| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Paediatrics (R.H.W., S.W.K.d.K., D.C.M.v.d.K., A.C.S.H.-K.), Division of Endocrinology, Erasmus MC Sophia, 3015 GJ Rotterdam, The Netherlands; and Dutch Growth Foundation (D.C.M.v.d.K., A.C.S.H.-K.), 3001 KB Rotterdam, The Netherlands
Address all correspondence and requests for reprints to: Ruben Willemsen, M.D., Erasmus MC Sophia, Room SB-2603, Dr. Molewaterplein 60, 3015 GJ Rotterdam, The Netherlands. E-mail: r.h.willemsen{at}erasmusmc.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: Our objective was to investigate whether prematurity has an independent influence on several cardiovascular risk factors within a population of short SGA children.
Design: A cross-sectional observational study was performed.
Patients: A total of 479 short SGA children (mean age 6.8 yr), divided into preterm (<36 wk) and term (
36 wk) children, was included in the study.
Outcome Measure: Insulin sensitivity, β-cell function, body composition, and lipid levels were studied in subgroups, and blood pressure (BP), anthropometry at birth and during childhood in the total group.
Results: Preterm SGA children were significantly lighter and shorter at birth after correction for gestational age than term SGA children (P < 0.001) but had a comparable head circumference. In preterm SGA children, we found a significantly higher systolic (P = 0.003) and diastolic BP SD score (P = 0.026), lower percent body fat SD score (P = 0.011), and higher insulin secretion (P = 0.033) and disposition index (P = 0.021), independently of the degree of SGA. Insulin sensitivity, serum lipid levels, muscle mass, and body fat distribution were comparable for preterm and term SGA children.
Conclusions: Within a population of short SGA children, preterm birth has divergent effects on several cardiovascular risk factors. Whereas preterm SGA children had a higher systolic and diastolic BP, they also had a lower percent body fat and a higher insulin secretion and disposition index than term SGA children.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We and others have previously shown that short small-for-gestational-age (SGA) children have a lower insulin sensitivity (Si) (4, 5) and a higher systolic blood pressure (BP) (4, 6). Recent reports indicate that prematurity itself also might have adverse consequences with regard to cardiovascular risk factors. Compared with controls born at term, premature born young adults had higher BP and fasting glucose levels, regardless of being SGA (7). In a cohort of young adult Swedish men, prematurity was recognized as a risk factor for high BP, independent of birth weight SD score (SDS) (8).
In contrast, others reported that among children born preterm, only those who were born SGA had an increased systolic BP and pulse wave velocity, which is a measure for arterial stiffness (9). Singhal et al. (10) found comparable levels of fasting glucose, insulin, and lipids between preterm SGA and AGA adolescents.
Most aforementioned studies investigated the influence of SGA within a population of preterm born or "low birth weight" subjects. However, because only a minority of all preterm children is born SGA (
2–3%), these studies may not have been suitable to investigate whether a combination of SGA and preterm birth is worse with regard to cardiovascular risk parameters than being born SGA at term.
The aim of our study was to investigate whether prematurity has an independent influence on Si, β-cell function, and other cardiovascular risk factors within a large population of short SGA children. We hypothesized that being born both preterm and SGA is associated with a worse cardiovascular risk profile than being born SGA at term. To test our hypothesis, we compared preterm short SGA children with term short SGA children.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study group comprised 479 prepubertal short children born SGA. All children fulfilled the same inclusion criteria: 1) birth length and/or birth weight SDS below –2 for gestational age (11); 2) height SDS below –2 according to Dutch standards (12); 3) height velocity SDS below zero to exclude children with spontaneous catch-up growth (12); 4) prepubertal stage, defined as Tanner breast stage I for girls and testicular volume less than 4 ml for boys (13); and 5) an uncomplicated neonatal period without signs of severe asphyxia (defined as Apgar score < 3 after 5 min), sepsis, or long-term complications of respiratory ventilation, such as bronchopulmonary dysplasia. Children with endocrine or metabolic disorders, chromosomal defects, syndromes, and growth failure caused by other conditions (e.g. emotional deprivation, severe chronic illness, chondrodysplasia) were excluded, with the exception of Silver-Russell syndrome. When we performed the analyses after exclusion of the Silver-Russell subjects, the results were the same. The following data have been published before: Si and β-cell function data from 28 of 77 children (4); body composition from 30 of 149 children (14, 15); BP, total cholesterol, high-density lipoprotein-cholesterol (HDL-c), low-density lipoprotein-cholesterol (LDL-c), and triglyceride (TG) levels of 159 children (4, 6, 16); and free fatty acid (FFA) levels of 28 children (4). However, these papers had another objective and did not focus on the influence of prematurity. The study was approved by the Medical Ethics Committees. Written informed consent was obtained from the parents or custodians of each child.
Study design
Children were divided into two groups on the basis of their gestational age: 1) preterm (gestational age < 36 wk), and 2) at term (gestational age 36 wk). The gestational age of the subjects was determined by ultrasound in the first trimester, if available, and otherwise calculated from the date of the last menstruation.
Anthropometry
Standing height was measured using a Harpenden stadiometer, and expressed as SDS for sex and chronological age using Dutch references (12). Body mass index (BMI) was calculated according to the formula weight/(height)2 and was expressed as SDS for sex and age (17). Systolic and diastolic BP was measured twice on the left arm with an automated device and appropriate cuff size (Dinamap Critikon; Southern Medical Corp., Baton Rouge, LA). The mean of two measurements was used for analysis. Because height is an important determinant of BP in childhood and adolescence, BP was expressed as SDS adjusted for height and sex (18).
Body composition
Dual-energy x-ray absorptiometry scans (type Lunar DPX-L; GE Healthcare, Madison, WI) were performed in a subgroup of 149 children. Lean body mass (LBM) and fat percentage were determined. All values were transformed into SDS for sex and chronological age using Dutch reference values for children, which were obtained using the same instrumentation and software (19, 20). Because body composition, particularly LBM, is strongly related to height, LBM expressed as SDS for age and sex might result in an underestimation in short stature. Therefore, LBM was also expressed as SDS for height and sex. Height-adjusted SDS was calculated as previously described (15).
Si and β-cell function
In a subgroup of 77 children, a modified frequently sampled iv glucose tolerance test (FSIGT) with tolbutamide was performed, as previously described (21, 22). Glucose and insulin levels were measured in all samples, and Si, glucose effectiveness (Sg), acute insulin response (AIR), and disposition index (DI) were calculated using Bergman’s Minimal Model Millennium software (23). Si quantifies the capacity of insulin to promote glucose disposal, and Sg reflects the capacity of glucose to mediate its own disposal. The AIR, an estimate of insulin secretory capacity, was measured as the area under the curve from 0–10 min corrected for baseline insulin levels. DI equals AIR x Si and indicates the degree of glucose homeostasis. Sex and ethnic distribution, gestational age, birth weight SDS, birth length SDS, birth head circumference SDS, height SDS, and BMI SDS were not significantly different between children who underwent an FSIGT and those who did not. Only age was slightly higher in the FSIGT group (7.3 vs. 6.7 yr).
Hormone and biochemical assays
All blood samples were taken after an overnight fast. Serum glucose and total cholesterol levels were measured as previously described (24). Insulin levels were all measured in one laboratory using the same method (immunoradiometric assay; Medgenix, Biosource Europe, Nivelles, Belgium). The intraassay coefficient of variation (CV) was 2–4.7%, and the interassay CV was 4.2–11.3%. TGs were measured on the Chem-1 analyzer (Technicon Instruments, Tarrytown, NY) according to the manufacturer’s instructions, and after 1998, on the Hitachi 917 analyzer (Roche Diagnostics, Mannheim, Germany) according to the manufacturer’s instructions. Both methods were comparable (y = x-0.030).
Nonesterified fatty acids (FFA) were measured in serum using an enzymatic colorimetric method (Wako Chemicals, Neuss, Germany). Apolipoprotein (apo) A-1, apo-B, and lipoprotein A [Lp(a)] were determined by rate nephelometry on the Immage Immunochemistry system (Beckman Coulter, Mijdrecht, Netherlands), according to the manufacturers’ instructions. Between-run CVs were 4.2, 2.8, and 6.9% for these lipoproteins at levels of 0.94, 0.53, and 0.35 g/liter respectively.
Statistical analysis
FSIGT parameters, TGs, FFAs, limb fat, and trunk fat were logarithmically transformed before analysis because of a skewed distribution. Because Lp(a) levels were very skewed, also after logarithmic transformation, we tested whether the percentage of individuals having a value of more than or equal to 0.3 g/liter was different between the groups. All data are presented as mean ± SD, except for the skewed parameters mentioned previously, which are presented as median and interquartile range. Differences between groups were tested using the Student’s t test for continuous variables and the 
2 test for categorical variables. Multiple linear regression (for continuous variables) and binary logistic regression analyses (for dichotomous variables) were used to adjust differences in outcome between the groups for possible baseline differences. Correlations were analyzed using Spearman’s correlation coefficient. With respect to body composition and BP data, a value of –2 to 2 SDS corresponds with a normal body composition corrected for age and sex, and a normal BP corrected for height and sex. The level of significance was determined at P < 0.05. Statistics were performed using the computer statistical package SPSS (12.0; SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Table 1
lists the clinical data of the total study group and the preterm and term subgroups. Birth weight SDS, birth length SDS, current height SDS, and BMI SDS were significantly lower than zero for both groups. Compared with term SGA children, preterm SGA children were significantly lighter and shorter at birth after correction for gestational age. Age and BMI SDS were significantly lower in preterm than term SGA children. Current height SDS was slightly higher in preterm SGA children than in term SGA children. Sex distribution and birth head circumference SDS were comparable.
|
Because preterm SGA children had a significantly lower BMI SDS, we were interested whether this was due to a lower fat or a lower muscle mass. Preterm SGA children had a significantly lower percent body fat SDS than term SGA children, also after adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, and birth length SDS) (Table 2). Both trunk fat and limb fat were significantly lower in preterm SGA children, though the difference for limb fat did not reach significance anymore after adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, birth length SDS, and total body weight). Fat distribution, defined as trunk fat to total fat ratio, was comparable between preterm and term SGA children. Lean mass SDS was comparable, also when height-adjusted instead of age-adjusted SDS was used.
|
Table 3 shows the results of the FSIGT tests. Si was comparable for preterm and term SGA children. Insulin secretion (AIR) tended to be higher, and Sg tended to be lower in preterm than term SGA children, but the difference was not statistically significant. DI was significantly higher in preterm than in term SGA children. After adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, and birth length SDS), Si, Sg, and AIR were comparable for both groups, but DI was still higher for preterm SGA children, though this did not reach significance (P = 0.066). After additional adjustment for percent body fat SDS and height SDS, both AIR and DI were significantly higher in preterm than in term SGA children.
|
BP
Preterm SGA children had a significantly higher systolic (P = 0.010) and diastolic (P < 0.0005) BP SDS than term SGA children (Table 4). This remained so after adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, birth length SDS, BMI SDS, and height SDS). In addition, the percentage of children with a high systolic BP was higher in preterm SGA children (26.3%) compared with term SGA children (16.8%).
|
Preterm SGA children had significantly lower total cholesterol levels than term SGA children, but the difference was small and disappeared after adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, birth length SDS, BMI SDS, and height SDS) (Table 4
). Serum levels of HDL-c, LDL-c, TG, FFA, apo-A1, apo-B, the apo-B to apo-A1 ratio, and the percentage children with a Lp(a) level more than or equal to 0.3 g/liter were similar for both groups, also after adjustment for possible confounding factors (age, sex, ethnicity, birth weight SDS, birth length SDS, BMI SDS, and height SDS).
Multiple regression analyses
Because we were also interested in the relative contribution of several parameters, such as anthropometry at birth and gestational age, to the various cardiovascular risk parameters, we performed backward multiple regression analyses on the total study group (Table 5). The multiple regression analyses indicated that gestational age was indeed a significant contributor to the variance in percent body fat SDS, insulin secretion, DI, and systolic and diastolic BP SDS. Gestational age was not a significant determinant of Si.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our study shows that premature SGA children had a more severe degree of growth retardation. Previously, we showed that SGA children with a more severe phenotype were more likely to be delivered by an elective cesarean section (25). It is likely that the preterm elective delivery of these SGA children was prompted by the severe growth retardation. Preterm SGA children had a lower percent body fat SDS than term SGA children. Because preterm SGA children also had a more severe degree of SGA, as indicated by their lower birth weight SDS and birth length SDS, we adjusted for these possible confounding factors. After adjustment for anthropometry at birth, premature SGA children still had a lower percent body fat SDS than term SGA children. Because the postnatal period of children born preterm is often characterized by nutritional problems (26), this might be held responsible for the reduced percentage of body fat in preterm SGA children.
Si was comparable between preterm and term SGA children, but percent body fat SDS, which is known to strongly correlate with Si, was lower in preterm SGA children. In addition, insulin secretion tended to be higher in preterm SGA children. Both findings might indicate a relative insulin resistance in preterm SGA children. However, after adjustment for percent body fat SDS, Si remained comparable between preterm and term SGA children. Therefore, we cannot conclude that preterm SGA children have a relative insulin resistance.
An unexpected finding was that the DI was higher in preterm than in term SGA children. The DI reflects the capacity of pancreatic islets to compensate for a lower Si (27). Our data are in line with those of the Dutch famine studies. Ravelli et al. (28) showed that glucose tolerance was worse in those that were exposed to famine during mid or late gestation. Hofman et al. (5, 29) also reported a higher insulin secretion and DI in preterm SGA children than in term SGA children, but unfortunately, they did not provide statistical comparisons. The difference in DI between preterm and term SGA children might be explained by the timing of prenatal growth retardation. All children in our study were born SGA. Therefore, the children that were born prematurely must have had retarded growth before the third trimester. Of the children born at term, it is not known when they had their growth retardation, but it seems likely that if growth was impaired early in pregnancy, these children would not have been born at term. The higher DI in preterm SGA children indicates that their insulin secretion is higher than would be expected from their level of Si. This could be due to either a greater insulin secretion or a lower hepatic insulin extraction (30). It has been suggested that a lower liver mass might disadvantageously affect hepatic insulin extraction, and it is not unthinkable that this is the case in preterm SGA subjects. Due to the wide range and the modest difference in DI between preterm and term SGA children, the clinical relevance is questionable. Nevertheless, our data show that short preterm SGA children had a glucose homeostasis that was not worse than that of term SGA children.
Preterm SGA children had a significantly higher systolic and diastolic BP. Notably, more than 25% of all premature SGA children had high systolic BP according to the modified Adult Treatment Panel III criteria for children (31). An elevated BP in childhood is known to be associated with an increased risk for the development of hypertension in adulthood (32). In previous studies, prematurity itself has been associated with increased BP (7, 8), regardless of the degree of SGA. However, other reports suggested that among preterms, only those born SGA had abnormalities in their BP and vascular function (9, 10). The etiology of the relationship between small size at birth and an elevated BP is still unclear, but there are several hypotheses. According to Brenner and Chertow (33), intrauterine growth retardation leads to a reduced number of nephrons. This may lead to a reduced filtration surface area, renal sodium retention, and, ultimately, hypertension. Rodríguez et al. (34) demonstrated in renal autopsy tissue that glomerulogenesis continues after preterm birth, but stops after 40 d. Preterm infants had less glomeruli than infants born at term. So, both intrauterine growth retardation and prematurity may lead to a reduced number of nephrons. In our study we demonstrate that both preterm and term SGA children have an increased systolic and diastolic BP but that preterm SGA children are more affected than term SGA children.
The multiple regression analyses indicated that gestational age was indeed a significant contributor to the variance in percent body fat SDS, insulin secretion, DI, and systolic and diastolic BP SDS. The regression models for insulin secretion and DI resulted in an explained variance of approximately 30%. Because the models for systolic and diastolic BP SDS explained not more than 7.0% of the total variation, we must, however, conclude that gestational age is not a major factor in determining childhood BP levels. A limitation is that we measured BP by two measurements, and this might not reflect the 24-h BP. Other parameters were measured in a subgroup, which means that the results might not be applicable to the total study population. However, subjects of the subgroups were recruited randomly and consecutively, and there was no evidence that these were different from the others.
In conclusion, within a population of short SGA children, preterm birth has divergent effects on several cardiovascular risk factors. Whereas preterm SGA children had a higher systolic and diastolic BP, they had a lower percent body fat and a higher insulin secretion and DI than term SGA children.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online November 20, 2007
Abbreviations: AIR, Acute insulin response; apo, apolipoprotein; BMI, body mass index; BP, blood pressure; CV, coefficient of variation; DI, disposition index; FFA, free fatty acid; FSIGT, frequently sampled iv glucose tolerance test; HDL-c, high-density lipoprotein-cholesterol; HOMA-IR, homeostasis model of assessment of insulin resistance; LBM, lean body mass; LDL-c, low-density lipoprotein-cholesterol; Lp(a), lipoprotein A; SDS, SD score; Sg, glucose effectiveness; SGA, small-for-gestational-age; Si, insulin sensitivity; TG, triglyceride.
Received August 27, 2007.
Accepted November 14, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. G Ijzerman, C. D A Stehouwer, E. H Serne, J. J Voordouw, Y. M Smulders, H. A Delemarre-van de Waal, and M. M van Weissenbruch Incorporation of the fasting free fatty acid concentration into quantitative insulin sensitivity check index improves its association with insulin sensitivity in adults, but not in children Eur. J. Endocrinol., January 1, 2009; 160(1): 59 - 64. [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 |
