| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Pediatrics (B.C.G., C.L.R.), Statistics (R.F.), and Neonatology (P.B.), University of Bonn, 53113 Bonn, Germany; and Department of Obstetrics and Fetal Medicine (A.H., K.H.), University Clinic Hamburg-Eppendorf, 20246 Hamburg, Germany
Address all correspondence and requests for reprints to: Dr. B. Gohlke, Zentrum für Kinderheilkunde, der Universität Bonn, Adenauerallee 119, 53113 Bonn, Germany. E-mail: gohlke-bonn{at}t-online.de.
| Abstract |
|---|
|
|
|---|
Patients and Methods: Twenty-seven twins (15 with discordant growth) who have been treated for severe twin-to-twin transfusion syndrome by laser coagulation were studied. Cord blood samples were analyzed for IGF-I, IGF-II, IGF-binding protein-2, and ghrelin. Intertwin difference (
) of birth weight was correlated to
of the parameters analyzed. The
weight after 1 yr was correlated with
birth weight and all hormones.
Results: The
birth weight was positively correlated with
IGF-I (r = 0.66; P < 0.0002) and negatively correlated with
IGF-binding protein-2 levels (r = 0.68; P < 0.001) but with neither
IGF-II nor
ghrelin. There was a strong intertwin correlation for all hormones. By comparing the growth in the first year, we found an overall reduction of the relative weight difference between the twins of 57%. ANOVA was used to calculate factors for prediction of postnatal catch-up growth. Besides the birth weight difference (R2 = 0.84; P < 0.0001), only ghrelin was of prognostic value for postnatal catch-up growth (R2 = 0.94; P = 0.0035).
Conclusion: These data confirm the importance of IGF-I in contrast to IGF-II for fetal weight. Additionally, ghrelin seems to be involved in fetal and probably postnatal growth.
| Introduction |
|---|
|
|
|---|
Intrauterine growth is regulated by fetal as well as maternal and environmental factors. Little is known of their relative contribution and importance because singleton studies are limited in this regard. Several studies of singleton pregnancies have shown that various hormones are associated with birth weight. However, singleton studies fail to address the question of whether alterations in fetal nutrition leading to impaired birth weight are due to disturbance in placental transport or in fetal hormone function. Our study population recruited only monozygotic (MZ) but hemodynamically dichorionic twins with discordant birth weight. MZ twinning is a powerful clinical model because of the identical genetical background, the same maternal nutrition, and only the individual differences for the fetal environment. Therefore, they are an excellent model to study the interaction between genetic and environmental factors and their effects on fetal growth (5, 6).
The IGFs are involved in the regulation of growth during pregnancy as well as in early embryonic and fetal development. In most studies, cord blood IGF-I but not IGF-II concentrations correlate with birth weight (7, 8, 9, 10, 11). In contrast, others have shown normal IGF-I (12, 13) and reduced IGF-II levels in singleton intrauterine growth retardation fetuses (14, 15). There are only a few studies in twins with interpair variation in birth weight. Studies in concordant twins suggest that fetal circulating IGF-I levels may be genetically determined (16).
Ghrelin is a peptide predominantly produced by the stomach, hypothalamus, and adipose tissue (17, 18). It displays strong GH-releasing action (19). In addition, ghrelin is also actively synthesized by the placenta (20) and can be detected in cord blood at 30 wk gestational age, which indicates that it may play a role in fetal development (21, 22, 23). Its various endocrine and nonendocrine actions have been subjects of recent scientific work (24). Previous studies indicate that ghrelin levels are influenced by acute and chronic feeding state, with increase by fasting and energy restriction (25, 26) but decrease by food intake and glucose (27, 28, 29).
In the present study, we measured IGF-I, IGF-II, IGF binding protein (IGFBP)-2, and ghrelin concentrations in cord blood in MZ monochorionic twins and examined its relationship to birth weight and growth after 1 yr of life.
| Patients and Methods |
|---|
|
|
|---|
Twenty-seven women with monochorionic MZ twin pregnancies and TTTS were studied. The diagnosis of TTTS was made by the combination of single monochorionic placenta, polyhydramnios and oligohydramnios, stuck-twin, and an initial diagnosis before 25 wk gestation. These fetuses were treated by means of laser coagulation between 17 and 25 wk gestation. All the fetuses were delivered in the local referring hospital.
Fifteen twin pairs presented with discordant growth [birth weight difference > 15%, one twin being small for gestational age (SGA, birth weight < 2 SD for gestational age), the other appropriate birth weight (AGA)], and 12 twin pairs were concordant concerning their growth (AGA-AGA). All twins were measured regularly at 2, 4, 6, and 12 months of age. Nineteen twin pairs who were older than 1 yr were analyzed concerning their postnatal growth.
Collection of samples
Fetal cord blood was obtained from each twin from the umbilical venous blood from clamped segment of cord at the birth. The samples were centrifuged, and serum was stored at 70 C until a batch assay was performed. Informed consent for collection of cord blood samples was obtained from all parents. The study protocol was approved by the local ethics committee.
Immunoassays
All hormone concentrations were measured by radioimmunometric assays using commercially available kits (Mediagnost, Reutlingen, Germany). IGF-I (nanograms per milliliter) was measured by RIA after employing excess IGF-II to saturate IGFBPs. The detection limit was 0.02 ng/ml. IGF-II (nanograms per milliliter) was measured after separation from IGFBP by acid chromatography according to Blum and Breier (30). The detection limit was 0.1 ng/ml. The sensitivity of the IGFBP-2 assay was 0.2 ng/ml. Interassay variances were 7.4 (IGF-I), 7.9 (IGF-II), and 9.6% (IGFBP-2), respectively. Intraassay variances were 5.6 (IGF-I), 5.4 (IGF-II), and 8.5% (IGFBP-2), respectively. Immunoreactive ghrelin concentrations were measured in duplicate using a commercial RIA (Linco Research, Inc., St. Charles, MO). The antibody used in the assay is a rabbit polyclonal antibody against full-length octanoylated human ghrelin. Intra- and interassay coefficients of variation were 3.3 and 17.8%, respectively.
Statistical analysis
Clinical data and hormone concentrations are expressed as medians and ranges. Delta values (
) indicate differences between the twins. For parametric data, the paired t test was used to compare values within twin pairs and Students t test between groups. Spearman correlation was calculated between relative intertwin birth weight difference (percent) and
of the hormones. The percent growth discordance was defined as the difference in birth weight expressed as a proportion of the birth weight of the larger twin. Intertwin difference of IGF-I, IGF-II, IGFBP-2, and ghrelin was expressed in
. Statistical analysis was performed by the SAS system (SAS Institute, Cary, NC). Multiple linear regression analysis was used to calculate the predictors for weight difference at birth and after 1 yr.
| Results |
|---|
|
|
|---|
|
In the discordant group, fetal IGF-I concentrations in SGA twins were significantly lower than those in the AGA cotwin (P > 0.01) (Table 1
). No difference was observed in the concordant group (P = 0.76). Spearman correlation coefficient (r) between
birth weight and
IGF-I was 0.66 (P < 0.001) (Fig. 1A
).
|
Fetal IGF-II levels in recipient and donor twins in both groups were comparable (P = 0.79 for discordant twins, P = 0.4 for concordant twins; Table 1
). Spearman correlation coefficient (r)
birth weight to
IGF-II difference was 0.08 (P = 0.71; Fig. 1B
).
IGFBP-2
In the discordant group, fetal IGFBP-2 concentrations in SGA twins were significantly higher than those in the AGA cotwin (P < 0.01; Table 1
). Spearman correlation coefficient (r) of
birth weight to
IGFBP-2 was 0.68 (P < 0.001). The
IGFBP-2 levels were significantly correlated to
IGF-I (r = 0.5; P = 0.01) but not to
IGF-II.
Ghrelin
Spearman correlation coefficient showed a negative but not significant correlation between
birth weight and
ghrelin levels (r = 0.39, P = 0.10).
Correlations among study hormones and between twin pairs
We examined univariate correlations among study hormones and found IGF-I to be negatively correlated to IGFBP-2 (r = 0.53; P < 0.01). There was no correlation among the other hormones at birth. Table 2
shows Spearman correlation coefficients among all measured hormones.
|
|
Nineteen twin pairs were analyzed concerning their postnatal weight and length development, and an overall reduction of 57.1% of relative weight difference was observed. Single data are shown in Fig. 3
. Three of the nine pairs of twins with discordant weight at birth did not show a significant reduction of relative weight difference.
|
1 yr) and relative birth weight difference (R2 = 0.84; P < 0.0001) (Fig. 4
ghrelin level (R2 = 0.94; P < 0.0035).
|
| Discussion |
|---|
|
|
|---|
In accordance with most published data (7, 8, 9, 10, 11) and in contrast to Bajoria et al. (5), we found a strong correlation of IGF-I levels and birth weight. Among the group of discordant twins, the normally grown twin, in all but one case, had significantly higher cord serum IGF-I levels than the growth-restricted cotwin. There was no significant intertwin difference in the cord blood IGF-I levels in the concordant twin pairs. These data demonstrate that IGF-I is important in the regulation of fetal growth, and its mechanism appears to be, at least in part, through an endocrine action. There is also evidence that the weight difference between twins suffering from chronic TTTS cannot be explained only by the transfer of blood and nutrients along the placental vascular anastomoses. Although all our patients had a laser coagulation of the connecting vessels, more than half of the twins presented with discordant birth weight.
In addition, we found a strong intertwin correlation for IGF-I for all twin pairs, confirming the underlying genetic influence on the IGF-I level (31).
In IGF-II a strong intertwin correlation was found, but it was not related to fetal weight or length in our study population. IGF-II may play a role as a fetal somatomedin during earlier periods of human intrauterine life.
IGFBP-2 is one of the major IGFBPs in the fetus (32). It is known to correlate negatively to birth weight, indicating an antagonistic regulatory mechanism for the effect of IGF. Alternatively, IGFBP-2 levels might be up-regulated in the presence of low IGF-I levels. In disease states it has been shown that if IGF-I levels were suppressed, IGFBP-2 levels were elevated (33).
Umbilical cord blood concentrations of ghrelin were inversely related to birth weight. This is in accordance with very recent data demonstrating that cord ghrelin concentrations were higher in SGA neonates, compared with AGA/large-for-gestational-age neonates (22, 23, 34). A similar inverse relationship is found in patients with anorexia nervosa having higher and obese patients having lower ghrelin levels than controls (25, 26, 29). We cannot add information about change of ghrelin concentrations throughout gestation due to our small patient numbers.
The source(s) of circulating ghrelin in fetuses remains unclear. It could originate from the placenta (20), the stomach (17), or other fetal organ tissues (35) that are known to synthesize ghrelin during early fetal life. Our study population recruited only MZ but hemodynamically dichorionic twins after laser coagulation. We suggest that the SGA fetus itself had developed a compensatory mechanism to the negative energy balance, and the increased ghrelin level would be an indication of an adaptive response.
The most important predictor for catch-up growth between the discordant twin pairs was the relative weight difference at birth, a greater difference at birth being associated with greater weight difference after 1 yr. Although IGF-I was positively associated with birth weight, it was not of additional prognostic value for the weight development during the first year of life. Further information was received only by including intertwin difference of ghrelin. SGA twins with a high ghrelin concentration had a better chance for catching up of weight after 1 yr. We suggest that higher ghrelin concentrations remained high throughout the first year of life, leading to an increased appetite and different satiety regulation in those patients followed by rapid postnatal catch-up growth. This is in accordance with recent findings of Iniguez et al. (36). They observed a significantly smaller glucose-induced drop in ghrelin concentrations in 1-yr-old infants born SGA who had experienced catch-up growth, compared with those who had not, and proposed that higher postprandial ghrelin concentrations may have resulted in greater weight gain early in life.
In conclusion, our data confirm the importance of IGF-I vs. IGF-II for fetal weight development, although IGF-I levels were significantly higher in the AGA twin, compared with the SGA cotwin, whereas these levels were similar in the AGA twin pairs. Furthermore, this demonstrates the strong contribution of the uterine environment for IGF-I. However, a significant intertwin correlation was found for both growth factors, showing also the genetic influence on IGF-I and -II.
Furthermore, our data show that ghrelin is negatively correlated to birth weight difference with higher levels in the lighter twin. Birth weight difference was the strongest predictor for weight difference after 1 yr, but prediction of catch-up growth was dependent on ghrelin levels with high cord blood ghrelin associated with improved catch-up growth than lower levels. These data confirm the importance of both genetic and intrauterine factors in early human development.
| Acknowledgments |
|---|
| Footnotes |
|---|
Abbreviations:
, Differences; AGA, appropriate birth weight; IGFBP, IGF binding protein; MZ, monozygotic; SGA, small for gestational age; TTTS, twin-twin-transfusion syndrome.
Received June 22, 2004.
Accepted January 21, 2005.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
O. M. Faye-Petersen and T. M. Crombleholme Twin-to-Twin Transfusion Syndrome: Part 1. Types and Pathogenesis NeoReviews, September 1, 2008; 9(9): e370 - e379. [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 |