| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Hospital for Children and Adolescents (E.K., T.H., P.H., S.A.) and Department of Obstetrics and Gynecology (S.A.), Helsinki University Central Hospital, 00029 HUS Helsinki, Finland; and National Public Health Institute (E.K.), 00300 Helsinki, Finland
Address all correspondence and requests for reprints to: Eero Kajantie, M.D., National Public Health Institute, Mannerheimintie 166, 00300 Helsinki, Finland. E-mail: eero.kajantie{at}helsinki.fi.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We measured plasma adiponectin concentrations by ELISA in cord vein of 197 infants. Of them, 122 were born preterm (gestational age, 22–32 wk), and 75 at term (49 from a healthy and 26 from a diabetic pregnancy, with similar findings, and thus all data from term infants pooled).
Mean adiponectin concentrations increased from less than 1 µg/ml at 24 wk gestation to approximately 20 µg/ml at term. One week increase in gestational age corresponded in preterm infants to 43% increase (95% confidence interval 34–53%; P < 0.0001) in adiponectin and term infants to 21% increase (12–31%; P < 0.0001). In preterm infants, one unit increase in birth weight SD score corresponded to 42% increase (22–66%; P = 0.0001) in adiponectin, and females had 57% higher adiponectin concentrations (0–146%; P = 0.05) than males. These differences were not seen in term infants. Adiponectin levels were lower in preterm infants with recent (<12 h) exposure to maternal betamethasone but were unrelated to mode of delivery, preeclampsia, or impaired umbilical artery flow.
In conclusion, adiponectin concentrations in fetal circulation show a 20-fold rise between 24 wk gestation and term and, in preterm infants are associated with birth weight SD score, sex, and glucocorticoid exposure. Adiponectin may play an important role in regulating fetal growth and explaining its links to the metabolic syndrome and its consequences during adult life.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Adiponectin is present in abundant concentrations in cord blood of term infants (5, 6), but little is known about its role earlier during the fetal period. It has, however, several features that make its role during the fetal period an intriguing object to study. A body of evidence from experimental and epidemiological studies have shown that the metabolic syndrome and its components are associated with small size at birth, a proxy of intrauterine conditions that result in slow fetal growth (7, 8, 9, 10). Fetal growth is to a great extent controlled by actions of insulin (11). Adiponectin as a key regulator of insulin sensitivity could thus be expected to have significant effects on fetal growth and development. If so, it would constitute an important candidate to explain the link between small size at birth and adult metabolic syndrome. A further intriguing feature of adiponectin is that, although it is the product of the most abundant gene transcript in adipocytes (12), in children (13) and adults (14, 15, 16) its concentrations show paradoxically an inverse correlation with body fat percentage, suggesting that the amount of fat may exert negative feedback on adiponectin synthesis (3). Because the proportion of body fat rises throughout mid- to late gestation, preterm infants across a wide range of gestational ages constitute a useful model to test this hypothesis. With this background, we set out to study how gestational age and clinical conditions in preterm and term pregnancies affect plasma adiponectin concentrations in a cohort of infants with a wide range of gestational ages and prematurity-associated morbidity.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study population comprised 197 newborn infants born at the Department of Obstetrics and Gynecology, Helsinki University Central Hospital, Helsinki, Finland. To elucidate the role of circulating adiponectin during normal and pathologic development of the mid- and late pregnancy fetus, we chose a study population consisting of three separate groups. The preterm group encompassed 122 infants born between 22 and 32 wk gestation. The 49 healthy term infants born after 36 wk gestation from an uncomplicated pregnancy served as a reference group. The possible effects of maternal diabetes were studied in 26 infants from a full-term diabetic pregnancy. Table 1
shows the clinical data.
|
Betamethasone (12 mg im twice at 24-h intervals) served as an antenatal glucocorticoid treatment when a preterm delivery was imminent. The treatment was repeated in 7–10 d, if necessary. Of the 122 preterm infants, nine received no betamethasone, 89 infants one course, 21 infants two courses, and three infants three courses. The time between the last betamethasone dose and birth and the number of betamethasone treatments were both considered variables in the data analysis.
Adiponectin measurements
Immediately after birth, umbilical vein samples were drawn into EDTA-containing tubes with plasma frozen without delay and stored at –20 C until analysis. Adiponectin concentrations were measured by ELISA (R&D Systems Inc., Minneapolis, MN).
The study protocol was approved by the Ethics Committee of the Department of Obstetrics and Gynecology, Helsinki University Central Hospital.
Data analysis
Adiponectin concentrations were right skewed and thus log transformed into normality. Simple and multiple linear regressions were used to assess correlation between variables. Unless otherwise stated, regression equations are adjusted for gestational age, relative birth weight, and infant sex. Because of the logarithmic transformation of the independent variables, each regression coefficient indicates percent change in plasma adiponectin concentration associated with one unit change in the dependent variable. To allow for possible nonlinear effects, a reference cell dummy coding was employed to account for the time between the last betamethasone dose and birth. Subjects were divided into four groups: 1) less than 12 h (n = 26); 2) 12–72 h (n = 33); 3) 72 h to 7 d (n = 28); and 4) the reference group with more than 7 d or no betamethasone (n = 35), who are no longer expected to show any effect of administered betamethasone (20). As a consequence of the reference cell dummy coding, the regression coefficient of each dummy variable in groups 1, 2, and 3 denotes difference to the reference group (21).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. Term infants from pregnancies with maternal diabetes had similar adiponectin concentrations as healthy term infants (18.7 vs. 23.7 µg/ml, P = 0.2), with no difference between infants of mothers with gestational and type 1 diabetes (P = 0.8). Therefore, we present a pooled analysis for infants from healthy and diabetic term pregnancies. Gestational age
Plasma adiponectin concentrations were closely correlated with gestational age (Fig. 1 and Table 2). In preterm infants, 1 wk of increase in gestational age was associated with 43% [95% confidence interval (CI) 34–53%; P < 0.0001] increase in plasma adiponectin, whereas the relationship was less steep in term infants, 21% (12–31%; P < 0.0001; P for interaction preterm vs. term infants = 0.03) (Table 2). With preterm and term infants combined, the relationship between gestational age and plasma adiponectin gave a good fit to a quadratic model [log (adiponectin) = –19.56 + 1.14 * gestational age – 0.0143 * gestational age2; adjusted r2 = 0.68; P < 0.0001] (Fig. 1).
|
|
In an unadjusted regression analysis, birth weight was associated with cord plasma adiponectin in preterm infants, in whom a 100-g increase in birth weight corresponded to 26% increase (95% CI 22–31%, P < 0.0001) in adiponectin and also in term infants, a corresponding increase being 3.5% (95% CI 1.1–5.9%, P = 0.005). Table 2 shows that the relationship between size at birth and plasma adiponectin was independent of gestational age in preterm but not term infants. In preterms, one SD score (SDS) increase in birth weight SDS was associated with 42% increase (95% CI 22–66%; P = 0.0001) in adiponectin (Fig. 2). This relationship remained similar when adjusted for gestational age and sex. No relationship with birth weight SDS was, however, seen in term infants (P = 0.2; P for interaction preterm vs. term = 0.05) (Table 2). Moreover, in both preterm and term infants, no association was seen between ponderal index and plasma adiponectin, whether adjusted for gestational age, birth weight SDS, and sex. Adiponectin concentrations were as well unaffected by maternal height (P = 0.4 for preterm and 0.2 for term infants) or prepregnancy body mass index (P = 0.2 for preterm and 0.3 for term infants).
|
Preterm girls had 57% higher plasma adiponectin concentrations (95% CI 0–146%; P = 0.05) than preterm boys. Adjusted for gestational age and relative birth weight, the difference was 47% (11–95%; P = 0.008). There was no sex difference in all term infants combined (P = 0.4; P for interaction preterm vs. term = 0.09) (Table 2) or in infants from healthy or diabetic pregnancies.
Mode of delivery
Plasma adiponectin concentrations were similar in infants born with cesarean section, compared with vaginal delivery (P = 0.7 for preterm and 0.6 for term infants).
Preeclampsia and fetal distress
We assessed whether fetal distress per se is associated with plasma adiponectin concentrations in preterm infants of mothers with associated conditions. Compared with other preterm infants, adjusted for gestational age and sex, infants of mothers with maternal hypertension had 36% (95% CI 10–54%, P = 0.01), with preeclampsia 28% (–10 to 54%, P = 0.1) and impaired umbilical artery flow 50% lower (25–67%, P = 0.001) adiponectin concentration. However, this association was entirely attributable to the growth restriction associated with these conditions: further adjustment with birth weight SDS resulted in no association with maternal hypertension (P = 0.7), preeclampsia (P = 0.8), or increased umbilical artery resistance (P = 0.4). Moreover, in both preterm (P = 0.6) and term (P = 0.8) infants, cord vein adiponectin was unrelated to cord artery pH.
Maternal betamethasone treatment
Most preterm infants had been exposed to maternal betamethasone treatment, administered to reduce neonatal morbidity and mortality. We assessed its effects by comparing plasma adiponectin concentrations with those of preterm infants unexposed to betamethasone with at least 7 d after last betamethasone exposure, who are expected to have no residual betamethasone activity (20). Compared with this group, as shown in Fig. 3, adiponectin concentrations were reduced in infants with less than 12 h after last betamethasone dose, similar in those with 12–72 h after last betamethasone and increased in those with 72 h to 7 d after last betamethasone. Plasma adiponectin concentrations were unrelated to the number of betamethasone treatments given.
|
To demonstrate the combined effects of clinical variables on cord plasma adiponectin concentrations, we calculated multiple regression models shown in Table 2. In both preterm and term infants, adjustment of the effects of gestational age, birth weight SDS, and infant sex on plasma adiponectin to each other produced only a moderate change in the associations. We further checked whether any two- or three-way interactions are present between the effects of these variables on plasma adiponectin. No interaction was seen in preterm infants, but in term infants the interaction between the gestational age and birth weight SDS was statistically significant (P = 0.007). To examine this interaction, we divided the term group into two groups according to gestational age. In term infants born before 40 wk gestation, one unit in birth weight SDS corresponded to 14% increase (95% CI –5 to 36%, P = 0.1) in adiponectin concentration, whereas in infants born after 40 wk gestation, the corresponding change was –7% (–17 to 4%, P = 0.2).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In a term newborn, adipose tissue is a crucial element for survival as an energy reserve, although little has so far been known about its role as an endocrine organ. Adipose tissue is present at least from 14 wk gestation, and its morphologic differentiation mostly takes place during the second trimester, by the end of which adipose tissue is present at principal body fat deposit areas (23, 24). Most of the increase in the amount of adipose tissue occurs during the last trimester, paralleling closely the gestational age increase we found in adiponectin concentrations. This is in marked contrast with adults (14, 15, 16) and children (13), in whom adiponectin concentrations paradoxically show an inverse correlation with body fat percentage. Together with the correlation with birth weight SDS we observed in preterm infants, this finding implies that the proposed inhibition of adiponectin synthesis by the amount of fat mass, possibly through the production of other adipocytokines (3), is not yet operative in the fetus before 32 wk gestation.
At term, however, we found adiponectin concentrations to be unrelated to relative birth weight, although there was a significant interaction between the effects of gestational age and birth weight SDS, suggesting a weak positive relationship that fades away with increasing gestational age. A modest positive relationship between birth weight and cord vein adiponectin was also shown in a recent study of healthy term infants, although it remains uncertain how much of this relationship was attributable to increasing gestational age, which was not accounted for in that study (6). Moreover, we found no relationship between adiponectin and maternal diabetes (gestational or type 1), and another study has shown only a minor decrease in adiponectin in term infants of mothers with type 1 diabetes (5). The faintness of these relationships may be explained by the fact that none of these studies included a sufficient number of term infants born small for gestational age. Alternatively, they may suggest that the phenomenon of reduced adiponectin concentration per amount of adipose tissue is becoming apparent already at term.
Apart from the energy-storing white adipose tissue, newborn infants possess an amount of brown adipose tissue whose main function, nonshivering thermogenesis, makes it crucial for body temperature homeostasis after birth. The amount of brown adipose tissue increases significantly during late gestation, but during childhood this tissue tends to become atrophied (25). Brown adipocytes exhibit abundant expression of the adiponectin gene (26). Interestingly, its regulation appears to differ from that in white adipocytes. For example, the effect of insulin on adiponectin synthesis is inhibitory in white (27) and stimulatory in brown adipocytes (26). It is thus possible that these differences in regulation of adiponectin synthesis may in part explain why the negative association between corpulence and adiponectin concentration in children (13) and adults (14, 15, 16) is not seen in newborns.
An obvious reservation to these conclusions is that we do not have a specific measure of the amount of adipose tissue. However, particularly in ventilated small preterm infants, specific measurements such as dual-energy x-ray absorptiometry are highly impractical, and there is no consensus on how anthropometric variables such as skinfolds should be interpreted. Yet in preterm and term neonates, body weight has been shown to explain, together with infant sex, 85% of variation in fat mass and 72% of variation in percent body fat as assessed by dual-energy x-ray absorptiometry (28), and thus it constitutes a satisfactory estimation of body fat content.
The reduced adiponectin concentrations we found in growth-retarded preterm infants are intriguing in light of the enhancing effect of adiponectin on insulin sensitivity. The regulation of fetal growth differs considerably from that during the postnatal period. The IGF system is mostly controlled by insulin, which allows rapid responses to nutritional fluctuations, in contrast to the more stable actions of the GH system, which predominates during the postnatal period (11). The sensitivity of fetal tissues to insulin action has thus potentially profound effects on fetal growth. Rare inherited disorders with impaired insulin sensitivity, such as leprechaunism caused by homozygous mutations in the insulin receptor gene, are associated with marked intrauterine growth retardation (29). A significant number of small preterm infants requires postnatal insulin infusions to maintain glucose homeostasis (30). However, the role and mechanisms of the regulation of insulin sensitivity during the fetal period have been poorly understood. The present study suggests that adiponectin is a key candidate for further studies on the regulation of insulin sensitivity in the growing fetus.
In adults, adiponectin concentrations are consistently lower in males than females. A similar difference has been observed in a group of infants of mothers with type 1 diabetes, with mean gestational age of 37 wk 6 d (5) but not in term infants from healthy pregnancies (5, 6). These findings, together with our result of a comparable difference in preterm but not in term infants, suggest that the mechanism(s) responsible for the sex difference in the fetus could be associated with gestational age. In adults, the suppression of adiponectin synthesis by androgens has been suggested to account for the sex difference (22). During midpregnancy male fetuses exhibit 5-fold higher mean testosterone concentrations than females (31), a difference that is reduced to less than 1.5-fold by term (32). It is therefore possible that a similar regulation is operative in the growing fetus, but to confirm that requires further study.
We found no association between adiponectin concentrations and mode of delivery. This suggests that fetal stress associated with vaginal delivery does not have a significant effect on adiponectin concentrations, implying that concentrations measured from cord vein samples obtained after birth are a reliable indicator of adiponectin concentrations during similar conditions in utero. This relative stability is consistent with findings of no changes in adiponectin concentrations between birth and 4 d of age in term infants (6) or in adults after physical exercise (33) or a high-fat meal (34) and only modest changes during a diurnal cycle (35).
Low adiponectin concentration was associated with preeclampsia, maternal hypertension, or impaired umbilical artery flow, which is an end-stage feature of these and other conditions characterized by severe placental dysfunction. However, all these associations became nonsignificant when adjusted for birth weight SDS. Fetal growth impairment is a key feature of these conditions, which in turn are most important causes of intrauterine growth restriction, particularly in preterm infants. This makes it difficult to distinguish whether the low adiponectin concentration is a consequence of fetal growth restriction in general or a specific characteristic of these disorders. A specific mechanism is argued for by the association of preeclampsia with impaired placental 11ß-hydroxysteroid dehydrogenase 2 function (36, 37). This results in increased fetal exposure to maternal cortisol, which would be expected to reduce adiponectin concentrations. However, the role of adiponectin in the pathophysiology of preeclampsia may be more complex. The reduction in 11ß-hydroxysteroid dehydrogenase 2 function is less pronounced before 32 wk gestation (38), and despite the well-known association of preeclampsia and maternal insulin resistance (39), there is at least one report (40) showing increased adiponectin concentrations in maternal circulation during preeclampsia, compared with normal pregnancies.
Glucocorticoids are known inhibitors of adiponectin synthesis (41). Consistent with that, we found reduced adiponectin concentrations in preterm infants exposed to betamethasone less than 12 h before birth. It is thus possible that the adverse metabolic effects of perinatal glucocorticoid administration, such as decrease in insulin sensitivity and increase in blood pressure, could in part be mediated through decreased adiponectin concentrations. The increased adiponectin concentrations we found in infants born between 3 and 7 d after last betamethasone are more difficult to explain. Whether they are related to the overall favorable effects of antenatal glucocorticoids in reducing neonatal morbidity remains to be studied.
In conclusion, adiponectin concentrations in fetal circulation show a 20-fold rise between 24 wk gestation and term. They are unrelated to the mode of delivery implying that cord plasma concentrations at birth represent circulating concentrations in the growing fetus. In preterm infants, low adiponectin concentrations are also associated with low-birth-weight SDS, male sex, and recent exposure to maternal betamethasone. These findings suggest that the proposed inhibition of adiponectin by the amount of body fat is not yet operative before 32 wk gestation. Adiponectin may play a significant role in regulating fetal growth and explaining its links to the metabolic syndrome and its consequences during adult life.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
E.K. and T.H. contributed equally to this work.
Abbreviations: CI, Confidence interval; SDS, SD score.
Received January 6, 2004.
Accepted April 27, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. Troisi, N. Potischman, and R. N. Hoover Exploring the Underlying Hormonal Mechanisms of Prenatal Risk Factors for Breast Cancer: A Review and Commentary Cancer Epidemiol. Biomarkers Prev., September 1, 2007; 16(9): 1700 - 1712. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Baviera, F. Corrado, C. Dugo, M. L. Cannata, S. Russo, and D. Rosario Midtrimester Amniotic Fluid Adiponectin in Normal Pregnancy Clin. Chem., September 1, 2007; 53(9): 1723 - 1724. [Full Text] [PDF]
|
|
|
|
 |
|
 A Klamer, K Skogstrand, D M Hougaard, B Norgaard-Petersen, A Juul, and G Greisen Adiponectin levels measured in dried blood spot samples from neonates born small and appropriate for gestational age Eur. J. Endocrinol., August 1, 2007; 157(2): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Siahanidou, H. Mandyla, G.-P. Papassotiriou, I. Papassotiriou, and G. Chrousos Circulating levels of adiponectin in preterm infants Arch. Dis. Child. Fetal Neonatal Ed., July 1, 2007; 92(4): F286 - F290. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. Willemsen, M. van Dijk, Y. B. de Rijke, A. W. van Toorenenbergen, P. G. Mulder, and A. C. Hokken-Koelega Effect of Growth Hormone Therapy on Serum Adiponectin and Resistin Levels in Short, Small-for-Gestational-Age Children and Associations with Cardiovascular Risk Parameters J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 117 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kadowaki, M. Waguri, I. Nakanishi, Y. Miyashita, M. Nakayama, N. Suehara, T. Funahashi, I. Shimomura, and T. Fujita Adiponectin Concentration in Umbilical Cord Serum Is Positively Associated with the Weight Ratio of Fetus to Placenta J. Clin. Endocrinol. Metab., December 1, 2006; 91(12): 5090 - 5094. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Bansal, V. Charlton-Menys, P. Pemberton, P. McElduff, J. Oldroyd, A. Vyas, A. Koudsi, P. E. Clayton, J. K. Cruickshank, and P. N. Durrington Adiponectin in Umbilical Cord Blood Is Inversely Related to Low-Density Lipoprotein Cholesterol But Not Ethnicity J. Clin. Endocrinol. Metab., June 1, 2006; 91(6): 2244 - 2249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. K. Ong, J. Frystyk, A. Flyvbjerg, C. J. Petry, the Avon Longitudinal Study of Parents and Childre, A. Ness, and D. B. Dunger Sex-Discordant Associations With Adiponectin Levels and Lipid Profiles in Children. Diabetes, May 1, 2006; 55(5): 1337 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kajantie, R. Kaaja, O. Ylikorkala, S. Andersson, and H. Laivouri Adiponectin Concentrations in Maternal Serum: Elevated in Preeclampsis But Unrelated to Insulin Sensitivity Reproductive Sciences, September 1, 2005; 12(6): 433 - 439. [Abstract] [PDF]
|
|
|
|
|
|
J. Verhaeghe, R. van Bree, E. van Herck, and W. Coopmans Exogenous Corticosteroids and in Utero Oxygenation Modulate Indices of Fetal Insulin Secretion J. Clin. Endocrinol. Metab., June 1, 2005; 90(6): 3449 - 3453. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Corbetta, G. Bulfamante, D. Cortelazzi, V. Barresi, I. Cetin, G. Mantovani, S. Bondioni, P. Beck-Peccoz, and A. Spada Adiponectin Expression in Human Fetal Tissues during Mid- and Late Gestation J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2397 - 2402. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
