ESPE Abstracts (2019) 92 FC9.4

Prenatal Environment and Genetic Background Influence Urinary Steroid Excretion in Monozygotic Twins with Intra-Twin Birth-Weight Differences

Sandra Schulte1, Joachim Woelfle1, Felix Schreiner1, Charlotte Kasner1, Mathias Gruenewald1, Michaela F. Hartmann2, Stefan A. Wudy2, Peter Bartmann3, Bettina Gohlke1


1Children's University Hospital Bonn, Dept. Paediatric Endocrinology and Diabetology, Bonn, Germany. 2Center of Child and Adolescent Medicine, Justus Liebig University Giessen, Steroid Research and Mass Spectrometry Unit, Paediatric Endocrinology and Diabetology, Giessen, Germany. 3Children's University Hospital Bonn, Dept. Neonatology, Bonn, Germany


Background: Low birth-weight and unfavourable intrauterine conditions are associated with a subsequent impact on the endocrine system. Many studies reported hyperandrogenaemia and precocious adrenarche in children born small for gestational age (SGA). However, little information is available on steroid profiles in these subjects.

Objective and Hypotheses: We followed genetically identical twins with intra-twin birth-weight differences from birth until adolescence to objectify the impact of a lower birth-weight on development and health in later life.

Method: 68 monozygotic twin pairs where included in this study. Birth-weight difference of <1SDS was defined concordant (n=41, 18 female), birth-weight difference ≥1SDS discordant (n=27, 15 female). Spontaneous urine samples were collected at a mean age of 14.9 yrs and gas chromatography–mass spectrometry (GC-MS) was used to analyse urinary steroid profiles (36 metabolites).

Results: We found pronounced intra-twin correlations in steroid metabolites – nearly all measured metabolites were significantly correlated within the twin-pairs. In the concordant twin-pairs all but three (92%) were significantly correlated (30 metabolites correlated highly significant with P<0.01, and three significantly; p≤0.05). In the discordant group, 28 out of 36 metabolites (78%) were significantly correlated (27 highly significant; P<0.01, one significantly; p≤0.05). In the concordant group with birth-weight differences of < 1 SDS no significant differences in urinary steroid metabolites were detected. In contrast to this, we found significant intra-twin differences for two DHEAS metabolites in the twins with marked birth-weight differences of more than 1 SDS ("discordant group"): 5-Androstene-3β,16α-diol-17-one (P=0.009, 822.24 vs 737.83 µg/l) and 5-androstene-3β,16α,17β-triol (androstenetriol-16α) (P=0.038, 489.45 vs 437.51 µg/l). All former smaller twins showed higher concentrations of DHEAS-metabolites than their larger co-twins.

Conclusion: In this special group of monozygotic twins with intra-twin birth-weight differences, we could show that birth-weight has a long-lasting impact on steroid profiles. We detected significant differences regarding DHEAS-metabolites with higher concentrations in the former smaller twins and fewer significant intra-twin correlations regarding all metabolites analysed. However, most metabolites analysed in both groups showed highly significant intra-twin correlations, suggesting a major genetic determination of steroid hormone concentrations.