinant GH commonly results in an increase in growth
velocity to a variable degree (3, 5, 6), with a mean result of a
doubling of the pretreatment growth rate in most studies during the
first year of treatment (3, 5). Therapeutic efficacy tends to diminish
thereafter, as seen in other populations treated with GH (7, 8). In a
recent study (2), it was estimated that the predicted final height in
some children without GH deficiency treated with pharmacological doses
of GH increased by approximately 10 cm. The most recent analysis of the
Genentech-sponsored trial (4), shows that GH treatment of patients with
marked short stature (
2 SD score) results in an increase
in mean final height of 9.1 cm for boys (95% confidence limits,
5.412.8) and 5.6 cm for girls (95% confidence limits, 2.09.3)
compared with untreated historial controls whose baseline height was
below -2 SD score for age. However, these patients still
do not achieve their midparent target height (4).
Although these clinical investigations focusing on auxological issues continue, concerns about the metabolic side-effects of GH treatment prompted our examination of the metabolic consequences of GH therapy for idiopathic short stature in a large multicenter trial. As we explore the widening indications for GH use in non-GH-deficient children (1, 9, 10, 11) and the clinical utility of varying GH doses in nonresponders (3), safety aspects in the context of metabolic side-effects need to be carefully investigated.
We studied the effects of 5 yr of GH treatment in children with ISS on levels of insulin-like growth factor I (IGF-I), L-T4, cholesterol, triglycerides, sodium, and potassium. We also measured effects on carbohydrate metabolism, as assessed by glucose and insulin levels in the fasting state and 120 min after a standard oral glucose load, as well as hemoglobin A1c levels.
| Materials and Methods |
|---|
|
|
|---|
|
In subsequent years, the patients were randomized to receive 0.3 mg/kg·week recombinant human GH given either daily or tiw. Of the 121 patients who entered the study, a total of 80 achieved final or near-final height, as defined by a bone age of 16 yr or more in boys and of 14 yr or more in girls (4). Laboratory data were evaluated at 6-month intervals during GH treatment. Blood chemistry, liver function tests (serum glutamic oxaloacetic transaminase, serum glutamic pyruvic transaminase, alkaline phosphatase, and bilirubin), serum T4, blood glucose, insulin, and hemoglobin A1c levels were measured at Smith-Kline Bioscience Laboratories (Van Nuys, CA). Hemoglobin A1c was determined using a chromotographic method (normal, <6.2%). IGF-I was measured using acid chromotography extraction (4, 12).
Glucose challenge was measured at 0 min in the fasting state and at 120 min after a standard glucose load consisting of 1.75 g glucose solution/kg, with a maximum of 75 g/dose. A standardized comparison of changes in biochemical parameters between baseline and treatment values was made using Students two-tailed t test. Results are expressed as the mean ± SD. P values < 0.05 were considered significant. For the 5-yr analyses, patients were included for each metabolic measure if data were available at 0 and 60 months. Because of skewing of data, statistical analyses were performed using median or log-transformed values for insulin and triglyceride measurements. Insulin values are shown as antilog in the figures to facilitate graphical depiction in units.
As daily vs. tiw GH treatment showed no difference in metabolic effects, only combined data for both groups are shown. The auxological difference in the two treatment groups has recently been reviewed (13).
| Results |
|---|
|
|
|---|
Mean IGF-I levels rose from baseline levels during treatment in
both pubertal and prepubertal subjects (Table 2
). Using age-appropriate standard
reference levels, the data, expressed as the SD score for
age and sex, are shown in Fig.
1.
|
Mean T4 levels did not show any
significant changes during the 5-yr follow-up period. There was a
significant increase in T4 levels at 6 months in the group
receiving daily GH. These changes were not accompanied by clinical
symptoms and reverted to pretreatment levels by 12 months (Table 2
).
Cholesterol and triglycerides
There were modest changes in mean total cholesterol and mean
triglyceride concentrations during treatment that were not
statistically or clinically significant (Table 2
). At all times mean
serum levels for both cholesterol and triglycerides remained in the
normal range when applied to recently developed age-appropriate
standards (14).
Na, K, and blood pressure
There were no significant changes in mean serum sodium or mean
serum potassium during the 5 yr of treatment (Table 2
). There was no
change in mean blood pressure during the entire observation period.
Blood chemistries
Mean serum glutamic oxaloacetic transaminase, serum glutamic pyruvic transaminase, and bilirubin levels showed no significant changes during the 5-yr period. Mean alkaline phosphatase rose in the treatment group (data not shown).
Carbohydrate tolerance and hemoglobin A1c
Mean baseline and 2 h postprandial glucose levels remained
unchanged throughout the treatment period, whereas insulin showed a
rise during treatment from low normal levels into the normal range
(Table 2
). After year 3 of the study, mean basal and postglucose load
insulin levels showed no further rise (Fig. 2
, a and b). The increases in mean
postprandial insulin levels were 7-fold at 3 yr and 9-fold at 4 yr
compared with control values. Follow-up of individual patients showed
that no study subject became hyperinsulinemic in either the fasting or
the postprandial state. Mean fasting and postprandial insulin levels
remained within the normal range. Hemoglobin A1c levels
(Fig. 3
) remained stable in the normal
range throughout the study. Glucose metabolism data were available for
a smaller number of patients (see Table 2
). There was, however, no
statistically significant difference between the baseline
characteristics of patients with or without glucose metabolism results.
Thus, these results are representative of the general patient
population.
|
|
| Discussion |
|---|
|
|
|---|
IGF-I
In response to GH administration, serum levels of IGF-I rise, as expected. The mean levels achieved remain within the normal range using age-appropriate standards. In response to GH administration, serum levels of IGF-I rose significantly. However, standardized scores returned to baseline levels by 48 months. IGF-I levels were less than those achieved in GH-deficient patients taking similar doses, possibly due to partial GH insensitivity in some ISS patients (16). IGF-I monitoring also may prove to be a useful tool to evaluate compliance with GH therapy.
T4
Administration of human GH to patients with GH deficiency has been reported to produce a variety of perturbations of thyroid function. A recent analysis of the effects of recombinant human GH on thyroid function showed that after 4 days of GH treatment (0.125 mg/m2·day), mean serum T4 levels decreased by 8%; in contrast, mean serum T3 levels increased by 21%. The increase in mean T3 levels led to a 54% decrease in mean TSH levels. These data are most consistent with enhanced extrathyroidal (including intrapituitary) conversion of T4 to T3 with a compensatory decrease in TSH secretion (17). In the same study an inconsequential short term rise in the T3/T4 ratio was seen that was reversed during prolonged treatment. In other long term studies, no significant changes in thyroid hormone levels were observed (18).
Five of the original patients in this study (6) were started on thyroid hormone replacement during the first 3 yr of the study. Their serum T4 levels ranged from 4.45.5 µg/dL at the time that thyroid replacement was initiated. The transient changes in thyroid hormone measurements seen with GH therapy are generally not associated with clinical disease.
Cholesterol/triglycerides
GH administration had a minimal effect on cholesterol and triglyceride levels when measured at annual intervals (14). In a study of normal individuals, GH administration decreased cholesterol levels after 1 week of very high doses, but not after 6 months using lesser amounts (19). In contrast, large doses of GH raised the triglyceride concentration, whereas 6 months of small doses had no effect (19).
The effects of exogenous GH on site-specific adipose tissue distribution (20) were not evaluated in our patients. Changes in adipose tissue distribution during GH therapy may be due to a GH-mediated decrease in insulin responsiveness that takes place in parallel to a decrease in the de novo triglyceride synthesis (20). Body mass index remained normal in the group of children evaluated here.
Fluid, electrolytes, and blood pressure
Mean sodium and potassium concentrations showed no significant change during the 5-yr course of study. Transient sodium and water retention after pituitary GH administration was described 30 yr ago (21). In acromegaly, both intra- and extracellular fluid volumes are increased. Studies of GH-deficient adults using recombinant human GH (22) yielded similar results. In 6 of 24 adults, fluid retention was apparent after 4 weeks of GH treatment, as evidenced by swelling of hands and feet. In a study that sought to determine whether 3 weeks of treatment with GH might facilitate the preservation of nitrogen and accelerate the loss of body fat during dietary restriction, the injection of GH resulted in cessation of weight loss for at least 1 week in 5 of 8 study subjects, and 5 subjects developed mild edema during the first treatment week. Not surprisingly, after GH was terminated, all subjects experienced a water diuresis (mean, 2.06 ± 1.25 kg). When higher doses of GH (10 mg/day for 7 days) were used in a study by Manson and Wilmore (24), water retention and edema formation became even more prevalent.
Fluid retention has been described in children with GH deficiency after both short and long term use of pituitary GH when using antipyrine kinetics as an indirect measure of total body water (25). Recent data by Ho et al. (26) show that the administration of hGH at a dose of 0.1 mg/kg·day for 5 days in adult volunteers acutely leads to an activation of the renin-angiotensin axis, which reached the maximum level 35 days after the first GH dose. Blood pressure did not change however. Although fluid retention commonly accompanies GH excess, it is unlikely that the chronic volume expansion seen in acromegaly arises solely from GH-induced activation of mineralocorticoid secretion. It is well known that escape from sodium retention occurs promptly after protracted aldosterone elevation (26). In a recent study assessing the 1-yr effects of GH on blood pressure, atrial natriuretic factor, aldosterone, and PRA, no changes compatible with sodium retention were seen (27). The relationship between transient sodium retention and pseudotumor cerebri accompanying GH administration remains to be explored (15).
Carbohydrate tolerance and hemoglobin A1c
In hypopituitary children, GH treatment has been associated with normalization of glucose turnover and insulin secretion and has led to either unchanged or increased insulin responses to oral glucose administration (28, 29, 30). Hindmarsh et al. (31) measured glycemic responses to GH therapy in short normal children. Fasting serum insulin concentrations increased from 5.4 ± 3.5 to 15.3 ± 8.7 mU/L at the end of the first year of treatment, which is very similar to our findings.
Fasting levels decreased slightly during the second year. No postprandial insulin levels were reported by these investigators. They conclude that tolerance to the effects of GH on carbohydrate metabolism occurs (31). In children with ISS, Lesage et al. (2) observed nearly complete reversal of hyperinsulinism within 12 months of discontinuation of GH therapy in their group of 10 prepubertal children treated with 23 times the dose of recombinant GH used here. Similarly, in a group of Turner syndrome patients we recently showed a progressive normalization of insulin levels after GH treatment was discontinued (32).
The rise in insulin levels shown here is substantive, although levels remain within the normal range for postprandial insulin levels as defined by the reference laboratory we used. The clinical significance, if any, of these changes is unclear. Follow-up of patients without GH deficiency treated for a similar time period with comparable doses of GH shows a decline in postprandial insulin levels after GH therapy is discontinued (32). These data, therefore, do not support the concept that GH therapy of this duration will lead to ß-cell exhaustion.
Heptulla et al. (33) recently reported that the effects of GH on insulin sensitivity in prepubertal children with ISS anticipate the changes in carbohydrate tolerance seen typically during normal adolescence. GH treatment in ISS may lead to a prolongation of the physiological state of insulin resistance seen in normal puberty (34, 35). The long term implications of prolonged insulin resistance are not known. It is reassuring, however, that despite uninterrupted GH treatment, the insulin levels when measured fasting and 2 h after a standard glucose load showed no further rise after the first 3 yr of GH therapy. More detailed analysis of the insulin resistance associated with puberty using euglycemic and hyperglycemic clamp techniques (34, 35) suggests that GH-induced insulin resistance spares hepatic action of insulin and is limited to peripheral glucose metabolism. The insulin resistance is thus associated with a greater fall in branched chain and other essential amino acids. The hyperinsulinemia seen in puberty or during GH therapy may, therefore, amplify the anabolic effects of insulin on protein metabolism in puberty (33, 34).
It is currently not known whether the metabolic effects of GH are dependent on the mean serum value of GH over a given time period, the total GH circulating in plasma during that period (area under the curve), the number and/or amplitude of GH peaks achieved, or the maximum GH levels achieved. Using the dosage schedule described in this study, peak GH levels as high as 32 ng/mL have been measured after sc injections (36). This order of magnitude of GH levels is seen in 24-h studies of GH dynamics in normal healthy control children (37). In addition, IGF-I concentrations or the concentration of IGF-binding proteins may modulate the metabolic effects of GH (38).
In summary, careful monitoring has not revealed any currently discernible metabolic side-effects of clinical significance during this long term study of GH therapy in children with ISS. In particular, insulin levels increase but remain within the normal range, as do glucose and hemoglobin A1c. Although GH treatment in this group of children appears to be safe, continued surveillance is necessary.
|
| Footnotes |
|---|
2 The Genentech Collaborative Study Group is composed of: A. J.
Johanson (Genentech, South San Francisco, CA), J. Baptista (Genentech),
J. Kuntze (Genentech), R. Blizzard (University of Virgina,
Charlottesville, VA), J. Cara (Wright State
University, Detroit, MI), S. Chernausek (Childrens Hospital,
Cincinnati, OH), M. Geffner (University of California-Los Angeles
Medical Center), J. Gertner (Cornell Medical Center, New York, NY), N.
Hopwood (University of Michigan, Ann Arbor, MI), S. Kaplan
(University of California-San Francisco Medical Center), B. Lippe,
University of California Los Angeles Medical Center), L. Plotnick
(Johns Hopkins Hospital, Baltimore, MD), and A. Rogol (University of
Virginia, Charlottesville, VA). ![]()
Received December 5, 1997.
Revised April 10, 1998.
Accepted May 22, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. A. Quigley, A. M. Gill, B. J. Crowe, K. Robling, J. J. Chipman, S. R. Rose, J. L. Ross, F. G. Cassorla, A. M. Wolka, J. M. Wit, et al. Safety of Growth Hormone Treatment in Pediatric Patients with Idiopathic Short Stature J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5188 - 5196. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. L Weise and M. C Nahata Growth Hormone Use in Children with Idiopathic Short Stature Ann. Pharmacother., September 1, 2004; 38(9): 1460 - 1468. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. O. Reiter, K. M. Attie, T. Moshang Jr., B. L. Silverman, S. F. Kemp, R. B. Neuwirth, K. M. Ford, and P. Saenger A Multicenter Study of the Efficacy and Safety of Sustained Release GH in the Treatment of Naive Pediatric Patients with GH Deficiency J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4700 - 4706. [Abstract] [Full Text] [PDF] |
||||
![]() |
Growth-Promoting Strategies in Turner's Syndrome J. Clin. Endocrinol. Metab., December 1, 1999; 84(12): 4345 - 4348. [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |