Benefit of early commencement of growth hormone therapy in children with Prader-Willi syndrome.

Prader-Willi syndrome (PWS) is a chromosomal disorder and growth failure is a common presentation. Growth hormone (GH) treatment is beneficial in PWS although the optimal age for starting GH is unknown. We investigated whether GH response in PWS was associated with the age of GH commencement by comparing 16 children who commenced GH before 3 years of age (early group) with 40 children who commenced GH after 3 years of age (late group) from the Ozgrow database. Height SDS, body mass index (BMI) SDS, bone age (BA)-chronological age (CA) ratio, change in height (delta Ht) SDS and change in BMI during 4 years of GH treatment were compared between the groups. The early group had better height SDS and delta Ht SDS. BA delay was more pronounced in the early group but BA did not mature beyond CA with GH therapy in either group. Although the initial GH dose for the early group was lower than that of the late group, the former had better height outcome. The starting GH dose seen in the database is lower than the dose used by international centres.

21-hydroxylase genotyping in Australasian patients with congenital adrenal hyperplasia.

Mutations in CYP21 (21-hydroxylase) lead to congenital adrenal hyperplasia (CAH). We genotyped 26 probands with CAH by PCR-sequencing the entire CYP21 gene. 25/26 had homozygous or compound heterozygous mutations. The frequencies of mutations were similar to other populations with deletion/hybrid, I2 G splice and I172N the most common. Five patients with a I172N allele predicting simple-virilising CAH had a salt-wasting phenotype. Two other probands also had a more severe phenotype than predicted by genotype. Two families had both non-classic and salt-wasting phenotypes arising from combinations of three deleterious alleles. Two novel CYP21 alleles were detected: D106N and a large deletion encompassing CYP21 and adjacent pseudogene. Two rare CYP21 alleles were also found. Three of these four novel/rare alleles were only detected as a result of sequencing the entire CYP21 gene. Entire CYP21 sequencing will increase the number of mutations detected in CAH, and in combination with functional studies should contribute a greater understanding of phenotype-genotype correlations.

Androgen receptor genotyping in a large Australasian cohort with androgen insensitivity syndrome; identification of four novel mutations.

We genotyped the androgen receptor (AR) gene in 31 Australasian patients with androgen insensitivity syndrome (AIS). The entire coding region of AR was examined including analysis of polymorphic CAG and GGN repeats in all patients. AR defects were found in 66.7% (6/9) of patients with complete AIS (CAIS) and 13.6% (3/22) of patients with partial AIS (PAIS). A novel deletion (N858delG) leading to a premature stop codon was found in CAIS patient P1. CAIS patient P2 has a novel deletion (N2676delGAGT) resulting in a stop at codon 787. These mutations would result in inactivation of AR protein. A novel insertion of a cysteine residue in the first zinc finger of the AR DNA-binding domain (N2045_2047dupCTG) was found in CAIS patient P3. PAIS patient P4 has a novel amino acid substitution (Arg760Ser) in the AR ligand binding domain, which may impair ligand binding. Five patients were found to have previously reported AR mutations and no mutations were identified in the remaining patients.

Parental pre-pregnancy BMI is a dominant early-life risk factor influencing BMI of offspring in adulthood.

OBJECTIVE: We examined parental and early-life variables in order to identify risk factors for adulthood overweight and obesity in offspring. We report here on the longitudinal prevalence of overweight and obesity in Australian children born between 1989 and 1991 and followed from birth to age 22. METHODS: Data were analysed on 1355 participants from the Western Australian Pregnancy Cohort (Raine) Study, with anthropometry collected during pregnancy, at birth, one year and at three yearly intervals thereafter. Multivariate analyses and cross-sectional logistic regression quantified the timing and contribution of early-life risk factors for overweight and obesity in young-adulthood. RESULTS: At five years of age 12.6% of children were overweight and 5.2% were obese. By early adulthood, the prevalence of obesity had increased to 12.8%, whilst overweight remained relatively stable at 14.2% (range from early childhood to adulthood 11-16%). Parental pre-pregnancy body mass index (BMI) was the strongest determinant of adult offspring BMI. Although rapid first year weight gain was associated with increased offspring BMI, the impact of first year weight-gain diminished over childhood, whilst the impact of parental BMI increased over time. CONCLUSIONS: Parental pre-pregnancy BMI and rapid early-life weight gain predispose offspring to obesity in adulthood.

Regulation of the growth hormone-independent growth factor-binding protein in children.

Regulation of the diurnal variation of the GH-independent insulin-like growth factor-binding protein (BP-28) was studied in 53 children who underwent various investigations for possible endocrine abnormalities. The plasma BP-28 levels increased 12-fold from 8 +/- 2 (+/-SE) micrograms/L at 2100 h to a peak level of 109 +/- 15 micrograms/L between 0600 and 0800 h. This rise was inversely related to plasma insulin levels and was unrelated to plasma cortisol levels. The overnight rise of plasma BP-28 was significantly altered in children who ate a light meal at 0130 h; in them BP-28 levels started to fall after 0300 h, reached nadir levels at 0400 h, began to rise again by 0700 h, and returned to control levels by 0800 h. Such changes did not occur in children given water alone. From the peak early morning level, plasma BP-28 fell to basal levels in children given oral glucose at 0800 h; the t1/2 of the fall was 55 +/- 9 (+/-SE) min. In children who continued to fast, plasma BP-28 did not fall but, rather, increased from 144 +/- 12 micrograms/L at 0800 h after 10 h of fasting to 239 +/- 30 micrograms/L by 1600 h. The induction of hypoglycemia by insulin given at 0945 h after an overnight fast caused a similar but more rapid rise in plasma BP-28 to 668 +/- 317 micrograms/L (range, 208-1763 micrograms/L) by 1230 h. These results suggest that the diurnal variation of plasma BP-28 concentrations in children is not due to an intrinsic rhythm, but is regulated by the metabolic status of the child.

The insulin-like growth factor (IGF)-binding proteins and IGF bioactivity in Laron-type dwarfism.

Laron-type dwarfism (LTD) is caused by a variable defect in the GH receptor gene and is, therefore, an ideal model to study the physiology of the insulin-like growth factors (IGFs) and their binding proteins (IGFBPs) in the complete absence of GH action. In this study we examined the overnight variation of the IGFs, IGFBPs, and IGF bioactivity in two prepubertal subjects with LTD. Subject 1 was a 14-yr-old female, 103 cm tall (-8.3 SD), and subject 2 was a 11.5-yr-old male, 103.6 cm tall (-5.9 SD). Both had serum IGF-I levels below 0.07 U/mL and low constant serum IGF-II levels overnight (185 +/- 10 and 232 +/- 8 micrograms/L), despite high serum GH levels [mean GH, 65 (32.5 micrograms/L) and 53 mU/L (26.5 micrograms/L)]. Serum IGFBP-1 levels increased overnight (from 24 and 22 micrograms/L at 2000 h to 83 and 110 micrograms/L at 0800 h) as serum insulin levels fell [from 19 (136 pmol/L) and 17 mU/L (122 pmol/L) at 2000 h to less than 2 (less than 14 pmol/L) and 5 mU/L (36 pmol/L) at 0800 h] in subjects 1 and 2, respectively. Serum IGFBP-2 levels remained constant overnight, as assessed on Western Ligand blotting and, despite the changes in IGFBP-1, remained the most prominent IGFBP throughout. On size separation, most of the IGF-II (greater than 60%) eluted with IGFBP-2 and the other low mol wt IGFBPs. Serum IGFBP-3 levels were reduced, and IGFBP-3 was not the major IGF carrier in LTD serum, in contrast to normal serum. An IGFBP-3-specific protease that was heat sensitive and cation dependent was identified as the cause of an apparent overnight rise of serum IGFBP-3 levels. No IGFBP-3 variation and no proteolytic activity was seen in normal serum or rapidly separated LTD plasma. Serum IGF bioactivity, measured in a porcine cartilage bioassay, was 0.18 and 0.55 U/mL in subjects 1 and 2; differences in bioactivity between subjects did not relate to serum IGF-II levels, but, rather, to differences in IGFBP-3 levels. Serum IGF bioactivity was not constant overnight and varied in a similar fashion in both subjects 1 and 2, with reduction in bioactivity between 0600-0800 h by 55% and 32%, suggesting the presence of inhibitory factors in the LTD serum; this decrease coincided with the rise in serum IGFBP-1 levels.(ABSTRACT TRUNCATED AT 400 WORDS)

Clinical features and endocrine status in patients with growth hormone insensitivity (Laron syndrome).

Twenty-seven patients with GH insensitivity were identified from 44 possible cases, using a scoring system based on height standard deviation score (SDS), basal GH, insulin-like growth factor-I (IGF-I), IGF-I response to IGF-I generation test, and GH-binding protein (GH-BP) determinations. The 27 cases were from 8 European countries and Australia. Clinical features were as follows: age 2.8-22.6 yr; 12 male, 15 female, 19 prepubertal. Birth weight was median -0.72 SDS (1.75(-)-3.29) and birth length, median -1.59 SDS (0.63(-)-3.63). Hypoglycemia had been documented in 33% of the cases, and micropenis was present in 58% of the males. At assessment, height was median -6.1 SDS (-3.8(-)-10.2), weight was median -3.2 SDS (-0.1 to -5.2), and percentage weight for height, median 111.3 (72-271). Puberty was absent in 2 boys aged 15 yr and in 3 girls aged 13 yr. Bone age was delayed in 19 of the 27 patients. Endocrine investigations showed basal serum GH median 17 micrograms/L (0.5-79), IGF-I values less than 5th percentile, and all except 2, age less than 8 yr, less than 0.1 percentile for age. Percentage increment of IGF-I during IGF-I generation test (hGH 0.1 U/kg body weight daily x 4) did not exceed twice the intraassay coefficient of variation, being less than 0.1 percentile for age. IGF-II was median 135.0 micrograms/L (62-232), all values being less than 5th percentile for age. Insulin-like growth factor binding protein-3 (IGFBP-3) values were median 0.53 mg/L (0.10-1.17 mg/L), all being less than 5th percentile for age. IGFBP-3 values after hGH remained less than 5th percentile. IGFBP-1 values showed the normal fall with age, some being above the normal range; IGFBP-2 values were normal. There was a positive correlation between height SDS and IGF-II SDS (r = 0.66, P < 0.001) and IGFBP-3 (r = 0.64, P < 0.001). Specific binding of [125I]hGH to GH-BP was undetectable in 18 patients and extremely low (< or = 5.6%) in 2. GH-BP was normal (14.2-45.9% radioactivity) in 7 subjects, all female, demonstrating that normal GH-BP does not exclude GH insensitivity.

The short-term effects of growth hormone therapy on height velocity and cardiac ventricular wall thickness in children with Noonan's syndrome.

Noonan's syndrome (NS) is associated with short stature and cardiac defects. Small studies reported linear growth increases with recombinant human GH (rhGH) therapy, but also raised concerns related to the anabolic effects of rhGH and the possible progression of ventricular hypertrophy. We report a multicenter study examining the efficacy and safety of rhGH (4 IU/m2.day, sc) in children with NS. Entry criteria were: NS confirmed by single observer, height SD score less than -2(UK Height Standards 1990), prepubertal, and normal maximal left ventricular (LV) wall thickness less than 1 cm by 2-dimensional echocardiography. Thirty subjects were recruited (19 males and 11 females), aged 8.9 +/- 0.5 yr (range, 4.8-13.7 yr). Growth was monitored for 12 months before and at 3-month intervals during therapy. Measurements of maximal LV wall thickness were taken at 0 and 12 months. Serum insulin-like growth factor I(IGF-I), IGF-II, and IGF-binding protein-3 levels were determined at 0, 3, 6, 9, and 12 months. Ten subjects with NS (4 females and 6 males), aged 8.8 +/- 0.7 yr (range, 6.3-11.8 yr), were monitored over the same period as a comparison group. In the treatment group, 27 subjects completed 12 months of therapy. Height SD score increased from -3.01 +/- 0.10 to -2.36 +/- 0.10 (P < 0.0001) after 12 months; height velocity (HV) increased from 4.9 +/- 0.2 to 8.9 +/- 0.3 cm/yr at 6 months and 8.1 +/- 0.4 cm/yr (P < 0.0001) from 6-12 months. The HV SD score increased from -0.7 +/- 0.15 to +2.42 +/- 0.32 over 12 months (P < 0.0001). The increase in HV was more than 2 cm/yr in 24 patients. IGF-I increased from 121 +/- 13 to 240 +/- 22 micrograms/L at 12 months (P < 0.0001), and IGF-binding protein-3 increased from 2.65 +/- 0.20 to 4.01 +/- 0.42 mg/L at 12 months (P = 0.0009). In the comparison group, there was no change in height SD score (-2.03 +/- 0.19), HV (4.4 +/- 0.24 CM/yr), or HV SD score (- 1.08 +/- 0.21). There was no increase in mean maximal LV wall thickness during the study in either the treatment group (12 month values were 0.63 +/- 0.02 cm at the mitral valve level and 0.66 +/- 0.02 cm at the papillary muscle level) or in the comparison group (0.63 +/- 0.04 cm at the mitral valve level and 0.61 +/- 0.03 cm at the papillary muscle level). In conclusion, rhGH was effective in 24 of the treated patients; these subjects achieved a significant increase in height SD score and HV over 1 yr. Abnormal anabolic effects of rhGH on myocardial thickness were not confirmed, and no patient developed features of hypertrophic cardiomyopathy.

Insulin and variation in glucose levels modify the secretion rates of the growth hormone-independent insulin-like growth factor binding protein-1 in the human hepatoblastoma cell line Hep G2.

The plasma level of the GH-independent insulin-like growth factor binding-protein-1 (IGFBP-1) is regulated inversely by insulin. In this study the effect of insulin and changes in the glucose concentration on in-vitro IGFBP-1 secretion by the Hep G2 cell line was studied. Media from confluent cells in 12 replicates were collected for consecutive periods: initial control (20 h), study (6 h) and recovery (20 h). Insulin suppressed IGFBP-1 secretion maximally at 100 mU/l (-32%) within 6 h. The secretion of IGFBP-1 was stimulated by a decrease in the glucose concentration in the medium, maximally (+25%) with a decrease from 24 to 6 mmol/l. Stimulation by varying glucose levels and suppression by insulin of IGFBP-1 secretion persisted on return to control conditions after the removal of physiological concentrations of glucose (4-12 mmol/l) and insulin (50-500 mU/l). The findings in the Hep G2 cell line that a variation in the physiological concentrations of glucose and insulin each independently regulate IGFBP-1 secretion suggest that this cell line may be a suitable model for further in-vitro studies of the regulation of secretion of IGFBP-1.

The induction of a specific protease for insulin-like growth factor binding protein-3 in the circulation during severe illness.

The insulin-like growth factors (IGF-I and IGF-II) are almost completely bound in the circulation to specific binding proteins (IGFBPs). These IGFBPs appear to play a pivotal role in maintaining circulating levels and modulating the delivery of the IGFs to the tissues. A large proportion of the circulating IGFs are bound with high affinity to one of the binding proteins. IGFBP-3. The mechanism by which these IGFs are transferred from the circulatory pool to the tissue receptors is at present unclear. Recent studies in late pregnancy have demonstrated the presence of specific proteases which may modify the IGFBPs such that their affinities for the IGFs are reduced. In this paper, we have demonstrated the presence of a heat-sensitive cation-dependent proteolytic enzyme specific for IGFBP-3 in the serum of five severely ill patients. The activity of this protease was found to vary in these patients, becoming more apparent during fasting than when studied after commencement of parenteral nutrition, indicating that one of the influencing factors in the activity of this protease is the nutritional intake of the patient. Age- and sex-matched healthy adults were also studied in a similar protocol, but no proteolytic modification of any of the IGFBPs was found in any of the samples examined. As the levels of both IGF-I and IGF-II were found to be low in the patients, the presence of a circulatory protease suggests that this may be an adaptive response to increase the bioavailability of the IGFs and possibly to improve the nitrogen retention and counter the catabolic state in severe illness.

The insulin-like growth factors and their binding proteins in a case of non-islet-cell tumour-associated hypoglycaemia.

Non-islet-cell tumours which induce hypoglycaemia are rare. Insulin-like growth factor-II (IGF-II) produced by some tumours is thought to be responsible for the hypoglycaemia and other systemic effects, despite normal or even low serum IGF-II levels. We studied a 44-year-old woman presenting with symptomatic hypoglycaemia associated with a large intraabdominal haemangiopericytoma. The serum IGF-II level was 455 micrograms/l when measured after acid-ethanol extraction (normal range (NR) 450-750 micrograms/l) and 1063 micrograms/l after acid chromatography (normal human serum pool 1068 micrograms/l). Levels of fasting plasma insulin, C-peptide, glucose and serum IGF-I levels were low before the operation (less than 2 mU/l (NR less than 2-14), 0.23 nmol/l (NR 0.4-1.2), 3.1 mmol/l, (NR 3.7-5.9) and 0.02 U/ml respectively). After tumour removal, the symptoms resolved rapidly and the patient made a full recovery. Secretion of both insulin and growth hormone was suppressed before the operation in response to a 75 g glucose meal and to an infusion of 100 micrograms GH-releasing hormone (GHRH) respectively in comparison with studies after the operation. Serum IGF-II levels 6 weeks and 12 weeks after the operation fell to 385 micrograms/l (777 micrograms/l; acid chromatography) and 280 micrograms/l (647 micrograms/l; acid chromatography) and serum IGF-I levels increased to 0.35 U/ml and 0.26 U/ml. Serum before the operation and tumour extract contained chiefly a large molecular weight precursor IGF-II (molecular weight 15,000-20,000) which disappeared from the serum after the operation. The IGF-binding proteins (IGFBP-1, IGFBP-2, IGFBP-3 and IGFBP-4) were examined.(ABSTRACT TRUNCATED AT 250 WORDS)

The induction of specific proteases for insulin-like growth factor-binding proteins following major heart surgery.

Insulin-like growth factors (IGF-I and IGF-II) circulate bound to specific high-affinity binding proteins (IGFBPs). Recent evidence has shown that in pregnancy and severe illness, specific proteases modify these binding proteins, reducing their affinity for IGFs. We have studied 12 patients, undergoing elective coronary artery vein-bypass graft surgery, for the appearance of these proteases and have demonstrated the induction of two independent, heat-labile, cation-dependent proteases. Proteolytic activity directed against IGFBP-3 was detected in all patients between 24 h and 5 days after surgery; the second IGFBP-4 specific protease was active 1 h after sternotomy. The total IGF-I levels were found to decrease following surgery, with the IGF-I distribution in the plasma being radically altered from that seen prior to the operation. One day after the operation the majority of the IGF-I, instead of being bound in the relatively inert 150 kDa complex, was associated with the smaller binding proteins which are more readily accessible to the tissues. These findings are in contrast to pregnancy where, despite similar proteases, the majority of the IGF-I remains in the 150 kDa complex. The alteration seen in IGF-I distribution after surgery did not appear to be a direct result of the IGFBP-3 proteolytic activity or an effect of the addition of heparin to the circulation. The potential increase in bioavailability of IGFs caused by the alteration in carrier protein may play a pivotal role in countering the catabolic state induced by surgery.

Girls with virilisation in childhood: a diagnostic protocol for investigation.

AIM: To analyse critically a protocol for the investigation of girls presenting with virilisation in childhood. METHODS: Twenty five girls aged 1.6-8.7 years with features of virilisation were evaluated. Twenty four had presented with pubic hair, eight with auxilliary hair, seven with facial acne, four with clitoromegaly, and 10 with tall stature. They underwent clinical assessment (height, weight, height velocity, staging of puberty, physical examination for acne, body odour, and clitoromegaly) and laboratory assessment comprising basal concentrations of cortisol, 17 OH-progesterone (17 OHP), androstenedione, dehydroepiandrosteronesulphate (DHEAS), testosterone, and oestradiol. The above steroids were also measured during the short synacthen test (0.25 mg intramuscularly) in 16 subjects and low dose dexamethasone suppression tests (0.5 mg at six hourly intervals over 48 hours). Pelvic ultrasound, computed tomography and magnetic resonance imaging of adrenals were carried out when the biochemical findings suggested that there might be an autonomous source of androgen secretion. RESULTS: Clinical and laboratory assessments differentiated the patients into three diagnostic categories: adrenarche (18 cases), congenital adrenal hyperplasia (five cases), and adrenocortical tumour (two cases). The last had elevated concentrations of DHEAS, 1.5 and 19.1 mumol/l (normal value < 0.5 mumol/l), androstenedione, 24.6 and 21.8 nmol/l (normal < 1 nmol/l), and testosterone, 4.5 and 2.4 nmol/l (normal < 0.8 nmol/l), with none suppressing on dexamethasone suppression. Congenital adrenal hyperplasia subjects had elevated basal serum concentrations of 17 OHP (n = 4): 250, 140, 14, and 14.1 nmol/l (normal < 10 nmol/l) and elevated peak values of 17 OHP after synacthen (n = 3): 76, 179.5, and 175 nmol/l. Adrenarche patients had elevated basal concentrations of DHEAS (median: 2.3 mumol/l; n = 17) and androstenedione (median 2.6 nmol/l; n = 17). Nine patients also had elevated basal serum testosterone concentrations (median 0.9 nmol/l). Peak values of 17 OHP after synacthen were significantly different from baseline (n = 12) and were < 50% of the lowest value in congenital adrenal hyperplasia. Serum DHEAS, androstenedione, and testosterone suppressed following dexamethasone suppression (n = 16), thereby distinguishing adrenarche patients from adrenal tumour patients. Clinical details did not distinguish patients, except for clitoromegaly which was present only in the tumour and congenital adrenal hyperplasia patients. CONCLUSIONS: This protocol proved useful and practical in cases of virilisation presenting particular diagnostic difficulty.

Construction of a community height surveillance programme: the Hackney growth initiative.

OBJECTIVES: To determine why, in the London Borough of Hackney before 1990, fewer children than expected were identified with remedial causes of short stature. To construct a practical model for height surveillance of 5 and 11 year old school entrants to improve the quality of child growth surveillance. SETTING: City and Hackney Borough, London, United Kingdom. METHODS: School nurses were trained by a clinical auxologist to measure children's height at school entry accurately and reproducibly. New procedures for measurement technique, plotting of data, referral, and audit were established. A reference manual was provided and a continuing training programme was started. RESULTS: During the first year the percentage of the target group measured was low. Changes in work practice led to improvements from 77% measured in the first year to 91% in the second year and 87% in the third year for 5 year olds. Improvements for 11 year olds were from 36% to 86% to 87% over the three years. Only 1.2% of 5 year olds and 2.6% of 11 year olds measured had height less than the third centile (compared with Tanner's height standards). CONCLUSIONS: School nurses measured height reliably. New audit procedures led to rapid changes in working practice and improvements in the percentage of children measured. The low numbers of short children previously identified with unrecognised abnormality may indicate an upward trend in height in this inner city population.

The geographic distribution of child abuse in an inner city borough.

The reported incidence of child abuse continues to rise. Present funding constraints make the introduction of child abuse intervention programs of proven effectiveness very difficult. This difficulty may lead to a change in direction away from population screening which looks for individuals at risk, towards the identification of geographic areas where child abuse is persistently concentrated (target areas). The method of plotting on a large scale map all registered cases of child abuse in an inner London borough is compared to three more complicated methods in its ability to accurately identify target areas. The geographic plotting of cases is favored because of its simplicity and our ability to update it annually as well as identify smaller areas such as housing estates.

The insulin-like growth factor I generation test in the investigation of short stature.

Genotypic and phenotypic heterogeneity in patients with growth hormone (GH) insensitivity syndrome suggests that partial defects exist in the GH receptor. The insulin-like growth factor I (IGF-I) generation test was assessed as a means of identifying partial GH receptor defects in a heterogeneous group of 22 prepubertal children with short stature. In a subgroup of nine patients with peak GH levels of 63.7 +/- 3.7 mU/l during a glucagon tolerance test, the response to the IGF-I generation test was no different from that for the group as a whole (peak GH, 43.3 +/- 4.5 mU/l), despite the fact that this subgroup exhibited a negative relationship between height SDS and peak GH and a positive relationship between height SDS and IGF binding protein-3. This preliminary study therefore suggests that the IGF-I generation test in its present form will not be useful as a primary screening test for partial GH insensitivity. Despite this, the IGF-I generation test has been extremely useful in the confirmation of the diagnosis of GHIS and may therefore also prove useful in the confirmation of partial defects in the GH receptor. A subgroup of short children with peak GH levels above 40 mU/l had some characteristics of partial GH receptor deficiency. These children, to whom GH therapy would not normally be given, may respond better to recombinant human IGF-I.

Improvement of diagnostic criteria in growth hormone insensitivity syndrome: solutions and pitfalls. Pharmacia Study Group on Insulin-like Growth Factor I Treatment in Growth Hormone Insensitivity Syndromes.

A survey to identify children and adolescents with primary growth hormone insensitivity syndrome (GHIS) yielded 38 patients who were positively identified using a scoring system that included five criteria: height, basal growth hormone (GH), GH binding protein, basal insulin-like growth factor I (IGF-I) and the increase of IGF-I after 4 days of GH administration (IGF generation test). Because of an overlap of the accepted and excluded groups with respect to points scored, an attempt was made to improve the scoring system. The new criteria were: height below -3 SDS, basal GH 4 mU/l or above, GH binding below 10%, basal IGF-I and basal IGF binding protein-3 (IGFBP-3) below the 0.1 centile for age, an increase of IGF-I in the IGF generation test less than 15 micrograms/l, and the increase of IGFBP-3 less than 0.4 mg/l. With this scoring system, a clear separation between the accepted and the excluded groups was obtained. IGFBP-3 was included to give the GH-dependent parameters of the IGF system more weight and because the accuracy of IGFBP-3 in the IGF generation tests was greater than the accuracy of IGF-I, when the group of patients with GHIS was compared with a group of patients with GH deficiency. Unexpectedly, the IGF generation test was unable to segregate both cohorts completely. In the GHIS-positive group, a significant correlation was found between basal IGF-I or IGFBP-3 levels corrected for age (SDS) and height SDS (r = 0.49, p < 0.002 and r = 0.61, p < 0.0001, respectively). There was also a significant correlation between the changes of IGF-I or IGFBP-3 in the IGF generation test and height SDS. That is, the patients with a slight response to GH were those with the least growth retardation, suggesting the existence of partial GH insensitivity.