Somatomedin activity in human cord plasma and relationship to birth size,insulin, growth hormone,and prolactin

Somatomedin activity was determined by a rabbit chondrocyte bioassay in cord plasma from babies of between 37 and 41 wk gestation. A positive correlation (P<0.001) was found between plasma somatomedin activity and birthweight. The mean somatomedin activity in infants whose birthweights were within 1 SD of the mean (3293 g) was 0.76 ± 0.27 U/ml. Mean somatomedin activity in infants whose weight was (a) greater than the mean weight ± 1 SD was 1.3 ± 0.17 U/ml, and (b) less than the mean weight ? 1 SD was 0.48 ± 0.15 U/ml. Plasma somatomedin activity was also correlated with placental weight, P< 0.02 and gestational age, P<0.05. No correlation was found between plasma somatomedin activity and birth length, OFC, most measurements of skinfold thickness, cord plasma, growth hormone, prolactin or insulin.     

The short child with subnormal plasma somatomedin C

In 59 male and 59 female healthy children of average stature between 7 and 10 yr old, the normal range of plasma somatomedin C was investigated. The 95% tolerance limits narrowed progressively when the child's plasma somatomedin C status was described by the mean of one, two, three, or four determinations at 6-wk intervals. The 95% tolerance limits were therefore based on the mean of four determinations. In 97 children, age 7 to 10, below the 3.0 percentile in height, 44 had an average plasma somatomedin C below the 2.5 percentile. Among these hyposomatomedinemic short children, 19 were partially or totally deficient in growth hormone, 20 had normal immunoreactive growth hormone responses to dopa, glucagon, and sleep (nongrowth hormone deficient), and five had borderline provocative tests. Both growth hormone deficient and nongrowth hormone deficient children showed significant linear growth responses to 6-month courses of human growth hormone (0.16 to 0.70 unit/kg/wk). The responses of the latter group were 50 to 90% as great as those of the former.     

Empty sella in short children with and without hypothalamic-pituitary abnormalities

A study was conducted on growth hormone (GH) response to oral clonidine (0.15 mg/m²), GH and cortisol responses to i.m. glucagon (0.1 mg/kg), and glucose response to an oral load of glucose (1.75 g/kg). Measurements were made on the circulating concentrations of free thyroxine (FT4), thyroid stimulating hormone (TSH) and different growth parameters and CT sellar images in 25 GH deficient children (Peak GH response to clonidine and glucagon<7 ug/ml), 15 growth retarded children (Ht<5th percentile for age and gender) with sickle cell disease (SCD) and GH deficiency, 30 randomly selected children with normal variant short stature (NVSS) (HtSDS 2SD below the mean for age and gender with normal GH response to stimulation (>10 ug/ml) and 20 age-matched normal children were evaluated. Out of the 25 children with GH deficiency, five had multiple pituitary hormonal deficiency (GH<TSH and/or ACTH. deficiencies), and 20 had isolated GH deficiency. Empty sella, either complete or partial, was detected in 9 out the 20 children with isolated GH deficiency (45%), 4 out of the 5 children with multiple pituitary deficiency (80%), all the children with SCD and GH deficiency (100%), 3 out of the 30 children with NVSS (10%) and in none of the normal children. The insulin-like growth factor-1 (IGF-I) concentrations were significantly lower in the two groups of children with GH deficiency compared to those with NVSS. The height standard deviation scores (HTSDS) were significantly lower and the annual growth velocity was slower in children with idiopathic GH deficiency and empty sella compared to those with NVSS and those with empty sella associated with SCD. The bone age delay (yr) did not differ among the 3 groups of children with short stature. All children with isolated GH deficiency associated with empty sella had normal body mass indices (BMI), while all the children with SCD and empty sella had BMI below the 5th percentile for the corresponding age and gender. None of the children had glucose intolerance. In conclusion, children with growth retardation and abnormal hypothalamic pituitary functions have high incidence of empty sella. However, empty sella is detected in considerable number (10%) of short children with normal hypothalamic pituitary function.     

Growth hormone deficiency after chemotherapy for acute lymphoblastic leukemia in children who have not received cranial radiation

Chemotherapy-related growth failure is a significant problem in children with acute lymphoblastic leukemia (ALL) and other childhood cancers. Growth impairment after cranial radiation (CR) can result in diminished adult height, but growth failure following chemotherapy without CR is usually followed by catch-up growth and normal adult height.1 A retrospective review of 347 ALL survivors registered in our Long Term Follow Up (LTFU) Clinic, since 1997 revealed that 109 had received CR; 3, total body irradiation (TBI); and 235, neither CR nor TBI. For patients whose growth velocity slowed, growth hormone (GH) levels and pediatric endocrinology referrals were obtained. Among the 112 ALL survivors who had received some form of CR, 5 had significant growth failure with growth hormone deficiency (GHD). Among the 235 ALL survivors treated with chemotherapy without CR, 2 were diagnosed with growth failure and GHD. We report the two survivors of childhood ALL treated with chemotherapy without CR who required GH replacement due to absence of catch-up growth. A 15-year-old boy and a 12-year-old girl, off therapy for 9 and 6 years, respectively, were evaluated for decreased growth velocity and failure of catch-up growth. Peak GH responses to stimulation using arginine and clonidine were 3.4 and 3.0 ng/ml, respectively (normal >10 ng/ml). Other causes of growth failure were ruled out, and GH replacement therapy was instituted. Their chemotherapy had included methotrexate, 6 mercaptopurine, vincristine, adriamycin, cyclophosphamide, L-asparaginase, dexamethasone, cytarabine, 6 thioguanine, and intrathecal methotrexate. The growth of all children treated with intensive chemotherapy, regardless of whether CR was administered, should be closely monitored with measurement of standing height at 6 months intervals until growth is complete.     

Subnormal Growth During Puberty in Children Treated for Acute Lymphoblastic Leukemia

Growth was studied longitudinally in 19 children who were long-term survivors after acute lymphoblastic leukemia (ALL). Of the children, 13 were girls; 6 were boys. They had all undergone a 3-year cytostatic treatment period which included vincristine, adriamycin, asparaginase, methotrexate, purinethol, and prednisone. Prophylactic cerebral irradiation (20-24 Gy) had been given to all children; 4 of them had also been given irradiation to the spine (10 Gy). The pattern of growth was nearly identical in girls and boys. Growth in relation to the therapy was almost normal, whereas growth during puberty was subnormal and final height was 1.3 SD less than expected at onset of disease. The growth pattern was the same for children with cerebrospinal irradiation as for those with cerebral irradiation. In view of the present results and previous studies on growth hormone (GH) secretion after cerebral irradiation, we suggest that treatment with luteinizing hormone releasing hormone (LHRH) or GH could be considered at puberty for children who have been treated for ALL, including cerebral irradiation, and who have a poor prognosis for final height.     

Normal growth despite abnormalities of growth hormone secretion in children treated for acute leukemia.

We have studied the relationship between abnormalities of the growth hormone-somatomedin axis and growth in 26 children previously treated for acute lymphatic leukemia. Each child had previously received cranial irradiation, was in complete clinical and hematologic remission, and off all drugs.The mean standing height SDS of the 26 children was significantly less than normal. There was no significant difference between the mean standing height SDS, height velocity SDS, somatomedin activities, and degree of bone age retardation between the 17 children who received the higher dose of cranial irradiation (Group 1) and the nine who had the lower dose of cranial irradiation (Group II). Furthermore, there was no significant reduction in mean height velocity SDS, somatomedin activity, or bone age in either group when compared to normal age-matched controls. The peak GH responses to both insulin hypoglycemia and an arginine test were significantly lowered in Groups I and II when compared to a control group of children. We conclude that only a minority of children, who previously received cranial irradiation for ALL were clinically GH deficient and, therefore, likely to benefit from GH therapy despite the finding that the majority of these children had reduced GH responses to pharmacologic stimuli.     

Chronic Growth Retardation with Normal Growth Hormone Response to Provocative Stimuli and Low Somatomedin Activity

ABSTRACT. Four prepubertal children with chronic growth retardation (growth velocities ≤4 cm/yr), normal growth hormone (GH) response to provocative stimuli and low basal but increased somatomedin activity values after GH administration, received continuous GH-therapy (4 IU/three times a week) for an 18–24-month period. Growth velocity doubled during the first 12 months of therapy and remained 4–6 cm/yr until the end. Bone age progressed according to chronological age and adult height predictions improved. No thyroid function or carbohydrate and lipid metabolism anomalies were observed. After completion of this GH-therapy period, patients remained off treatment during the following six months. Growth velocities were similar to pre-GH-treatment values in two patients, lower in the third and higher in the fourth, who was by then pubertal. Thus, in these patients, long-term GH-therapy promoted growth and improved adult height prediction.     

Constitutional delay of growth and pubertal development: growth hormone secretory pattern and possible therapy

Constitutional delay of growth and pubertal development is a frequent cause of short stature. These children have a significant retardation of skeletal age and delayed sexual development. They generally maintain a normal growth curve and tend to attain normal adult height. Although children with constitutional delay of growth are believed to have no medical or endocrine abnormality to explain their short stature, some controversy regarding their growth hormone secretory status has recently surfaced; some authors have reported low growth hormone levels to provocative stimuli and decreased growth hormone secretion during sleep, as well as low somatomedin C values in some children with constitutional delay of growth. Others, however, have found the growth hormone secretory status to be normal and similar to that of a control population. The implications of these findings, particularly in regard to possible forms of therapy, are discussed in some detail.     

The relation between the secretion of growth hormone (GH), somatomedin C/IGF I (IGF I) and steroids before and after the onset of puberty in patients of small stature

Somatomedin C/IGF I, dehydroepiandrosterone sulfate (DHAS), testosterone (T) or estradiol (E2) have been measured in 154 patients of a previous study in which growth hormone (GH) responses to classical pharmacologic stimuli and spontaneous growth hormone secretion during sleep were compared in short children before and at the beginning of puberty. Five groups were identified: Group I, normal growth hormone secreting children; group II, completely growth hormone deficient; group III, partially growth hormone deficient; group IV, with normal sleep secretion and low responses to stimuli; group V, with the reverse situation. The somatomedin C/IGF I levels were widely dispersed. In group I, the mean +/- SEM levels of somatomedin C/IGF I were 0.77 +/- 0.047 U/ml before puberty and 1.36 +/- 0.142 U/ml in early pubertal patients, with a relation to age (r = 0.52, p less than 0.001). The difference between prepubertal and pubertal patients was significant. In groups II to V, there was no pubertal rise of somatomedin C/IGF I. In group II, the mean IGF I level was 0.48 +/- 0.05 U/ml, significantly lower than in prepubertal patients of group I. In groups III, IV and V, it was 0.7 +/- 0.069 U/ml, 0.8 +/- 0.059 U/ml, and 0.73 +/- 0.059 U/ml respectively, not different from prepubertal patients of group I, but significantly lower than in early pubertal patients of the same group. In prepubertal patients, somatomedin C/IGF I was slightly but highly significantly correlated to growth hormone sleep secretion (r = 0.27, p less than 0.001) and to dehydroepiandrosterone sulfate (r = 0.36, p less than 0.001), but growth hormone and dehydroepiandrosterone sulfate were not correlated with each other.(ABSTRACT TRUNCATED AT 250 WORDS)     

Growth Hormone Response to Growth Hormone-Releasing Hormone (hp GHRH-1-44) as an Index of Growth Hormone Secretory Dysfunction after Prophylactic Cranial Irradiation for Acute Lymphoblastic Leukemia (24 Grays)

ABSTRACT. The growth hormone response to growth hormone releasing hormone hp GHRH_1-14 (2 μg/kg i.v.) was studied in 19 prepubertal children who had been irradiated with 24 Gy for acute lymphoblastic leukemia (ALL) or lymphosarcoma (LS) at a mean chronological age of 410/12 years (limits 10/12 to 9 years). They were evaluated after a mean time interval of 4 8/12±3/12 years and compared to 14 prepubertal children with constitutional short stature (CSS). The individual responses to GHRH were decreased in all but three of the irradiated children. The mean GH response was 16.7±2.5 ng/ml as compared to 52.6±8.5 ng/ml in the control group ( p <0.001). The GH response to GHRH was not correlated with the GH response to arginine-insulin tolerance test (AITT). A decreased response to GHRH with values between 12.5 and 19.4 ng/ml was observed in four cases with normal growth rates and normal GH responses to AITT. These results suggest that an impaired GH response to GHRH is a frequent finding after cranial irradiation for ALL or LS and may be the only sign of GH secretory dysfunction. It is probably indicative of early hypothalamic impairment of GH secretion.     

Growth-stimulating effects of recombinant human growth hormone in children with end-stage renal disease

Because patients with uremia have evidence for growth hormone resistance, we investigated whether this resistance can be overcome by administration of recombinant human growth hormone in supraphysiologic doses in children with severe uremia. Nine stunted children with end-stage renal disease (median age 5.8 years, median bone age 2.7 years) were treated with recombinant human growth hormone, 4 IU/m2/day subcutaneously, for a period of 1 year. Median height velocity was increased from 4.4 cm/yr before therapy to 8.0 cm/yr during treatment. Negative values for height velocity standard deviation scores for chronologic age were improved from a median of -2.6 to +1.5 without advancing bone age more than chronologic age. The growth hormone-insulin-like growth factor I resistance may be explained in part by the increased serum concentration of the high molecular weight insulin-like growth factor binding protein despite normal insulin-like growth factor I serum concentration. Treatment with recombinant human growth hormone improved the ratio between the serum concentrations of insulin-like growth factor I and its binding protein, and normalized the somatomedin bioactivity in the growth cartilage bioassay.     

Plasma somatomedin activity in an infant with Beckwith—Wiedemann syndrome

Plasma growth hormone levels and somatomedin activity were determined in a child with Beckwith-Wiedemann syndrome at birth and at 8 mth of age. Birthweight and length were above the 97th centile. Somatomedin activity in the cord plasma was elevated (2.8 U/ml) compared with controls (0.15-1.3 U/ml; n=15). Growth hormone was also high (76 ng/ml compared with control group range of 5.5–42.1 ng/ml, n=26). At 8 mth of age both somatomedin activity and plasma growth hormone had fallen to normal levels and weight and length were on the 75th centile. It is suggested that the high somatomedin activity may have been a contributing factor in the excessive fetal growth of this child.     

Diagnostic limitations of spontaneous growth hormone measurements in normally growing prepubertal children

To evaluate whether the measurement of the spontaneous overnight growth hormone secretion in prepubertal children clearly separated normal children from subjects with growth hormone deficiency, we studied 45 prepubertal normally growing children (10 with normal height and 35 with constitutional growth delay) and compared their overnight growth hormone secretion with that of a group of subjects with either isolated growth hormone deficiency or neurosecretory dysfunction. Peak growth hormone levels (greater than or equal to 10 ng/mL) following oral clonidine administration were normal in individuals with normal height, constitutional growth delay, and neurosecretory dysfunction, as was the basal somatomedin C concentration; subjects with growth hormone deficiency had low peak growth hormone levels (less than 10 ng/mL) following oral clonidine administration as well as low basal somatomedin C values. The mean 9-hour overnight growth hormone concentration, total growth hormone output, total number of nocturnal pulses, and the mean peak growth hormone response during nocturnal sampling were similar in the normal height and constitutional growth delay groups and significantly greater than those seen in subjects with either growth hormone deficiency or neurosecretory dysfunction. Twelve (26.6%) of 45 normally growing children (4 to 10 normal height and 8 of 35 constitutional growth delay), however, had low overnight growth hormone levels (less than 3 ng/mL), which overlapped results obtained in the growth hormone-deficient or neurosecretory dysfunction groups. Frequent overnight growth hormone (GH) sampling does not always separate normal-growing children from those with partial or complete GH deficiency. In our this study over one quarter of the normally growing children had overnight GH levels in the range of children with either GH deficiency or neurosecretory dysfunction. These findings, in addition to the cost and difficulty in performing this test, do not support the measurement of spontaneous GH as a routine test in short but normally growing prepubertal children.     

A cross-sectional study of growth, puberty and endocrine function in patients with thalassaemia major in Hong Kong

Methodology: A cross-sectional study of growth, puberty and endocrine function was performed on 35 girls and 33 boys with thalassaemia major.
Results Despite regular transfusion and chelation therapy, 75% of the girls and 62% of the boys over the age of 12 years were below the third percentile for height. Hypogonadotropic hypogonadism was found in a similar percentage of patients. Moderate to marked zinc deficiency secondary to chelation therapy was considered unlikely because normal serum zinc levels were found in all but three of our patients, but we could not exclude the possibility of a marginal status of zinc nutrition causing growth failure. Growth hormone deficiency and diabetes mellitus were sometimes encountered but hypothyroidism, hypoparathyroidism and adrenal insufficiency were rare among our patients. Most of the patients with growth failure had normal growth hormone (GH) response to insulin induced hypoglycaemia. The serum insulin-like growth factor-1 (IGF-1) levels were low in our patients and no significant difference in the serum IGF-1 levels was found between prepubertal children with or without growth failure (0.4±0.1 mU/mL vs 0.37±0.11 mU/mL, P = 0.39). Similarly, no difference in the serum IGF-1 levels was found between pubertal children with or without growth failure (0.48 ± 0.2 U/mL vs 0.56 ±0.14 U/mL, P= 0.26).
Conclusions Delayed sexual maturation and a possible defect in growth unrelated to the GH-IGF-1 axis may be responsible for the growth failure in adolescent children with thalassaemia major.     

Growth hormone therapy in Silver Russell Syndrome: 5 years experience of the Australian and New Zealand Growth database (OZGROW)

Data were analysed on 33 children (22 males) with Silver Russell syndrome treated with growth hormone for periods up to 5 years. Baseline data (medians) at commencement of growth hormone (GH) therapy were age 6.7 years, bone age delay 1.7 years, height standard deviation score (SDS)-3.2, weight SDS –3.1, and growth velocity 5.7 cm/ year. All were prepubertal. Median birth weight SDS for gestational age was –3.2. GH was commenced at 14 IU/m² per week and subsequently adjusted according to response. Growth velocity and growth velocity SDS for chronological age (CA) improved over baseline and gains in height SDS for CA were 1.0, 1.5 and 1.8 SD over 3, 4 and 5 years respectively (P < 0.001). No significant increase in height SDS for bone age was observed. Increased GH doses were required after the 1st year to maintain growth rates. Mean bone age advancement was 3.1 years after 3 years of treatment, and 6.0 years after 5 years treatment. Younger age was a predictor of the growth response over the 1st year. Predictors of response after 3 years were catch-up growth, low weight SDS at birth and low height SDS for CA. Age at onset of puberty was normal, but height at onset of puberty was lower than normal means. Conclusion We have demonstrated significant improvement in growth in Silver Russell syndrome after 3 years of GH therapy, however data on estimated mature height and final height are insufficient to conclude final outcomes. Further follow up is required to assess the long-term benefit. Received: 19 July 1995 Accepted: 4 March 1996     

Insulin-like growth factors I and II in evaluation of growth retardation

Plasma samples from 68 growth hormone (GH)-deficient children (provocative serum GH level less than 7 ng/ml), 44 normal short children, and 197 children with normal height were assayed by specific radioimmunoassays for the somatomedin peptides, insulin-like growth factors (IGF)-I and -II. Eighteen percent of the GH-deficient children had IGF-I levels within the normal range for age, whereas 32% of normal short children had low IGF-I levels. Low IGF-II levels were found in 52% of GH-deficient children, but also in 35% of normal short children. However, only 4% of GH-deficient children had normal plasma levels of both IGF-I and IGF-II. Furthermore, only 0.5% of normal children and 11% of normal short children had low plasma levels of both IGF-I and IGF-II. We conclude that plasma levels of either IGF-I or IGF-II overlap in GH-deficient and normal short children, but that the combination of radioimmunoassays may permit better discrimination among normal, normal short, and GH-deficient children.     

Investigation of Growth Hormone Secretion in Patients with Intrauterine Growth Retardation

Rochiccioli, P., Tauher, M., Moisan, V. and Pienkowski C. (Department of Paediatrics, CHU Rangneil, Toulonse Cedex, France). Investigations of growth hormone secretion in patients with intrauterine growth retardation. Acta Paediatr Scand [Suppl] 349: 42, 1989.
Growth hormone (GH) deficiencies have rarely been reported in intrauterine growth retardation (IUGR). This study has investigated GH secretion using GH provocation tests, 24-hour GH secretory profiles, and insulin-like growth factor I (IGF-I) measurements in 24 children with intrauterine growth retardation. The criteria for diagnosis were a birth length and weight below the 10th percentile for gestational age. The average age at investigation was 5.5 years, and the average growth retardation was -3.3 SD. Twenty children had shown catch-up growth between the ages of 6 months and 3 years, followed by varying decreases in growth velocity. Studies of GH secretion demonstrated GH deficiency in 16 patients, with neurosecretory dysfunction in six. Treatment with pituitary GH in nine children increased mean growth velocity from 3.5 cm/year to 7 cm/year. GH therapy should thus be effective in improving the height prognosis of children with intrauterine growth retardation.     

Deficiency of non-suppressible insulin-like activity in thalassaemia major.

The commonly occurring short stature in the condition of thalassaemia major was investigated with respect to the possible role of the somatomedin growth factor low molecular weight non-suppressible insulin-like activity (NSILAs). Nineteen affected patients (12 boys, 7 girls) aged between 2 and 21 years were studied. Twelve of them were on or below the 10th centile for height of whom 7 were on or below the 3rd centile for height. Serum immunoreactive growth hormone responses to exercise were normal in 9 of 11 subjects tested. Using an isolated fat cell bioassay NSILAs was undetectable in 10 and was more than 2 SD below the normal mean value in the other 9 subjects. High molecular weight NSILA (not a growth factor) was very low or undetectable in all 9 subjects tested. Low molecular weight NSILAs did not show the normal correlation with age in childhood, nor was there any correlation with height, height velocity, or bone age. The 2 children above the 50th centile for height had undetectable NSILAs. There was no evidence of iron or ferritin interfering in the bioassay, and mixing experiments showed no evidence of inhibitory activity towards NSILAs in thalassaemic sera. Low circulating levels of the somatomedin NSILAs may contribute to the short stature in thalassaemia major, but other factors may permit normal growth in some affected children.     

Deficiency of non-suppressible insulin-like activity in thalassaemia major

The commonly occurring short stature in the condition of thalassaemia major was investigated with respect to the possible role of the somatomedin growth factor low molecular weight non-suppressible insulin-like activity (NSILAs). Nineteen affected patients (12 boys, 7 girls) aged between 2 and 21 years were studied. Twelve of them were on or below the 10th centile for height of whom 7 were on or below the 3rd centile for height. Serum immunoreactive growth hormone responses to exercise were normal in 9 of 11 subjects tested. Using an isolated fat cell bioassay NSILAs was undetectable in 10 and was more than 2 SD below the normal mean value in the other 9 subjects. High molecular weight NSILA (not a growth factor) was very low or undetectable in all 9 subjects tested. Low molecular weight NSILAs did not show the normal correlation with age in childhood, nor was there any correlation with height, height velocity, or bone age. The 2 children above the 50th centile for height had undetectable NSILAs. There was no evidence of iron or ferritin interfering in the bioassay, and mixing experiments showed no evidence of inhibitory activity towards NSILAs in thalassaemic sera. Low circulating levels of the somatomedin NSILAs may contribute to the short stature in thalassaemia major, but other factors may permit normal growth in some affected children.     

Arm span, serum IGF-1 and IGFBP-3 levels as screening parameters for the diagnosis of growth hormone deficiency in patients with myelomeningocele – preliminary data

Short stature is a common problem in patients with myelomeningocele (MMC) and hydrocephalus. We evaluated auxological and laboratory parameters to differentiate short stature due to neurological defect from short stature additionally caused by growth hormone deficiency (GHD). In a group of 38 prepubertal patients with MMC and hydrocephalus aged 3.8–11.0 years, auxological parameters, including arm span and bone age, and serum insulin-like growth factor 1 (IGF-1) and insulin-like growth factor binding protein-3 (IGFBP-3) levels were measured. Patients with normal supine length (n = 15) had normal arm span. Serum IGF-1 and IGFBP-3 levels were normal (≥ 10th percentile) in 14/15 patients. Twenty-three MMC patients had short stature (height SDS < −2), 11/23 patients had reduced arm span (SDS < −2), and 12/23 had normal arm span. Serum IGF-1 and IGFBP-3 levels were normal in 10/12 of short statured patients with normal arm span, but low (<10th percentile) in those patients with reduced arm span (IGF-1: 8/11 patients, P<0.05; IGFBP-3: 9/11 patients, P<0.005). In 7/11 short statured MMC patients with reduced arm span and low serum IGF-1 and IGFBP-3 levels, growth hormone secretion was investigated. All had a disturbed growth hormone secretion (GHD: n = 4; neurosecretory dysfunction: n = 3). Conclusion Arm span, serum IGF-1 and IGFBP-3 levels are estimated to be appropriate screening parameters for GHD in patients with MMC. Initiating growth hormone therapy should be considered not only according to endocrine findings but also with respect to neurological and orthopaedic anomalies. Received: 27 March 1997 / Accepted: revised form 18 September 1997