Effects on growth and metabolism of growth hormone treatment for 3 years in 36 children with Prader-Willi syndrome.

BACKGROUND/AIMS: Prader-Willi syndrome (PWS) is a complex genetic disorder whose many manifestations include obesity and short stature. Diabetes, osteoporosis, and scoliosis are common. We evaluated the effects of human growth hormone (hGH). METHODS: A prospective cohort study of 36 children (1-15 years of age) with genetically confirmed PWS who were given hGH (mean dose 0.033 ± 0.006 mg/kg/day) for 36 months. At baseline and once yearly, we evaluated growth, insulin-like growth factor-1 (IGF-1), body composition, bone mineral density (BMD), glucose tolerance, serum lipids, and spinal radiographs. RESULTS: Height gain over the 3-year period was 1.2 SD score. Lean body mass increased significantly during each treatment year. Total body fat decreased by 5.42 and 1.17% in the 1st and 2nd years, respectively. BMD remained unchanged during therapy. IGF-1 and homeostasis model assessment index of insulin resistance increased, and glucose intolerance was found in 22.7% of patients at baseline and 0% at 3 years. None of the patients had diabetes. Their lipid profile improved. Scoliosis was present in 27.8% of the patients at baseline and 47.2% at 3 years. CONCLUSION: GH treatment in children with PWS has multiple beneficial effects on growth and body composition. Tolerance is good, with an improvement in glucose metabolism, although IGF-1 levels and insulin resistance parameters should be monitored closely. The high rate of scoliosis warrants monitoring by a pediatric orthopedic surgeon.

Bone markers after total body irradiation in childhood.

Total body irradiation (TBI) can cause short stature because of decreased growth hormone (GH) and skeletal abnormalities. To evaluate the plasma concentrations of markers of bone formation (osteocalcin and procollagen type 1 amino-terminal propeptide, P1NP) and resorption (carboxy-terminal telopeptide, CTX), in patients (n=65) who had been given TBI at 6.6+/-0.4 years were evaluated at 9.8+/-0.4 years. Patients given single 10 Gy or fractionated 12 Gy TBI had similar characteristics, except that plasma insulin-like growth factor (IGF-1) was lower in those given a single 10 Gy. Seven had lower osteocalcin and two had higher CTX than controls. Bone markers (as zs) were positively correlated (osteocalcin with P1NP, rho=0.42, P=0.0007; osteocalcin with CTX, rho=0.3, P<0.02), but not P1NP with CTX. Plasma osteocalcin and CTX were also positively correlated with plasma IGF-1, but not with growth rate during the first year on GH (n=28). Adult height was -2.5+/-0.2 s.d.s. (n=49). Those irradiated when young (P=0.0002) or given single TBI lost more height between TBI and adult height. Most TBI patients had normal bone formation and resorption markers. Thus, impaired bone turnover is probably not the cause of their short stature and poor response to GH.

Familial membranous glomerulopathy, toxic exposure and/or genetic sensibility?

Membranous nephropathy rarely occurs as a familial disease. We report two siblings (brother and sister) who presented with nephrotic syndrome and many vascular complications. HLA identities and potential toxic exposure may be concurring in these cases.

Serum procollagen type 1 amino-terminal propeptide (P1NP) as an early predictor of the growth response to growth hormone treatment: Comparison of intrauterine growth retardation and idiopathic short stature.

UNLABELLED: There is no way to predict early the growth response to growth hormone (GH) treatment in short children with intrauterine growth retardation (IUGR) or idiopathic short stature (ISS). OBJECTIVE: To evaluate the capacity of the procollagen type 1 amino-terminal propeptide (P1NP), a new marker of bone formation, to help in this prediction. PATIENTS AND METHODS: Longitudinal study of 30 patients treated at 7.7 (range: 2.2-12.5) years for IUGR (n=16) or ISS (n=14) with GH (0.47 and 0.33 or 0.4mg/kg/week respectively). P1NP and insulin-like growth factor I (IGF I) were measured before and after 3-6 months of GH treatment. RESULTS: Before treatment, IUGR patients were younger and shorter than ISS patients, but their other characteristics were similar. IGF I Z-score (ZS) and P1NP concentrations were positively correlated in the whole population (Rho=0.48; P=0.01). After 3-6 months of treatment, both concentrations increased in IUGR and ISS (P<0.01). They remained correlated only in ISS (Rho=0.54; P<0.05). P1NP before treatment was negatively correlated (Rho=-0.67, P=0.015) with the growth rate (SD) during the first year of treatment in ISS but not in IUGR; IGF I ZS was not. The changes in P1NP for the whole population over 3-6 months, but not the changes in IGF I ZS, were positively correlated with the growth rate (Rho=0.41, P=0.03). CONCLUSIONS: Lower basal plasma P1NP concentrations predict better growth response to GH treatment during the first year in ISS children. Greater increases in its concentrations after 3-6 months of GH treatment may also predict a better growth response in both ISS and IUGR.

Final height and gonad function after total body irradiation during childhood.

Short stature and gonad failure can be a side effect of total body irradiation (TBI). The purpose of the study was to evaluate the factors influencing final height and gonad function after TBI. Fifty young adults given TBI during childhood were included. Twenty-seven had been treated with growth hormone (GH). Those given single 10 Grays (Gy) or fractionated 12 Gy TBI had similar characteristics, GH peaks, final heights and gonad function. After the end of GH treatment, 11/20 patients evaluated had GH peak >10 microg/l. Final height was <-2s.d. in 29 (58%). The height loss between TBI and final height (2.4+/-1.1 s.d.) was greater in those who were younger when irradiated (P<0.0001). When the GH-treated and -untreated patients were analyzed separately, this loss was correlated with the age at TBI at 4-8 years for the GH-treated and at 6-8 years for the untreated. Boys showed negative correlations between testicular volume and plasma follicle-stimulating hormone (FSH, P=0.0008) and between plasma FSH and inhibin B (P=0.005) concentrations. We concluded that the indications for GH treatment should be mainly based on the age at irradiation, taking into account the GH peak. The plasma FSH and inhibin B concentrations may predict sperm function.

[Central precocious puberty in girls: prediction of the etiology]. / Pubertés précoces centrales des filles: prédiction de l'étiologie.

Precocious puberty (PP) is defined in girls by the occurrence of pubertal development before the age of 8. This development raises 3 questions: 1) Is it abnormal puberty or variant of the normal? 2) If abnormal puberty, is it of central, hypothalamic-pituitary, or peripheral, ovarian or adrenal origin? 3) If central, is it idiopathic or due to a lesion, and is there indication to treat it? The PP in a girl with no previous medical history is usually of central and idiopathic origin. However, isolated central PP may reveal a CNS lesion, particularly an optic glioma with its risk of blindness. Two independent predictors of CNS lesion are the age at PP onset of less than 6 years old, and increased plasma estradiol concentration. The selection of the girls for neuroradiological imaging should be based on these two parameters. However, neuroradiological imaging remains necessary until the prospective confirmation of their predictive value.

Management of boys with short stature and delayed puberty.

OBJECTIVE: To review the management of boys with short stature and delayed puberty and the testosterone priming protocol. METHODS: In 148 boys aged > 14 years seen for height < -2 SDS and constitutional delayed puberty we evaluated growth hormone (GH) secretion and final height (80 boys). RESULTS: The GH peak was < 10 microg/l after arginine-insulin tests performed with testosterone heptylate priming in 8/32 (25%) and without in 62/153 (41%), including first and second evaluations. It was low in 7/11 boys given 2 x 100 mg testosterone (14.7 +/- 1.7 microg/l) and in 1/21 given 4 x 100 mg (21.3 +/- 2.0 microg/l, p = 0.04). It was low during sleep in 4/29 (14%) boys, all having basal plasma testosterone below 3.5 nmol/l. The basal insulin-like growth factor (IGF)-I concentration was below -2 SDS in 22% of the boys evaluated. Final height was -0.8 +/- 0.1 SDS. It was similar in those with low (n = 9) and normal (n = 71) GH peak, and in those treated (n = 22) or untreated (n = 58) with testosterone. It was over 1 SDS lower than the target height in 20% and than the predicted height at the initial evaluation in 14% of the boys. Pubertal growth was not correlated with the GH peak or plasma IGF-I. CONCLUSIONS: The GH peak during the sleep is more frequently normal than the peak after stimulation. The number of testosterone doses influences the quality of priming. The medical problems involved in treating boys with delayed puberty are excluding disease and deciding on testosterone treatment.

Idiopathic male pseudohermaphroditism: variations in presentation and management.

UNLABELLED: Male pseudohermaphroditism (MPH) is the abnormal development of genitalia in an individual with a 46,XY chromosome complement and testicular tissue. The etiology of MPH is unknown in most cases, which are defined as idiopathic. OBJECTIVE: To analyze the data for cases of idiopathic MPH. PATIENTS AND METHODS: A retrospective study of 29 patients with idiopathic MPH and no uterus. RESULTS: Four patients had a family history of abnormal sexual development and five had low birth weight. The initial manifestations were sexual ambiguity (26), microphallus and hypospadias (2), and primary amenorrhea (1). Basal and/or stimulated testosterone concentrations showed insufficient testosterone secretion in three patients. Genitography showed a vagina in 13 patients. Male genitoplasties were performed on 21 out of the 24 patients reared as males and female genitoplasties on five patients. Histological studies of the gonads of these showed streak gonads in one, normal gonads in one and signs of testicular dysgenesis in three others. Molecular studies on the SRY gene (17) showed no mutation. CONCLUSIONS: Idiopathic male pseudohermaphroditism is a heterogeneous condition, even within families with a history of this condition. We propose a set of guidelines for the management of these patients.

Congenital gonadotropin deficiency in boys: management during childhood.

OBJECTIVE: To analyze the features of boys with congenital gonadotropin deficiency (CGD), and to determine the value of plasma inhibin B and anti-Müllerian hormone (AMH) for predicting testicular function and the effect of testosterone treatment. PATIENTS: We followed 19 boys for CGD, including five with Kallmann syndrome. RESULTS: The boys were seen before 14 years of age for micropenis (9 boys) or later for delayed puberty (10 boys). No testis was palpable in the scrotum in 13 patients, bilaterally in seven of them. Luteinizing hormone (LH) peak after a gonadotropin releasing hormone (GnRH) test was between 0.5 and 5.6 U/l. Plasma inhibin B was low in the four patients evaluated at less than 1 year old. AMH was low in one of them and normal in four others. Of the older patients, three lad low plasma inhibin B and four had normal concentrations; plasma AMH was low in three of them and increased in four. Testosterone treatment restored penis length to normal in all patients. CONCLUSIONS: Low plasma inhibin B and AMH concentrations may indicate testicular damage in boys with CGD.

Extreme short stature after intrauterine growth retardation: factors associated with lack of catch-up growth.

UNLABELLED: The factors associated with lack of catch-up growth after intrauterine growth retardation (IUGR) are unknown. OBJECTIVE: To identify these factors by analyzing the clinical features and growth hormone (GH)-insulin-like growth factor I (IGF-I) axis. METHODS: 95 patients with height <-3 SD after IUGR were assigned to group 1 without (n = 50) or group 2 with (n = 45) malformations. Twenty-one in group 1 and 19 in group 2 were treated with GH. RESULTS: They were seen at 5.3 +/- 0.5 and 4 +/- 0.5 year (p = 0.02) with heights at -3.4 +/- 0.1 and -3.9 +/- 0.2 SD (p = 0.03). Group 1 differed from group 2 in having a lower frequency of consanguinity (2 vs. 28.9%, p < 0.001), and higher frequencies of target heights (26.5 vs. 6.7%, p = 0.02) and mothers' heights (34.7 vs. 8.9%, p < 0.01) <-2 SD, multiparity (26 vs. 8.9%, p < 0.05), prematurity (36 vs. 15.5%, p < 0.05) and cesarean section birth (42 vs. 17.8%, p = 0.01). The GH-IGF-I axis data and the height increases after 3 years of GH treatment (1.6 +/- 0.2 in group 1 and 1.1 +/- 0.3 SD in group 2) were similar. CONCLUSION: The short height of the parents, particularly of the mother, is associated with factors limiting the catch-up growth after IUGR of children without malformations, while the high frequency of consanguinity in those with malformations suggests that transmitted fetal factors affect organogenesis or development.

Extremely short stature: influence of each parent's height on clinical-biological features.

UNLABELLED: Idiopathic extremely short stature probably has several causes. OBJECTIVE: To evaluate the influence of each parent's height on clinical-biological features. METHODS: 57 patients without intrauterine growth retardation seen at 7.9 +/- 0.4 years for height < or = -3 SD were classified according to the difference between their target height and actual height: < 2 SD in familial short stature (FSS, n = 28) and >2 SD in non-FSS (n = 29). RESULTS: Height decreased from -0.5 +/- 0.1 SD at birth to -2 +/- 0.2 SD at 1 year and -2.7 +/- 0.1 SD at 3 years, but the changes in the two groups were similar. FSS children were shorter than non-FSS children both at birth (p = 0.03) and as adults after growth hormone (GH) treatment (p < 0.05), but their plasma insulin-like growth factor I concentrations and GH peaks were similar. The FSS children fathers' heights were more frequently below -2 SD (64%) than the mothers' heights (35%) and were correlated with height at first evaluation (p < 0.05). For the whole population, the mothers' heights were correlated with birth weight (p < 0.05) and with height at first evaluation (p < 0.03). CONCLUSION: This study confirms the influence of the mother's height on birth weight and shows how of the father's height influences idiopathic extremely short stature.

Follow-up of 68 children with congenital adrenal hyperplasia due to 21-hydroxylase deficiency: relevance of genotype for management.

The phenotype of congenital adrenal hyperplasia (CAH) varies greatly. The purpose of this study was to optimize diagnosis and follow-up by comparing phenotype with genotype. Sixty-eight patients with CAH due to 21-hydroxylase deficiency were studied by clinical, hormonal, and molecular genetic methods. Patients were classified according to predicted mutation severity: group 0, null mutation (17.6%); group A, homozygous for IVS2 splice mutation or compound heterozygous for IVS2 and null mutations (33.8%); group B, homozygous or compound heterozygous for I172N mutation (14.7%); group C, homozygous or compound heterozygous for V281L or P30L mutations (26.5%); and group D, mutations with unknown enzyme activity (7.4%). All group 0 and A patients had the salt-wasting form, and group C had nonclassical forms. Group B included five salt-wasting and five simple virilizing forms. Groups 0 and A were younger at diagnosis (P < 0.02), and females were more virilized than those in group B. Group B had higher basal plasma 17-hydroxyprogesterone (564 +/- 162 nmol/liter) and testosterone (11 +/- 3 nmol/liter) levels than group C [59 +/- 13 nmol/liter (P < 0.001) and 1.4 +/- 0.2 nmol/liter (P < 0.005), respectively]. Hydrocortisone doses given to groups 0, A, and B were similar at all ages, but lower in group C (P < 0.01). Final height was below target height in classical (n = 16; -2 +/- 0.2 SD score; P < 0.02) and nonclassical (n = 11; -1.2 +/- 0.4 SD score; P < 0.03) forms. The severity of the genetic defects and the clinical-laboratory features are well correlated. Genotyping, combined with neonatal screening and optimal medical and surgical treatment, can help in the management of CAH.

Assessing short-statured children for growth hormone deficiency.

AIM: To optimize the workup of short-statured children by defining the most appropriate tools for diagnosing growth hormone (GH) deficiency. METHODS: Patients were assigned to prepubertal (n = 113) or pubertal (n = 112, including 25 boys primed with testosterone) age groups. Mean plasma GH concentration during sleep, GH peak after provocative test, and insulin-like growth factor I (IGF-I) were measured in a single evaluation. RESULTS: The mean GH concentration during sleep was more often normal (n = 155) than the GH peak after provocative tests (n = 105) or the IGF-I concentration (n = 88). Prepubertal patients with a normal body mass index (BMI) had mean GH concentrations during sleep that correlated positively with height, growth rate, GH peak after provocative tests, and IGF-I (p < 0.0005 for all) and negatively with the difference between target and patient heights (p = 0.01) and BMI (p < 0.05). Pubertal patients with a normal BMI had a mean GH concentration during sleep that correlated positively with GH after provocative tests (p < 0.0001) and IGF-I (p < 0.005). Mean GH concentration during sleep and IGF-I concentration for boys primed with testosterone were more often normal (n = 23) than the GH peak after provocative tests (n = 14). All 9 patients with pituitary stalk interruption had low IGF-I concentrations; 1 patient had a normal GH peak after provocative test, and 2 patients had normal mean GH concentrations during sleep. CONCLUSIONS: Measuring the GH concentration during sleep and priming boys with pubertal delay can help to exclude idiopathic GH deficiency. Magnetic resonance imaging is needed to exclude anatomic abnormalities when GH and/or IGF-I concentrations are low.

Factors influencing the growth hormone response to growth hormone-releasing hormone in children with idiopathic growth hormone deficiency.

OBJECTIVE: To evaluate the factors influencing the growth hormone (GH) response to GH-releasing hormone (GHRH) test in idiopathic GH deficiency. METHODS: 28 patients aged 4.9 +/- 0.7 years with certain GH deficiency were given GHRH (2 microg/kg). RESULTS: The GH peak after GHRH was correlated negatively with age at evaluation (r = -0.37, p < 0.05) and body mass index (r = -0.44, p = 0.02), and positively with anterior pituitary height (r = 0.47, p = 0.02), GH peak after non-GHRH stimulation (r = 0.78, p < 0.0001) and spontaneous GH peak (r = 0.82, p = 0.007). It was lower in the patients aged >5 years than in the youngest (p = 0.04), but it was similar in the patients with and without features suggesting a hypothalamic origin. CONCLUSION: The GH response to GHRH test cannot be used to differentiate between hypothalamic and pituitary forms of idiopathic GH deficiency, probably because the GH response decreases after the first 5 years of life, whatever the origin of the deficiency.

Growth hormone deficiency: permanence and diagnosis in young adults.

OBJECTIVE: To optimize the tools for diagnosing idiopathic growth hormone (GH) deficiency. METHODS: We compared the data of 43 young adults treated for GH deficiency before and after GH treatment and puberty. Those with organic lesions were assigned to group 1 (n = 9), those with certain GH deficiency (n = 11) to group 2 and those with no criterion of certitude of GH deficiency to group 3 (n = 23). RESULTS: Group 1 patients: the GH peaks at first [1.5 +/- (SE) 0.4 microg/l] and second (1.9 +/- 0.7 microg/l) evaluations before treatment were similar to those at the third evaluation (1.2 +/- 0.8 microg/l) after treatment. Group 2 patients: they had similar peaks (2.6 +/- 0.8, 2.9 +/- 0.5 and 5.5 +/- 1.4 microg/l). Group 3 patients: the peaks increased from 4.9 +/- 0.4 and 4.8 +/- 0.4 to 18.4 +/- 2.3 microg/l (p < 0.0001); 87% had a GH peak >10 microg/l at this evaluation. The plasma insulin-like growth factor 1 was initially below -2 z-score in 12/13 of these patients and similarly low in 4/17 patients at the third evaluation. The growth rates of the three groups before and their increase during the 1st year of treatment were similar. CONCLUSION: Almost all patients with GH deficiency before puberty without criteria of certitude had a normal GH peak after puberty. Some of these patients probably had a transiently low GH secretion.

Adult height in advanced puberty with or without gonadotropin hormone releasing hormone analog treatment.

Advanced puberty is defined as the onset of puberty in girls at 8-10 years of age and in boys at 9-11 years. This study analyzes adult height in 57 children with advanced puberty to evaluate the results of treating children (9 girls and 8 boys) with gonadotropin hormone releasing hormone (GnRH) analog and the impact of advanced puberty on adult height in untreated children (31 girls and 9 boys). For treated girls, adult height predicted at the onset of treatment (151.9+/-1.7 cm) was similar to the final adult height (155.3+/-1.4 cm), but lower than target height (157.2+/-1.6 cm, p = 0.04). For untreated girls, adult height predicted at the initial evaluation (156.7+/-1 cm) was also similar to adult height (157+/-1 cm), but lower than the target height (157.6+/-1 cm, p = 0.03). The adult heights of both treated and untreated girls were similar to their target heights. For treated boys, adult height predicted at the onset of treatment (173.2+/-3.1 cm) was greater than the final adult height (164.1+/-2.1 cm, p = 0.01), which was lower than target height (170.4+/-1.2 cm, p = 0.01). For untreated boys, adult height predicted at the initial evaluation (170.8+/-2.7 cm) was similar to both the adult height (169.1+/-1.9 cm) and target height (170.2+/-1.2 cm). Height gains between the onset of puberty and adult height were similar in treated (29.9+/-2.3 cm in girls and 29.8+/-1.7 cm in boys) and untreated (28.6+/-1 and 33.1+/-2 cm) children. When expressed as SD, the adult height was significantly shorter than that at 4 years in treated girls (difference 1 SD, p = 0.03), in untreated girls (difference 0.9 SD, p = 0.0002) and in treated boys (difference 0.9 SD, p = 0.02), but it was similar to that in untreated boys. Adult height was below target height by >5 cm in seven girls (two of them treated) and five boys (four of them treated). In conclusion, treating advanced puberty did not change the adult height reached by girls, and was associated with reduced growth potential in boys. The adult heights of untreated children were similar to those predicted at the initial evaluation and to target heights, but in girls they were 1 SD lower than the height at 4 years. These data suggest that advanced puberty decreases the growth potential by about 5 cm, and that GnRH analog treatment does not prevent this.

GH deficiency caused by cranial irradiation during childhood: factors and markers in young adults.

Cranial irradiation alters hypothalamic-pituitary function. We reevaluated 90 patients with GH deficiency caused by fractionated cranial irradiation performed at age 4.9 +/- 0.4 (SE) yr when they were 15.7 +/- 0.2 yr old. Group 1 received 18 Grays (Gy) (7 cases) or 24 Gy (21 cases) for acute lymphoblastic leukemia; group 2, 30-40 Gy for medulloblastoma (22 cases); group 3, 45-60 Gy for optic glioma and various tumors (30 cases); and group 4, 40-50 Gy for retinoblastoma (10 cases). The mean GH peaks after an arginine insulin test in group 3 (1.9 +/- 0.4 microg/liter) was lower than in groups 1 (4.8 +/- 0.5 microg/liter, P < 0.001) and 2 (3.4 +/- 0.5 microg/liter, P < 0.03). The mean plasma IGF-I concentrations in group 3 [-3.8 +/- 0.2 z score (zs)] was lower than in groups 1 (-2.4 +/- 0.3 zs, P < 0.001) and 2 (-3.1 +/- 0.2 zs, P < 0.02), as was the mean in group 4 (-3.9 +/- 0.3 zs, P < 0.01 compared with group 1 and P < 0.05 compared with group 2). GH peaks and IGF-I were correlated positively (P = 0.0001) and negatively with dose (P < 0.001 for GH and P = 0.0001 for IGF-I), but not with age at irradiation. Among the 43 patients with GH peaks below 3 microg/liter, 41 (95%) had plasma IGF-I less than -2 zs. The body mass index (BMI), plasma insulin, and leptin were similar in the four groups. They were positively correlated with each other (P < 0.001 for BMI compared with insulin and with leptin, respectively, and P < 0.01 for insulin compared with leptin), but not with age or dose of irradiation, or with markers of GH secretion. In conclusion, in patients with GH deficiency caused by cranial irradiation, the residual GH secretion and plasma IGF-I depend on the dose. Almost all the patients with severe GH deficiency had low plasma IGF-I. BMI, leptin, and insulin seem to be independent of GH status.

Factors affecting gonadal function after bone marrow transplantation during childhood.

Conditioning for bone marrow transplantation (BMT) may alter viability of germ cells and production of gonadal hormones. We analyzed the risk factors for gonadal failure after 12 Gy total body irradiation (TBI) given as six fractions (n = 31, group 1), 10 Gy (one dose) TBI (n = 20, group 2), 6 Gy (one dose) total lymphoid irradiation (TLI, n = 17, group 3) and chemotherapy alone (n = 7, group 4), given at 7.7 +/- 0.4 (0.6-13.6) years. Among the 34 girls, seven (20.6%) had normal ovarian function with regular spontaneous menstruation and normal plasma follicle-stimulating (FSH) and luteinizing (LH) hormones, five (14.7%) had partial ovarian failure with regular menstruation but increased FSH and/or LH, and 22 (64.7%) had complete ovarian failure. The 24 girls with chronological and bone ages >13 years included similar percentages, with increased FSH or LH in all four groups. There was a positive correlation between age at BMT and FSH (r = 0.54, P < 0.01), but not with lh, and between fsh and lh (r = 0.8, P = 0.0003). Plasma FSH concentrations had returned to normal spontaneously in six cases, and those of LH in two cases. Among the 41 boys, 16 (39%) had normal testicular function and 25 (61%) had tubular failure and increased FSH. Of these, 10 also had Leydig cell failure (three complete and seven partial). The 18 boys with chronological and bone ages >15 years included similar percentages with increased FSH or LH in groups 1 to 3, and testicular volume was significantly lower in group 2 than in group 3 (P = 0.008). There was no correlation between age at BMT and FSH, LH or testosterone, but there was a negative correlation between FSH and inhibin B (rho = -0.87, P < 0.003). we conclude that girls are more likely to suffer ovarian failure the older they are at bmt, and that early ovarian recovery is possible. the negative correlation between fsh and inhibin b in boys suggests that this parameter is an additional indicator of tubular function.