Growth hormone treatment of children with myelomeningocele

Presented at the National Cooperative Growth Study Ninth Annual Investigators Meeting, supported by an educational grant from Genentech, Inc.

Growth hormone treatment of children with myelomeningocele D e b o r a h Rotenstein, MD,° a n d Timothy J, Breen, PhDb From the Department of Pediatrics, Medical College of Pennsylvania and Hahnemann University, Allegheny Campus, Allegheny General Hospital, Pittsburgh, Pennsylvania, and the Department of Biostatistics, Genentech, Inc., South San Francisco, California

From the National Cooperative Growth Study database 106 patients (53 boys) with myelomeningocele who were treated with recombinant human growth hormone (GH) at 56 centers were identified. Eighty-one patients (41 boys) were prepubertal at enrollment. The mean pretreatment growth rate (GR) in these prepubertal patients was 4.5 ± 3.7 cm/yr, and the mean height SD score was -4.0 ± 1.2. The maximal stimulated GH level was less than 10 IJg/L in 71% of these patients and less than 7 IJg/L in 49%. The mean chronologic age was 6.5 ± 2.9 years, and the mean height age was 3.4 ± 1.7 years. After GH treatment the year I GR in those who remained prepubertal was 8.5 ± 3.3 cm/yr, a significant increase over baseline (p <0.0 I). This increase was sustained through year 4 and remained significant through year 3 (p <0.01). The height SD score showed sustained significant improvement through year 4, to -2.2 ± 1.4 (p <0.001). The GR and SD score for stature improve with GH treatment in children with myelomeningocele. (J Pediatr 1996;128:$28-31)

Myelomeningocele (myelomeningocele or spina bifida) has existed since prehistoric times but was first described by Peter Van-Forestus 1 in 1587. It is the most common major birth defect compatible with a reasonable quality of life. 2 In the absence of central nervous system infection, intelligence is in the normal range. The incidence in the United States has declined from 4 in 1000 births in 1970 to 0.32 in 1000;

Supported by Genentech, Inc., and Allegheny Singer Research Institute. Presented in part at the Ninth Annual Investigators Meeting, National Cooperative Growth Study, New York, N.Y., Oct. 12-15, 1995. Reprint requests: Deborah Rotenstein, MD, Department of Pediatrics, Allegheny General Hospital, 320 E. North Ave., Pittsburgh, PA 15212. aDeborah Rotenstein, MD, is a past recipient of a clinical trial grant from Genentech, Inc. bTimothy J. Breen, PhD, is employed by Genentech, Inc. Copyright © 1996 by Mosby-Year Book, Inc. 0022-3476/96/$5.00 + 0 9/0/72344

$28

however, in some countries the incidence is as high as 10 in 1000. 3 Survival is greater than 95% in the first 2 years of life.4 Although the cause is unknown but is likely to be multifactorial, occurrence is reduced by 70% with prenatal folic acid supplementation. 5 Traditionally defined as a disease of neural tube closure, myelomeningocele is characterized by a more global central nervous system defect. More than 80% of patients have hydrocephalus that necessitates a ventricuioperitoneal shunt, and nearly all have a Chiari II malformaGH GR NCGS SDS

Growthhormone Growthrate NationalCooperative Growth Study Standarddeviation score(s)

tion (medulla and pons are posteriorly bowed and often extend into the cervical spinal canal). 2 Abnormalities of the hypothalamic pituitary axis including central precocious puberty and tertiary hypothyroidism can be present. 6 Children and adults with myelomeningocele have extremely short stature, which has been presumed to be caused

The Jow~al of Pediatrics Volume 128, Number 5, Part 2

Rotenstein and Breen

S 29

Table I. Prepubertal patients with myelomeningocele and idiopathic GH deficiency: NCGS baseline data Idiopathic GH deficiency

Myelomeningocele

Age (yr) Height age (yr) Skeletal age (yr) Height SDS Weight-for-heightSDS

n

Mean

SD

n

Mean

SD

p Value

78 78 45 78 71

6.5 3.4 4.3 -4.0 -0.3

2.9 1.7 2.5 1.2 1.3

6020 5990 3888 5967 5322

8.6 6.0 6.7 -2.9 0.l

42 3.4 3.8 1.1 1.4

<0.001 <0.001 <0.001 <0.001 0.053

by spinal deformities and smaller appearing lower limbs] Associated vertebral anomalies and the frequent occurrence of a tethered spinal cord can lead to scoliosis, which may further create a smaller trunk, s It has been presumed that because of reports of normal arm length, children with myelomeningocele do not have growth hormone deficiency.9 It is believed that a hormonal abnon-nality would affect the growth of both the upper and the lower extremities. However, in short children with myelomeningocele, in 50% of whom the maximal stimulated GH levels were less than 7 gg/L, the radiographic lengths of the radius and the ulna were significantly less than the expected values, s° Hammock et a1.11 reported subnormal GH responses in two of three chip dren with myelomeningocele; however, M e y e r and Landau la described but did not quantify normal nocturnal GH secretion in three pubertal patients with myelomeningocele. Adults with myelomeningocele also have extremely short stature: 141.9 _+ 12 cm for women and 152.1 _+ 13 cm for men. 13 Six months of treatment with GH in seven prepubertal children resulted in a significant increase in the annualized growth rate, from 1.7 ± 0.2 to 7.9 _+ 3.4 cm/yr, l° Longer term treatment (mean duration, 37.6 months) in 22 patients at one center resulted in a sustained improvement in the GR and height SD score through year 4.14 The NCGS is a postmarketing surveillance study in children treated with GH. More than 24,000 patients have been enrolled at more than 400 centers, which makes a multicenter approach to evaluating the growth responses of children possible. This report is based on our evaluation of the baseline values and growth responses of children with myelomeningocele in the NCGS.

METHODS The database was searched to find patients with myelomeningocele who had not been treated with GH before enrollment in the NCGS. Patients in the NCGS were identified as having myelomeningocele if "myelomeningocele," "meningomyelocele," "spina bifida," or "neural tube defect" was reported in the medical history section of the NCGS enrollment form or was noted on a subsequent patient visit form. The data included age, height and skeletal age,

Table II. Maximal stimulated GH levels in prepubertal patients with myelomeningocele GH level (IJglL)

<7 7to<10 ~10

Boys

Girls

Total population (%) (n = 76)

18 9 11

19 8 11

37 (49) 17 (22) 22 (29)

SDS for weight and height, maximum stimulated GH level, average GH dose, baseline GR, and GRs for years 1 through 4. The data were then selected for all patients who were prepubertal at baseline and for those who remained prepubertal through treatment year 4. The data reported fer those who were prepubertal at baseline were reviewed for associated medical conditions and concomitant medications. Growth rates before treatment were calculated from the height at enrollment and the height at 3 to 18 months before enrollment but as close as possible to 1 year before. These pretreatment GRs may have been calculated over short intervals of 3 to 6 months and may have been obtained under conditions different from those during treatment. The method of measuremerit (height, length, arm span) is also not identified. Treatment GRs were calculated from measurements obtained at visits within 6 months and centered at the anniversary date of the first dose of GH. If several visits had been made, the visit closest to the anniversary date was used. The SDS for weight and height were calculated with data from the National Center for Health Statistics. 15 Data for the prepubertal patients with myelomeningocele were compared with data from the NCGS database for prepubertal patients with idiopathic GH deficiency. The maximal stimulated GH levels were analyzed as reported, but the type of assay (monoclonal or polyclonal antibody) used is not known. The GRs and SDS for height for treatment years 1 through 4 were compared with the pretreatment values. The dose of GH is reported as the average dose over the entire period of treatment converted to mg/kg per week. Frequency data are reported as percentages, and numeric variables are summarized as means and SD. Thep values for

S30

Rotenstein and Breen

The Journal of Pediatrics May 1996

Table Ill. Response to GH treatment in prepubertal patients with myelomeningocele Treatment year

No. of patients Growth rate, cm/yr (SD) No. of patients Height SDS (SD) *p <0.001. *p
Baseline

1

2

3

4

58 4.5 (3.7) 78 -4.0 (1.2)

46 8.5* (3.3) 54 -3.3* (1.3)

26 6.5? (2.7) 40 -2.8* (1.4)

15 7.0? (2.4) 19 -2.4* (1.2)

11 6.6 (1.6) 17 -2.2* (1.4)

the comparison of means and percentages, respectively, are based on t tests and chi-squared tests.

improvement in the height SDS was seen through year 4 (p <0.001).

RESULTS

DISCUSSION

From the NCGS database 106 children (53 boys) with myelomeningocele who were treated with GH were identified. The patients were treated at 56 different centers, each of which treated between 1 and 18 children. Eighty-one patients (41 boys) were prepubertal at enrollment, and at 36 of 49 centers on!y one prepubertal patient was treated. The prepubertal patients had associated medical conditions and were treated with a number of concomitant medications. Hypothyroidism was noted in seven patients and cardiac disease in two patients. Other conditions were congenital adrenal hyperplasia, Turner syndrome, hypoglycemia, asthma, and ambiguous genitalia; one case of each was reported. Only three patients who were treated with anticonvulsants were identified. The most frequently reported medications were urinary agents (oxybutynin chloride in 20 patients, imipramine hydrochloride in four patients, and hyoscyamine sulfate in one patient) and antibiotics (trimethoprim-sulfamethoxazole in 18 patients and nitrofurantoin in nine patients). Other medications were levothyroxine (seven patients), glucocorticoids (four patients), methylphenidate hydrochloride (three patients), chlorothiazide (two patients), and digoxin (one patient). Baseline data for prepubertal patients with myelomeningocele and prepubertal patients with idiopathic GH deficiency from the NCGS database are shown in Table I. The patients with myelomeningocele are younger and have a significantly greater height deficit. The maximal stimulated GH levels were less than 10 ~tg/L in 71% of the prepubertal patients and less than 7 ~ag/L in nearly half of them (Table II). The response to treatment over a 4-year period was assessed in those who remained prepubertal. The mean dose of GH was 0.31 ___0.08 mg/kg per week (range, 0.05 to 0.5 mg/kg per week). The GR increased from the pretreatment value through year 4, remaining significantly different through year 3 (p <0.01) (Table III). Sustained significant

It can be seen from the NCGS data that prepubertal children with myelomeningocele have extremely short stature and also have several associated medical conditions that require treatment. The maximal stimulated GH level of less than 7 ~ag/L in 49% of these patients suggests that GH deficiency should be included among the hypothalamic pituitary abnormalities associated with myelomeningocele. The GRs and height SDS in these prepubertal children are improved through year 4 in response to GH, but the effect on adult stature is not known. The centers that report to the NCGS do not necessarily have a uniform method of data collection. This drawback could be partially overcome if the patient population were large, but the number of prepubertal children with myelomeningocele in the NCGS is relatively small. However, it has been reported that the GRs and height SDS were increased in eight prepubertal children with myelomeningocele who were treated at one center where the same data-collecting methods were used uniformly over a 3-year interval, t4 The results in that report are similar to those observed with the NCGS data and aid in validating our findings. Treatment with GH has a significant effect on growth symmetry, obesity, and muscle strength in patients with myelomeningocele. Preliminary data from a clinical trial in nine prepubertal children with myelomeningocele who were treated with GH for 1 year showed an increase in their sitting height and upper and lower extremity lengths without a change in their upper-to-lower-segment ratio (D.R., unpublished data, 1995). An increase in their weight SDS also occurred without a change in their body mass index, and a decrease occurred in the sum of their triceps and subscapular skin fold thickness. An increase in their basal metabolic rate suggests an increase in lean body mass, and dynamometer readings before and during treatment showed an increase in muscle strength.

The Journal of Pediatrics Volume 128, Number 5, Part 2

Further study is necessary to determine the effects of G H in children with myelomeningocele in terms of growth and increased adult stature, body symmetry, obesity, muscle strength, and self-esteem and rehabilitative potential. We thank Julie Caligiuri and Barbara Mansfield, RN, for their technical and statistical assistance. REFERENCES 1. Van-Forestus P. Decapitis et cerebre morbis ac symptomatis. Observationum Medic inglium libri III. Leiden, Ex Off Plantiniana Raphelemagii, 1587. 2. Reigel D, Rotenstein D. Spina bifida. In: Cheek W, editor. Pediatric neurosurgery: surgery of the developing nervous system. 3rd ed. Philadelphia: WB Saunders, 1994:51-76. 3. Centers for Disease Control and Prevention. Spina bifida incidence at birth United States. MMWR 1992;4t:497-500. 4. McLone D. Continuing concepts in the management of spina bifida. Pediatr Neurosurg 1992;18:254-6. 5. Czeizel A, Dudas I. Prevention of the first occurrence of neural tube defects by periconceptional vitamin supplementation. N Engl J Med 1992;327:1832-5. 6. Perrone L, Det Gaizo D, D'Angelo E, Rea L, DiManso G, DelGado R. Endocrine studies in children with myelomeningocele. J Pediatr Endocrinol 1994;7:219-23.

Rotenstein and Breen

S31

7. Rosenblum MF, Finegold DN, Chamey EB~ Assessment of stature of children with myelomeningocele and usefulness of arm-span measurement. Dev Med Child Neurol 1983;25:33842. 8~ Reigel D. Tethered spinal cord. Concepts Pediatr Neuros~g 1983 ;4:142-64. 9. Greene SA, Frank M, Zachmann M, Prader M. Growth and sexual development in children with myelomeningocele. Eur J Pediatr 1985;144:146-8. 10. Rotenstein D, Reigel D, Flora L. Growth hormone treatment accelerates growth of short children with neural tube defects. J Pediatr 1989;317:417-20. 11. Hammock MK, Milhorat TH, Baron IS. Normal pressure hydrocephalus in patients with myelomeningocele. Dev Med Child Neurol Suppl 1976;18:55-68. i2. Meyer S, Landau H. Precocious puberty in myelomeningocele patients. J Pediatr Orthop 1984;4:28~31. 13. Rotenstein D, Adams M, Reigel D. Adult stature and anthropomorphic measurements of patients with myelomeningocele. Eur J Pediatr 1995;154:398-402. 14. Rotenstein D, Reigel D. Growth hormone treatment of children with neural tube defects: results from 6 months to 6 years. J Pediatr 1966;128:184-9. 15. Hamil] PVV. NCHS growth curves for children birth to 18 years: United States. Hyattsville, Md: National Center for Health Statistics, 1977; DHEW publication no. 78-1650. (Vital and health statistics; series 11; no 165):20-3.