Phenotypic variability in 17beta-hydroxysteroid dehydrogenase-3 deficiency and diagnostic pitfalls.

OBJECTIVE: 17beta-hydroxysteroid dehydrogenase type 3 isoenzyme (17beta-HSD3) is required to produce testosterone for male sex differentiation. Mutations in the HSD17B3 gene cause 17betaHSD3 deficiency and result in XY sex reversal of varying degree. We report the phenotypes of 14 subjects with 17betaHSD3 deficiency in relation to sex of rearing, androgen production, and HSD17B3 mutations. DESIGN: Cases were identified through the Cambridge Disorders of Sex Development Database where detailed clinical information was recorded, results of hCG stimulation tests were available, and HSD17B3 mutation was identified. RESULTS: Fourteen subjects from seven pedigrees (four consanguineous) had the following seven mutations: A56T, N130S, E215D, S232L, C268Y, V205E, and a novel mutation M197K. XY sex reversal was classified as complete in 10 infants at birth. Inguinal masses suggestive of androgen insensitivity syndrome (AIS) occurred in five infants. Contrasexual virilization reminiscent of 5alpha-reductase deficiency occurred in four subjects at puberty. The median (range) testosterone : androstenedione (T/A) ratio after a short hCG stimulation test was 0.32 (0.12-3.4). The S232L mutation identified in three affected family members caused isolated, severe hypospadias in one member who was raised male; virilization occurred despite in vitro studies showing an inactive mutant enzyme. Ratios of T/A in this pedigree were more than 0.8. CONCLUSION: XY sex reversal is sufficiently variable in 17betaHSD3 deficiency to cause problems in accurate diagnosis, particularly in distinguishing it from AIS. It should be considered in undervirilized male infants with normal Wolffian duct structures, absent Müllerian ducts, and normal adrenal steroid biosynthesis; or when an assigned female subject virilizes at puberty. Elevated hCG-stimulated T/A ratio may occur, and sex of rearing may not be concordant within affected families with the same HSD17B3 mutation. The T/A ratio, mutation analysis and functional analysis of the mutant enzyme taken in isolation, respectively, may not conclusively establish a diagnosis of 17betaHSD3 deficiency in undervirilized male subjects; the reasons for these discrepancies remain unknown.

The role of growth hormone in determining birth size and early postnatal growth, using congenital growth hormone deficiency (GHD) as a model.

OBJECTIVE: The role of GH in early human growth is unclear. Congenital GH deficiency (CGHD) provides a useful tool to explore this putative role. We have assessed the effects of CGHD on birth size and early postnatal growth, and the further impact of the presence of additional pituitary hormone deficiencies and midline brain defects on these parameters. DESIGN, PATIENTS AND MEASUREMENTS: Weight, length and BMI expressed as standard deviation scores (SDS), over the first two years of life, were retrospectively compared in 44 GH-deficient children (M:F 26 : 18). Thirty-eight of 44 patients underwent GH provocation testing and all patients had neuro-imaging of the brain. The patients were divided into three groups of increasing phenotypic complexity {group A [n = 12, isolated GHD, no midline defects], group B [n = 10, combined pituitary hormone deficiency (CPHD); no midline defects], group C (n = 22, CPHD with midline defects)}. RESULTS: Mean birth weight, length and BMI SDS were -0.4, -0.9 and +0.1 SDS, respectively. The differences were significant for weight (P = 0.03) and BMI (P = 0.003), but not length (P = 0.3) SDS, between groups A and C. Of the three groups, group A had a lower weight and BMI SDS than group C. The prevalence of postnatal complications (n = 25) was significantly different in the three groups [group A (8%), group B (80%), group C (73%); P < 0.001] and particularly between patients with isolated GH deficiency (IGHD) (group A) and CPHD (groups B and C; P < 0.0001). No patients in group A presented with neonatal hypoglycaemia as compared with 70% of those in group B and 59% in group C (P = 0.001). A reduced length SDS was observed in all patients within 6 months of birth and the reduction was greatest in group B (P = 0.03). Group C remained significantly (P < 0.05) heavier at 12, 18 and 24 months compared to group A. BMI SDS was significantly (P < 0.05) greater at all study points in CPHD patients (groups B and C) as compared with IGHD. Serum GH concentrations at testing did not correlate significantly with birth length (r = -0.08, P = 0.7), birth weight (r = -0.08, P = 0.6) or the age at induction of GH treatment (r = 0.12, P = 0.5). There were no significant differences between peak serum GH concentrations in patients in groups A (7.8 +/- 6.3 mU/l), B (3.9 +/- 4.8 mU/l) or C (8.7 +/- 5.4 mU/l). CONCLUSIONS: Length, weight and BMI data from our study groups suggest that GH per se has a minimal effect on intrauterine growth but a significant effect during the infancy period. Early growth may also be influenced by the complexity of the hypopituitary phenotype reflected by the presence of additional pituitary hormone deficiencies and midline forebrain defects.

Is the thyrotropin-releasing hormone test necessary in the diagnosis of central hypothyroidism in children.

To determine the value of the TRH test, we analyzed the unstimulated serum T(4) and TSH concentrations in 54 children with central hypothyroidism. A TRH test was performed in 30 patients. Midline brain defects (septo-optic dysplasia, 28; holoprosencephaly, 2) and combined pituitary hormone deficiencies were present in 30 and 52 patients, respectively. The mean serum free T(4), total T(4), and basal TSH concentrations were 0.6 ng/dl, 4.0 microg/dl, and 2.8 microU/ml, respectively. Five patients demonstrated elevated basal serum TSH concentrations. A normal TRH test [increase (delta) in TSH, 4.5-17.8], based on data from 30 controls, was documented in 23.3% of patients. Brisk (deltaTSH, >17.8), absent/blunted (deltaTSH, <4.5), and delayed responses were documented in 16.7%, 30%, and 30% of patients, respectively. The mean age at diagnosis was 2.8 yr, with 8 patients evolving into TSH deficiency. It was not possible to differentiate patients as having pituitary or hypothalamic disease based solely on the TRH test results. Patients with septo-optic dysplasia were diagnosed earlier and had elevated basal serum TSH and PRL concentrations, diabetes insipidus, and evolving disease. Although full pituitary function assessment is mandatory to identify combined pituitary hormone deficiencies, a TRH test is not essential, and the diagnosis should be made by serial T(4) measurements.