Main

Intrauterine growth retardation is known to be associated with hypoplasia of the fetal adrenal zone and with lower fetal serum concentrations of DHEAS(1, 2). The involution of the fetal adrenal after birth is reflected by a virtual disappearance of DHEAS from the circulation until adrenarche.

Adrenarche occurs exclusively in primate species. This phenomenon correlates histologically with the appearance and growth of the zona reticularis, within the adrenal cortex, and seems to be independent of pubertal development. Endocrine hallmarks of adrenarche are increasing circulating levels of DHEA, DHEAS, and androstenedione. DHEAS has the longer half-life in the circulation (8-11 h) and does not exhibit a circadian rhythm(3–5). DHEAS has been identified as a highly specific marker of the individual hormonal milieu in the adult(7).

The regulation of DHEAS secretion is still poorly understood. ACTH availability appears to be a prerequisite. Other possible regulators of adrenal androgen secretion are prolactin, dopamine, IGF-I, and a hitherto unidentified adrenal-androgen-stimulating hormone of pituitary origin. Nutritional state and intercurrent illness are also known to influence plasma DHEAS levels(3–8).

Recently, Barker et al.(9) reported increased rates of cardiovascular disease and non-insulin-dependent diabetes in adults born SGA and launched the concept of “programming” in fetal life. In this concept, the growth-retarded fetus adapts to undernutrition and survives by altering some endocrine and metabolic set points, which appear to remain altered postnatally(9, 10).

We postulated that fetal growth may be a modulator of adrenarche and tested this hypothesis by examining serum DHEAS concentrations in pairs of children who were born as discordant siblings from the same pregnancy.

METHODS

Thirteen pairs of discordant siblings, who were born in our hospital from twin (n = 8), triplet (n = 4), or quadruplet (n= 1) pregnancies were available for examination (after informed consent of both parents) at a median age of 8.2 y (range 5.8-16.0 y). At birth, one of each pair was SGA with a weight or length ≤ 2 SD(11), and the other was of appropriate weight (AGA). Birth weight of the smallest infant was a median 67% (range 33-80%) of that of the largest sibling.

Clinical examination of the children included measurement of height and weight, and pubertal staging according to Tanner. None of the children presented evidence of a systemic or syndromic disorder.

Serum DHEAS concentrations of SGA children and their AGA siblings were measured. In triplets and quadruplets, serum DHEAS of the siblings with the lowest and highest birth weight were compared. Serum levels of DHEAS were measured by RIA using the DSL DHEAS Radioimmunoassay Kit (Diagnostic Systems Laboratories, Webster, TX).

The intraassay coefficient of variation was 5.1% at 0.6 mmol/L and 4.1% at 6.1 mmol/L. The detection limit was 0.05 mmol/L. Reference ranges of DHEAS during childhood and adolescence were obtained from Reiter et al.(12). Paired t test was used for statistical comparison.

RESULTS

Clinical data and serum DHEAS concentrations are summarized in Table 1 and Figure 1. In all 10 pairs with similar weight (≤1 SD difference) at the time of sampling, serum DHEAS concentrations in SGA children were higher (median 2-fold increase; range 1.1-7; p = 0.002) than in AGA siblings. Conversely, in the 3 pairs with still discordant weight (>2 SD difference), serum DHEAS levels in SGA children were lower than in AGA children.

Table 1 Clinical data and serum concentrations of DHEAS in SGA and AGA siblings with similar weight at the time of sampling (upper part) and with still discordant weight (lower part)
Full size table
Figure 1
figure 1

Serum DHEAS concentrations in AGA and SGA siblings of twin (•), triplet (▴), or quadruplet (▪) pregnancies. Left panel: current difference in weight SDS between siblings ≤ 1 SD; right panel: current difference in weight SD score between siblings >2 SD. DHEAS levels of siblings are interlinked.

Full size image

DISCUSSION

In 10 pairs of children, born as discordant siblings but matched for current weight, serum DHEAS was consistently higher in the SGA child, compared with the AGA sibling. In three SGA children who were still discordant in weight, serum DHEAS was lower than in their AGA sibling.

During late childhood and adolescence, serum DHEAS is positively related to age and current weight(4, 5, 8). The study design and the adjustment for current weight unmasked the association between size at birth and serum DHEAS in childhood: prenatal growth restriction appears to result in a more pronounced adrenarche. It will be of interest to differentiate whether the relative increase of serum DHEAS results from an acceleration or from an amplification of adrenarche, as the former may be a temporary phenomenon limited to childhood and adolescence, whereas the latter may have lifelong consequences(13).

In siblings with still differing weight, the modulation of adrenarche by prenatal growth (negative relationship) appears to be overcome by the influence of current weight on circulating DHEAS (positive relationship).

The pathophysiologic mechanism underlying our findings is at present unclear. Both the fetal adrenal and the zona reticularis of the adult adrenal cortex are characterized by a low level of expression of 3β-hydroxysteroid dehydrogenase(14). It is therefore possible that the function of this enzyme plays a crucial role in the association between adrenarche and fetal growth.

Children born SGA present a particular pattern of growth and bone maturation. A subgroup of them is known to have an acceleration of bone maturation in late childhood, independently of pubertal development(15). The data reported here suggest that exaggerated adrenarche may participate in the pathogenesis of this phenomenon.

In conclusion, the presented findings, which account for both prenatal and postnatal weight gain, unmask a link between adrenarche and fetal growth. This relationship further supports the concept of early endocrine“programming” and extends this principle to adrenarche.