| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Paediatrics (M.C., M.M.v.W., H.A.D.-v.d.W.), Institute for Clinical and Experimental Neuroscience, and Department of Obstetrics and Gynaecology (J.P.W.V.), VU University Medical Center, 1081 HV Amsterdam, The Netherlands; and Department of Epidemiology (F.E.v.L.), Netherlands Cancer Institute, 1066 CX Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Mirjam M. van Weissenbruch, VU University Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands. E-mail: m.vanweissenbruch{at}vumc.nl.
| Abstract |
|---|
|
|
|---|
Objective: Our objective was to investigate several cardiometabolic measures in 8- to 18-yr-old IVF singletons and spontaneously conceived controls born from subfertile parents.
Design and Setting: This follow-up study was conducted at the VU University medical center, Amsterdam, The Netherlands.
Participants: Blood pressure was examined in 225 IVF-conceived children and 225 age- and gender-matched spontaneously conceived control children. Several indicators of insulin resistance were studied in a pubertal subpopulation (131 IVF children and 131 controls).
Main Outcome Measures: Blood pressure, fasting glucose, and fasting insulin were determined.
Results: Systolic and diastolic blood pressure levels were higher in IVF children than controls (109 ± 11 vs. 105 ± 10 mm Hg, P < 0.001; and 61 ± 7 vs. 59 ± 7 mm Hg, P < 0.001, respectively). Children born after IVF were also more likely to be in the highest systolic and diastolic blood pressure quartiles (odds ratio = 2.1, 95% confidence interval 1.4, 3.3; odds ratio = 1.9, 95% confidence interval 1.2, 3.0, respectively). Furthermore, higher fasting glucose levels were observed in pubertal IVF children (5.0 ± 0.4 vs. 4.8 ± 0.4 mmol/liter in controls; P = 0.005). Blood pressure and fasting glucose differences could not be explained by current body size, birth weight, and other early life factors or by parental characteristics, including subfertility cause.
Conclusions: These findings highlight the importance of continued cardiometabolic monitoring of IVF-conceived children and might contribute to current knowledge about periconceptional influences and their consequences in later life.
| Introduction |
|---|
|
|
|---|
In humans, little is known about the effects of poor periconceptional and/or preimplantation environment on postnatal cardiovascular and metabolic functioning. Concerns have recently been raised about the children born after subfertility treatment (9). Accumulating evidence suggests that in vitro fertilization (IVF) singletons are at increased risk for adverse perinatal outcome (10, 11). It is still unclear whether the IVF process in humans could affect the vulnerable processes occurring during early embryonic development with long-term health consequences. Therefore, we studied postnatal growth and development in 8- to 18-yr-old children born from subfertile parents who were either successfully treated with IVF or conceived spontaneously. The main objective of the present study was to investigate blood pressure and indicators of insulin resistance in IVF and control children.
| Subjects and Methods |
|---|
|
|
|---|
The OMEGA study is a Dutch retrospective cohort study aimed to examine long-term health effects of hormone stimulation. The cohort consists of 26,428 women diagnosed with subfertility problems in one of the 12 IVF clinics between 1980 and 1995; 19,840 women received IVF treatment, and 6,588 women did not (12, 13, 14). Eligible women had not achieved conception after at least 1-yr frequent unprotected intercourse at their first visit to the fertility clinic. Risk factor questionnaires to the women and detailed data collection from the medical records provided information on the children born from the OMEGA participants up to 1996–1997. The questionnaire response rate was 73% among subfertile women with children. The present study was restricted to IVF and spontaneously conceived children born from OMEGA participants who were treated for subfertility in the VU University medical center (VUmc). IVF children born from women treated in the VUmc who did not participate in the OMEGA study were also eligible for recruitment.
Approach of study subjects
From the 553 eligible singletons born after standard IVF treatment, we invited 95% of IVF children born between 1986 and 1991, 74% of IVF children born between 1992 and 1993, and 41% of IVF children born between 1994 and 1995 to achieve equal representation of all 1-yr age categories. For each participating IVF child, we searched one control child of the same gender and similar age (
3-month age difference) born after spontaneous conception from subfertile parents. In case an approached control child did not want to participate, the control recruitment process was repeated until an appropriate control child was found that did agree to participate. Between 2003 and 2006, 354 IVF and 454 control children and their parents were informed by letter about our study on growth and development of IVF children (Fig. 1
). Finally, 69% of the approached IVF children and 51% of the approached controls agreed to participate, resulting in 233 matched pairs. Subsequently, pubertal children were recruited for additional research, including the collection of a fasting blood sample [female criteria of puberty:
stage 2 breast development; male criteria:
stage 2 genital development and/or testis volume
4 ml (15)]. Additional research was restricted to pubertal children only to avoid low participation rates, especially among the youngest children, due to the invasive character of these tests. In total, 80% of the pubertal children underwent a blood withdrawal. All children and their parents gave written informed consent to participate in the study.
|
The study protocol was approved by the ethics committee of the VUmc and the National Medical Ethics Committee known as the "Centrale Commissie Mensgebonden Onderzoek" located in The Hague, The Netherlands.
Data collection
Systolic and diastolic blood pressure was measured twice at the nondominant arm in the sitting position using an automatic device with appropriate cuff size (Dinamap PRO 100; Criticon, Munich, Germany). The first measurement was performed after a 30- to 45-min interview and the second measurement within a few minutes after the first one. The mean of these two readings was used in analyses. Body weight and height were assessed to the nearest 0.1 kg and 0.1 cm using an electronic scale (SECA, Hanover, MD) and a stadiometer (Holtain Ltd., Crymych, Dyfed, UK), respectively. Skinfold thickness measurements (triceps, biceps, subscapular, and suprailiac) were collected by a Harpenden caliper (Harpenden, West Sussex, UK). Other body fat measures have been reported elsewhere.
Blood samples were drawn between 0900 and 1000 h after an overnight fast. Fasting glucose and insulin were determined using the Gluco-quant Glucose/HK, Roche assay kit (Roche Diagnostics GmbH, Mannheim, Germany) and Bayer/ACS Centaur immunoassay (Bayer Diagnostics, Mijdrecht, The Netherlands), respectively. The glucose to insulin ratio and the homeostasis assessment model (HOMA) were chosen as measures of insulin sensitivity. HOMA insulin resistance and β-cell function were calculated according the original formula (16). Laboratory measurements were performed at the Department of Clinical Chemistry of the VUmc.
Before the follow-up visit in the VUmc, a questionnaire was sent to the parents to gather information on various demographical, lifestyle, and medical factors, including the cause of subfertility, parental education level, maternal smoking during pregnancy, and birth weight and gestational age of the respective child. Maternal BMI and highest level of education completed by either parent were used as indicators of socioeconomic conditions (17). Information about drug use of the child and family history of disease in terms of diabetes type 2, cardiovascular disease, and hypertension among parents and grandparents was obtained by an interview. None of the children used medication that could have affected blood pressure. Birth weight, either extracted from VUmc birth certificates (49%) or outpatient clinic reports (38%), or self-reported by the parents (13%), was expressed as the SD score (SDS) to correct for gestational age and gender (18).
Statistical analysis
After exclusion of eight matched pairs due to missing blood pressure measurements, data of 225 IVF-control pairs and data of the subsequent pubertal subset consisting of 131 unmatched IVF children and 131 controls were analyzed (Fig. 1
). Differences between IVF-control pairs were tested using the paired t test for continuous variables and McNemars test for dichotomous variables. Metabolic data of pubertal IVF and control children were compared after correction for age and gender. Logistic regression analyses were performed to estimate crude odds ratios for being in the highest quartile of several outcome parameters associated with IVF conception. Furthermore, potential confounders of the association between blood pressure and indicators of insulin sensitivity on the one hand and IVF conception on the other hand were examined separately by regression analysis (e.g. gender, current weight, birth weight, gestational age, parity, maternal smoking during pregnancy, parental education, parental age, maternal BMI, subfertility cause, and family history of disease). Factors that changed the crude difference in outcome between IVF and control children with more than 10% were considered as confounders and included in the final regression model. Reported P values were based on two-sided tests of significance.
| Results |
|---|
|
|
|---|
114.5 mm Hg) and 1.9 times more likely to be in the highest diastolic blood pressure quartile (
65.5 mm Hg) than controls [highest quartile vs. lowest three quartiles: 95% confidence interval (CI) 1.4, 3.3; 95% CI 1.2, 3.0, respectively]. IVF children had a significantly higher sum of skinfolds compared with controls (40.5 ± 20.4 vs. 36.9 ± 17.5 mm; P = 0.04). In addition, higher fasting glucose levels were observed in IVF children (5.0 ± 0.4 vs. 4.8 ± 0.4 mmol/liter in controls; P = 0.005). IVF-conceived children were 2.5 times more likely to be in the highest fasting glucose quartile (
5.2 mmol/liter) than controls (highest quartile vs. lowest quartile: 95% CI 1.2, 5.2). No significant differences in fasting insulin concentrations, insulin resistance measures, height, weight, and BMI were found between both study groups.
|
|
|
| Discussion |
|---|
|
|
|---|
In clinical practice, the 3- to 4-mm Hg higher systolic blood pressure and the 1- to 2-mm Hg higher diastolic blood pressure in IVF children may seem like small increases, but at a population level, these differences might have a major impact on public health. A slight increase in blood pressure is associated with a remarkably increased risk of developing cardiovascular disease. For instance, lowering mean systolic blood pressure in adults by 2 mm Hg corresponds to an 8% reduction in the risk of stroke (19). Furthermore, it cannot be excluded that increased blood pressure after IVF may be amplified throughout life because blood pressure is known to track from childhood into adult life (20).
Over the past years, cardiovascular developmental consequences and potentially underlying mechanisms after environmental manipulation during early prenatal development have been documented in both human and animal studies (3, 5, 21, 22). The Dutch famine study demonstrated that exposure to malnutrition during early pregnancy is associated with an increased risk of coronary heart disease in adult life (22). Periconceptional undernutrition has been associated with the precocious activation of the hypothalamo-pituitary-adrenal (HPA) axis (23, 24, 25, 26). Gardner et al. (27) recently reported minor influences on HPA axis function in young adult sheep after periconceptional undernutrition. It has been suggested that the early activation of the HPA axis may not only lead to inappropriate elevation of prostaglandin levels and early birth but may also be associated with further programming effects due to inappropriate exposure of the fetus to glucocorticoids (28). Other targets, like the renin-angiotensin system and the sympathoadrenal axis, have also been associated with developmental origins of nutritional or other influences on cardiovascular function (29). Due to the complexity of the cardiovascular system, it is unlikely that the relation between periconceptional insults and postnatal cardiovascular dysfunction originates from one single cause. Early prenatal developmental plasticity in relation to environmental stimuli has been reported to lead to changes in fetal development through changes in imprinted gene expression, nutrient and stress-related signaling pathways, or cell cycle and apoptotic rates (9). Further research is necessary to investigate the role of these pathways in the development of cardiovascular dysfunction after periconceptional insults. In addition, in view of the present study, it remains to be elucidated whether increased blood pressure among IVF children originates from early prenatal life adaptations mediated through neuroendocrinal pathways related to the HPA axis and/or through one of the unidentified mechanisms. However, increased blood pressure levels in IVF children were to a large extent independent of birth weight, suggesting that the underlying mechanisms can modify the cardiovascular system even without affecting size at birth. This is in line with previous studies examining associations between early life factors and blood pressure among offspring (22, 30). It is important to realize that birth weight is just a proxy for fetal growth. Prenatal environmental insults that may affect embryonic and/or fetal growth trajectories can result in altered postnatal physiology without an effect on birth weight (31). Similarly, exposure to early prenatal life effects may even induce developmental adaptations in organ development and function that are not accompanied by changes in fetal growth characteristics.
Although exposure to adverse prenatal conditions, especially during late gestation, has been linked to decreased glucose tolerance in adults (32), it is unclear whether early prenatal insults in humans can influence postnatal glucose metabolism. Fasting glucose levels studied in the present study are within the normal range, and the difference in fasting glucose levels between pubertal IVF children and controls is small, not accompanied by differences in fasting insulin levels and other related measures. However, in view of the observed differences in blood pressure and body fat composition between IVF children and controls, considerable research is necessary to investigate the hypothesis that further changes in glucose metabolism might manifest in later life. Gold standard assessments, i.e. euglycemic hyperinsulinemic clamp tests and hyperglycemic clamp tests, will be useful to additionally investigate insulin sensitivity and β-cell capacity in IVF-conceived offspring.
When interpreting our results, the strengths and limitations of our study need to be considered. The strengths of our study include the relatively large study size and the comparison group consisting of spontaneously conceived children born from subfertile parents. Furthermore, we collected various variables related to blood pressure and metabolism that provided the opportunity to examine blood pressure and fasting glucose differences between IVF children and controls while adjusting for these potentially confounding factors. It is possible that we have underestimated the true association between IVF and outcome parameters by adjusting for intermediate factors (e.g. sum of skinfolds in case of evaluated blood pressure). In addition, the slight attenuation of the blood pressure differences by adjusting for birth weight and gestational age was to be expected. IVF is known to be associated with lower birth weight and shorter gestational age (11, 33), whereas these factors themselves have been found to increase blood pressure. Another limitation is potential selection bias because our study was based on 56% (n = 450) of the total number of subjects approached (n = 808). However, nonparticipation analysis yielded no significant differences between participants and nonparticipants in anthropometric measures. The main reason for nonparticipation was unwillingness of the child to participate. In addition, it cannot be excluded that the degree of subfertility in those who conceived after IVF was more severe compared with those who conceived spontaneously and that this difference in subfertility contributed to the physiological abnormalities observed in IVF offspring. Nevertheless, most control mothers who participated in the present study were diagnosed with subfertility in an era (1982–1990) when IVF was not a routine procedure and was not ethically acceptable to many women. Moreover, estimated differences in blood pressure between IVF and control children were hardly affected by adjustment for parental subfertility causes, rendering residual confounding very unlikely. Similarly, the fasting glucose difference between IVF and control children remained statistically significant after adjustment for subfertility cause.
In conclusion, increased blood pressure and fasting glucose levels among IVF children could not be explained by current risk indicators, early life factors, and parental characteristics. Although underlying mechanisms are largely elusive, the periconceptional period of IVF-conceived children might be a critical time window during which cardiometabolic function can be perturbed. Before definitive conclusions can be drawn, our results need to be reproduced by other prospective follow-up studies. Nevertheless, our findings underscore the importance of the continuing worldwide monitoring of postnatal development of IVF children and contribute to the current understanding of periconceptional exposure effects with regard to the development of both short- and long-term consequences in humans.
| Footnotes |
|---|
First Published Online February 19, 2008
Abbreviations: BMI, Body mass index; CI, confidence interval; HOMA, homeostasis assessment model; HPA, hypothalamo-pituitary-adrenal; IVF, in vitro fertilization; SDS, SD score; VUmc, VU University medical center.
Received November 1, 2007.
Accepted February 12, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Wagenaar, M.M. van Weissenbruch, D.L. Knol, P.T. Cohen-Kettenis, H.A. Delemarre-van de Waal, and J. Huisman Information processing, attention and visual-motor function of adolescents born after in vitro fertilization compared with spontaneous conception Hum. Reprod., April 1, 2009; 24(4): 913 - 921. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ceelen, M. M. van Weissenbruch, J. P.W. Vermeiden, F. E. van Leeuwen, and H. A. Delemarre-van de Waal Pubertal development in children and adolescents born after IVF and spontaneous conception Hum. Reprod., December 1, 2008; 23(12): 2791 - 2798. [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 |