Body composition by dual-energy X-ray absorptiometry in women with previous pre-eclampsia or small-for-gestational-age offspring

Authors


  • Part of this research was presented as an oral presentation at the European Congress of the International Society for the Study of Hypertension in Pregnancy, 24–26 May 2007, Reykjavik, Iceland and at the annual meeting of the Society for Gynecologic Investigation, 26–29 March 2008, San Diego, CA, USA.

Dr AL Berends, Department of Obstetrics and Gynaecology, Erasmus MC, University Medical Centre Rotterdam, Room Number 2224a, PO Box 2040, 3000 CA Rotterdam, the Netherlands. Email a.berends@erasmusmc.nl

Abstract

Objective  To investigate differences in body composition and fat distribution between women with previous pre-eclampsia or small-for-gestational-age (SGA) offspring and those with uncomplicated pregnancies.

Design  Cohort study.

Setting  Population-based study in a genetically isolated population in the southwest of the Netherlands.

Population  Women after pregnancies complicated by pre-eclampsia (n=45), SGA offspring (n=53) and uncomplicated pregnancies (n=106).

Methods  Women were compared for body composition and fat distribution variables, assessed by dual-energy X-ray absorptiometry (DXA) and anthropometrics at a mean follow-up time of 10.8 (SD ±5.9) years after pregnancy.

Main outcome measures  Total lean and fat mass, android fat mass, gynoid fat mass, android-to-gynoid fat ratio, waist and hip circumference, waist-to-hip ratio.

Results  Women with previous pre-eclampsia compared with controls had higher mean total fat mass index (11.5 ± 0.6 versus 9.7 ± 0.4 kg/m2; P = 0.03), lean mass index (15.8 ± 0.3 versus 14.5 ± 0.2 kg/m2; P =0.001) and body mass index ([BMI]; 28.4 ± 0.8 versus 25.4 ± 0.5 kg/m2; P = 0.005). Their waist circumferences (90.7 ± 2.0 versus 78.5 ± 1.3 cm; P < 0.001) and waist-to-hip ratios (0.86 ± 0.01 versus 0.77 ± 0.01; P < 0.001) were also higher as well as android fat mass (2.8 ± 0.2 versus 2.1 ± 0.1 kg; P = 0.01) and android-to-gynoid fat ratios (0.45 ± 0.02 versus 0.39 ± 0.01; P = 0.02). Mean total fat, lean and BMI was not significantly different between women with previous SGA offspring and controls, yet waist-to-hip ratios (0.83 ± 0.01; P < 0.001) were higher. The observed differences in waist and hip circumference, waist-to-hip ratio and gynoid fat mass could not be attributed to differences in BMI.

Conclusion  Women with previous pre-eclampsia or SGA offspring pregnancies compared with those with uncomplicated pregnancies have a preferential fat accumulation in the abdominal over hip region, which may explain, at least partly, their increased cardiovascular risk.

Introduction

There is growing evidence for an increased risk of future cardiovascular disease in women with a history of pregnancies complicated by pre-eclampsia or small-for-gestational-age (SGA) offspring.1–4 It is hypothesised that such adverse pregnancy outcomes unmask women with a predisposition to vascular or metabolic disease.5 Pregnancy, therefore, provides a unique opportunity for early identification of women at increased cardiovascular and metabolic risk and renders the possibility for risk reduction by preventive strategies in this specific group of women. To optimise preventive strategies, it is essential to know which cardiovascular risk factors, especially those that are treatable and modifiable, are present in women with previous pre-eclampsia or SGA offspring.

Obesity6,7 and in particular abdominal obesity8–10 is well recognised as major risk factor of cardiovascular disease and type 2 diabetes. It has been suggested that particularly intra-abdominal (visceral) adipose tissue is strongly associated with metabolic disturbances and cardiovascular disease,11 but the exact pathophysiological mechanism underlying this association remains subject of discussion.12

Body mass index (BMI) is most commonly used as a measure to define obesity. A limitation of BMI, however, is its inability to assess whether excess of body weight is because of excessive adipose tissue or muscle hypertrophy. Anthropometric measurements as waist circumference and waist-to-hip ratio are widely used to estimate abdominal obesity. A drawback of these measurements, however, is that they cannot discriminate between visceral and subcutaneous fat, or can they discriminate between fat and lean mass. Moreover, anthropometric measurements are subject to intra- and interexaminer variations. An alternative, more accurate, method that does discriminate between fat and lean mass is dual-energy X-ray absorptiometry (DXA).13 Although DXA can also not distinguish between visceral and subcutaneous fat, new DXA measurements that quantify adipose tissue in the lower trunk region correlate strongly with visceral fat.14–16

In our study population, we found an increased prevalence of metabolic syndrome in women with previous pre-eclampsia or growth restricted babies, as reflected by SGA offspring, compared with women with uncomplicated pregnancies.17 In that respect, we performed waist circumference measurements, which suggested a higher prevalence of abdominal obesity in these women. Abdominal obesity may well be an important contributor to the increased cardiovascular disease risk observed in women with pregnancies complicated by pre-eclampsia or intrauterine growth restriction.3,18 More such detailed information on the characterisation of women at future risk of cardiovascular disease is essential for preventive strategies to be implemented by clinicians caring for these women. Therefore, this study aimed to further explore body composition as well as fat distribution by means of DXA, in addition to traditional anthropometric measurements, in women with a history of pre-eclampsia, women with SGA offspring, and uncomplicated pregnancies.

Materials and methods

Subjects

Women with a history of pre-eclampsia or SGA offspring and women with a history of uncomplicated pregnancies only were recruited from a genetically isolated population in the southwest of the Netherlands.19 This population was founded around 1750 by a limited number of individuals (∼150) and has been characterised by minimal inward migration and rapid population growth over the past two centuries. Descendants of this population show less genetic diversity than outbred populations.19 This study is part of a larger research programme called Genetic Research in Isolated Populations (GRIP), which aims to identify susceptibility genes for complex disorders.20 All participants were of Caucasian origin. The scientific protocol of GRIP was approved by the Medical Ethics Committee of the University Medical Centre Rotterdam. All participants provided informed consent. Recruitment of participants has been described in detail elsewhere.17 In brief, 197 women with pregnancies complicated by pre-eclampsia or with SGA offspring, living in the isolated population at the time of delivery, were identified from National Birth Registration Records (anonymous data). Of those, 57 could not be traced because of unknown identity or absence of medical records. From the remaining 140 women who were approached for the study, 106 agreed to participate (response rate 76%). Fifty women had a history of pre-eclampsia of whom 43 (95.6%) were nulliparous at index pregnancy, and 56 had SGA offspring of whom 35 (66%) were nulliparous at index pregnancy. An equal number of unmatched controls were randomly selected from the midwife’s practice, located within the same community. Women who gave birth to children with congenital anomalies were excluded from the study. Only singleton pregnancies were included. Pre-eclampsia was defined as de novo hypertension (systolic ≥140/diastolic ≥90 mmHg) with proteinuria ≥300 mg per 24 hours or at least 1+ on semiquantitative analysis after 20 weeks of gestation or as superimposed pre-eclampsia when new onset proteinuria after 20 weeks of gestation occurred in case of chronic hypertension. SGA was defined as birthweight of newborns, born to women without severe nutritional deficiency, equal to or below the 5th percentile for gestational age at delivery, according to the Dutch fetal growth charts of Kloosterman.21 Additionally, we calculated the birthweight percentiles of these SGA babies according to customised birthweight percentiles (www.gestation.net). These growth curves adjust for physiological factors that affect fetal growth, such as maternal height, weight in early pregnancy, parity and ethnicity as well as the sex of the baby, thereby limiting misclassification of constitutionally small babies. Maternal prepregnancy weight or weight in early pregnancy was known for 27 of the 56 (48.2%) women with SGA offspring. The customised birthweight percentiles in this subgroup were all equal or below the 3rd percentile and therefore met the criteria of SGA.

Data on body composition were available for 45 women with a history of pre-eclampsia (3 excluded because of pregnancy and 2 due to technical error) and for 53 women with SGA offspring (3 missing due to technical errors), and for 106 controls.

Data collection

Participants were invited for examination at our research centre located within the community. All participants were interviewed by the research physician about their medical history, medication use, educational level, smoking habits and alcohol consumption. Diabetes mellitus was defined as the use of blood glucose-lowering medication. Educational level was categorised into low (primary school/lower vocational training), intermediate (secondary school/intermediate vocational training) and high education (higher vocational training/university). Participants were classified as nonsmokers or current smokers (≥1 cigarette/day). Alcohol consumption was defined as regular use of alcoholic drinks (≥1 unit/week).

Anthropometric measurements

Height and weight were measured with the participant dressed in light underclothing. Waist and hip circumference were measured on uncovered skin using a tape measure with the participant in upright position. Waist circumference was measured halfway between the rib cage and the pelvic bone. Hip circumference was measured at the maximal circumference of the hips. Waist-to-hip ratio was calculated from these measurements.

Dual-energy DXA

Total body and regional fat mass, lean mass and bone mineral content were obtained from DXA scans performed using a Prodigy™ total body fan-beam densitometer and analysed with the enCORE™ 2005 software V. 9.30.044 (GE Lunar Corporation, Madison, WI, USA). Total body scans were auto analysed by the software, which employs an algorithm that divides body measurements into areas corresponding to head, trunk, arms and legs. The trunk region was limited by an upper horizontal border below the chin (neck cut), vertical borders lateral to the ribs, and a lower border by the iliac crest. The arm region was limited by cuts that cross the arm sockets as close to the body as possible and separate the arms and hands from the body. The leg region is limited above by the oblique lines passing through the hip joint, and cuts that separate the hands and forearms from the legs and a centre leg cut that separates the right and left leg. Two additional regions were defined using the software provided by the manufacturer; the ‘android’ and ‘gynoid’ region (Figure 1). The ‘android region’ has a lower boundary at the pelvis cut and the upper boundary above the pelvis cut by 20% of the distance between the pelvis and neck cuts. The lateral boundaries are the arm cuts. The ‘gynoid region’ has an upper boundary between the upper part of the greater trochanters and a lower boundary defined at a distance equal to twice the height of the ‘android region’. The lateral boundaries are the outer leg cuts. The android and gynoid fat mass and android-to-gynoid fat ratio were calculated from these measurements. All analyses were verified by a trained technician who performed adjustments when necessary. Daily quality assurance tests were performed with a calibration block supplied by the manufacturer. Repeated measurements on the calibration block had coefficients of variation less than 1%.

Figure 1.

Two examples of DXA scans showing the special regions of interest corresponding with android fat mass and gynoid fat mass. (A) Gynoid fat deposition and (B) android fat deposition.

Definitions

Total body weight is the sum of total fat and lean mass and bone mineral content. BMI is total body weight divided by height squared. We calculated the following variables to describe body composition.

Fat mass index = total fat mass (kg)/height2 (m2)

Lean mass index = total lean mass (kg)/height2 (m2)

Total body fat percentage was defined as: (total fat mass × 100%)/(total fat mass + total lean mass + total bone mineral content).

Statistical analyses

General characteristics were compared between groups using Dunnett’s test, chi-square statistics and Fisher’s exact test where appropriate. Differences in body composition and fat distribution were analysed using a general linear model controlling for age, time interval between pregnancy and study, smoking status and educational level. Postmenopausal women were excluded from these analyses given the changes in body composition associated with this condition (one with prior pre-eclampsia, four with SGA offspring and five controls). Next, the relation between BMI and fat distribution variables (waist and hip circumference, waist-to-hip ratio, android fat, gynoid fat and android-to-gynoid fat ratio) was investigated using linear regression analysis with BMI as independent variable and the fat distribution variables as dependent variables. The analyses were adjusted for age, time interval between pregnancy and study, smoking status and educational level. A possible effect of pre-eclampsia or pregnancy with SGA offspring (disease status) in the relation between BMI and fat distribution variables was evaluated by adding an interaction term BMI × disease status to the model. Finally, we repeated our initial analyses on fat distribution with additional adjustment for BMI. For all statistical analyses, we used SPSS for Windows, version 11.0.1.

Results

General characteristics of participants are given in Table 1. At index pregnancy, women with pregnancies complicated by pre-eclampsia or with SGA offspring were on average 3 years older than controls. At the time of the study, women with a history of pre-eclampsia were younger than women with SGA offspring and controls. Accordingly, the time interval between delivery and study was significantly shorter in women with prior pre-eclampsia and SGA offspring compared with controls. Other significant differences between groups were found for educational level, use of antihypertensive medication and current smoking (Table 1). Hypertension defined as systolic blood pressure ≥140 mmHg and/or diastolic ≥90 mmHg and/or use of antihypertensive medication was significantly more prevalent among women with prior complicated pregnancies compared with controls. No significant differences were found between groups in use of lipid-lowering medication or diabetes mellitus (type 1 and 2).

Table 1.  General characteristics of women with previous pre-eclampsia, SGA offspring and controls
 Pre-eclampsia (n = 45)SGA (n = 53)Controls (n = 106)
  • Data are presented as means ± SD or percentages (absolute numbers). Differences between women with previous pre-eclampsia or SGA offspring and controls were examined with Dunnett’s test for continuous variables and with chi-square statistics or Fisher’s exact test for dichotomous variables.

  • * P < 0.001, **P < 0.05 and ***P < 0.01 compared with controls.

  • ****

    Hypertension defined as systolic blood pressure ≥140 mmHg and/or diastolic ≥90 mmHg and/or use of antihypertensive medication.

Index pregnancy
Age, years29.2 ± 3.9*29.9 ± 3.6*26.2 ± 4.3
Birthweight newborn, g2562 ± 902*2215 ± 555*3345 ± 379
Gestational age, weeks37 ± 3.7*38.6 ± 2.9**39.9 ± 1.4
Chronic hypertension6.7 (3)**5.7 (3)**0
Gestational diabetes4.4 (2)00
Current study
Age, years36.1 ± 5.5***39.3 ± 5.139.2 ± 5.6
Time since pregnancy, years7.0 ± 5.3*9.5 ± 5.1*13.1 ± 5.7
Premenopausal, %97.8 (44)92.5 (49)95.3 (101)
Educational level
 Low, %40 (18)*71.7 (38)72.6 (77)
 Intermediate, %46.7 (21)**18.9 (10)25.5 (27)
 High, %13.3 (6)***9.4 (5)**1.9 (2)
Hypertension,**** %46.7 (21)*28.3 (15)***8.5 (9)
Anti hypertensive drugs, %20 (9)*13.2 (7)***0.9 (1)
Lipid-lowering drugs, %2.2 (1)1.9 (1)0.9 (1)
Diabetes mellitus, %4.4 (2)3.8 (2)0
Current smoking, %22.2 (10)***54.7 (29)49.1 (52)
Alcohol consumption, %31.1 (14)28.3 (15)31.3 (33)

Anthropometric measurements

Weight, BMI, waist circumference and waist-to-hip ratio were higher in women with previous pre-eclampsia compared with controls (Table 2). In contrast, women who had SGA offspring did not differ significantly from controls with respect to body weight, BMI and waist circumferences. However, their hip circumferences were smaller and subsequently their waist-to-hip ratios were significantly higher than in controls (Table 2).

Table 2.  Body composition and fat distribution of women with previous pre-eclampsia or SGA offspring compared with controls
 Pre-eclampsia (n = 44)*P**SGA (n = 49)*P**Controls (n = 101)*
  • Data presented as adjusted means (SE).

  • *

    Postmenopausal women were excluded from analyses.

  • **

    Differences between women with previous pre-eclampsia or SGA offspring and controls were examined in a general linear model adjusted for age, interval between pregnancy study, smoking and educational level.

Antropometric measurements
Height (cm)164.2 (1.0)0.9162.6 (0.9)0.1164.4 (0.7)
Weight (kg)76.7 (2.4)0.0165.4 (2.2)0.268.8 (1.6)
BMI (kg/m2)28.4 (0.8)0.00524.8 (0.8)0.525.4 (0.5)
Waist circumference (cm)90.7 (2.0)<0.00181.2 (1.8)0.278.5 (1.3)
Hip circumference (cm)104.5 (1.8)0.397.7 (1.6)0.03102.0 (1.1)
Waist-to-hip ratio0.86 (0.01)<0.0010.83 (0.01)<0.0010.77 (0.01)
DXA measurements
Lean mass (kg)42.6 (0.9)0.00439.0 (0.8)0.739.4 (0.6)
Lean mass index (kg/m2)15.8 (0.3)0.00114.8 (0.3)0.414.5 (0.2)
Fat mass (kg)31.0 (1.7)0.0423.2 (1.5)0.126.4 (1.1)
Fat mass index (kg/m2)11.5 (0.6)0.038.8 (0.6)0.29.7 (0.4)
Android fat (kg)2.8 (0.2)0.012.1 (0.2)0.82.1 (0.1)
Gynoid fat (kg)5.8 (0.3)0.14.8 (0.2)0.15.3 (0.2)
Android-to-gynoid fat ratio0.45 (0.02)0.020.41 (0.02)0.30.39 (0.01)
Total body fat %39.1 (1.2)0.334.8 (1.1)0.0637.5 (0.8)

Dual-energy DXA

Women with a history of pre-eclampsia had an excess of total lean and fat mass compared with controls. After controlling for height, these differences remained significant, reflected by the higher lean and fat mass indices (Table 2). With respect to fat distribution, women with previous pre-eclampsia had an excess of fat in the android region and subsequently higher android-to-gynoid fat ratios than controls, whereas the total body fat percentage was not significantly different (Table 2). No significant differences were detected in body composition and fat distribution between women with SGA offspring and controls, except for borderline significant lower total body fat percentage (Table 2).

Fat distribution in relation to BMI

Next, we investigated for each group of women separately, the relation between BMI and the fat distribution variables: waist circumference, hip circumference, waist-to-hip ratio, android fat, gynoid fat and android-to-gynoid fat ratio. All fat distribution variables were positively associated with BMI (all P < 0.01). Slopes are given in Table 3. The associations between hip circumference and gynoid fat mass on the on hand and BMI on the other hand were significantly different in women with prior pre-eclampsia than in controls, as indicated by the significant interaction terms (Table 3). The association between BMI and gynoid fat was also significantly different in women with SGA offspring (Table 3).

Table 3.  Association between fat distribution variables and BMI stratified for previous disease status
Fat distribution variablesDisease statusSlope* (unit/kg/m2)Effect of disease status* BMI**P value***
  • *

    Values based on linear regression models with fat distribution variables as dependent variables and BMI as independent variable, adjusted for age, interval between pregnancy study, smoking and educational level.

  • **

    Regression coefficient indicating interaction effect of previous disease and BMI on fat distribution variables.

  • ***

    P value for the comparison with controls.

Waist circumferencePre-eclampsia2.30.260.2
SGA2.30.290.2
Controls2.0 
Hip circumferencePre-eclampsia1.7−0.40.01
SGA2.1−0.040.8
Controls2.0 
Waist-to-hip ratioPre-eclampsia0.0070.0040.06
SGA0.0050.0020.8
Controls0.004 
Android fatPre-eclampsia0.21−0.020.2
SGA0.21−0.020.2
Controls0.23 
Gynoid fatPre-eclampsia0.24−0.1<0.001
SGA0.27−0.090.002
Controls0.34 
Android-to-gynoid fat ratioPre-eclampsia0.0170.00060.8
SGA0.0160.00050.9
Controls0.017 

Given the positive association between BMI and the fat distribution variables, we repeated our analyses on fat distribution with additional adjustments for BMI. After controlling for BMI, both women with a history of pre-eclampsia and those with SGA offspring showed narrower hips, larger waist circumferences and waist-to-hip ratios (Table 4). In addition, we found that both women with previous pre-eclampsia and SGA offspring had less fat deposition in the gynoid region (Table 4).

Table 4.  Fat distribution of women with previous pre-eclampsia or SGA offspring compared to controls adjusted for BMI
 Pre-eclampsia (n = 44)*P**SGA (n = 49)*P**Controls (n = 101)*
  • Data are presented as adjusted means (SE).

  • *

    Postmenopausal women were excluded from analyses.

  • **

    Differences between women with previous pre-eclampsia or SGA offspring and controls were examined in a general linear model adjusted for age, interval between pregnancy study, smoking, educational level and BMI.

Antropometric measurements
Waist circumference (cm)85.4 (0.9)<0.00183.6 (0.8)<0.00179.6 (0.6)
Hip circumference (cm)99.7 (0.8)0.00199.8 (0.7)<0.001103.1 (0.5)
Waist-to-hip ratio0.85 (0.01)<0.0010.84 (0.01)<0.0010.77 (0.01)
DXA measurements
Android fat (kg)2.2 (0.08)0.82.3 (0.07)0.52.2 (0.05)
Gynoid fat (kg)5.1 (0.1)<0.055.1 (0.1)0.035.5 (0.09)
Android-to-gynoid fat ratio0.41 (0.02)0.50.43 (0.02)0.080.39 (0.01)

Discussion

In this study, women with pregnancies complicated by pre-eclampsia or with SGA offspring show an unfavourable fat distribution in comparison with women with uncomplicated pregnancies, marked by an excess of fat deposition in the abdominal region relative to fat deposition in the hip region. These differences could not be explained by differences in BMI.

As previously described,22,23 we found that women with a past history of pre-eclampsia have greater BMI compared with women with uncomplicated pregnancies. The association between obesity and pre-eclampsia has been well established.24–26 Little attention, however, has been paid to the actual composition of body weight in these women, despite the fact that it is excessive fat deposition rather than body weight per se that is independently associated to cardiovascular disease.27 The few studies that did investigate body composition in relation to pre-eclampsia by means of bioimpedance analysis or DXA were performed during pregnancy or very shortly thereafter.28–30 Therefore, these studies do not answer questions on body composition in these women in a nonpregnant state.

We found that women with previous pre-eclampsia have an excess of total fat mass accompanied by an increase of lean mass, resulting in a higher BMI compared with controls. Total body fat percentage was slightly, yet not significantly, higher. With regard to fat distribution, both DXA and anthropometric measurements revealed increased abdominal fat deposition in these women compared with controls, which is consistent with previously reported anthropometric data.22,23 Women who had SGA offspring did not differ significantly in total fat and lean mass from our controls. Unlike previous studies,24,31 we did not observe a significantly lower BMI in these women. However, women with SGA offspring may have lower body fat percentage, although the difference from controls was only significant at P = 0.06. Remarkably, we found that these women, despite their leanness, had an increased abdominal fat deposition reflected by higher waist-to-hip circumferences. We are aware of one other study examining fat distribution in women with low birthweight offspring at a large interval postpregnancy.32 These women, aged 70–79 years, had larger abdominal circumferences in comparison with women with normal birthweight offspring after adjusting for BMI.32

We evaluated whether the differences observed in fat distribution in both women with a history of pre-eclampsia or with SGA offspring were attributable to a possible confounding effect of BMI, by adjusting our analyses for BMI. Also then, our data indicated a preferential accumulation of fat in the abdominal over hip region.

With respect to cardiovascular disease, waist circumference and waist-to-hip ratio are well established as independent predictors,9,10 whereas such an association has not been studied yet for android fat mass measured by DXA. Therefore, our results strongly suggest that both women with a history of pre-eclampsia and those with SGA offspring are at increased risk of cardiovascular disease. The ultimate confirmation of this increased risk would have been to investigate cardiovascular disease as outcome in these women, yet due to the overall low incidence of cardiovascular disease in young women and due to limited numbers, this was not feasible. However, we previously described in this cohort that 40% and almost 30% of the women with prior pre-eclampsia and SGA offspring, respectively, were diagnosed with chronic hypertension within a decade after the index pregnancy.17 Furthermore, metabolic syndrome was diagnosed in nearly 40 and 20% of women with prior pre-eclampsia and SGA offspring, respectively.17

Whether the differences in body composition and fat distribution between women with previous pre-eclampsia or SGA offspring and women with uncomplicated pregnancies are causally related to pregnancy complications or are a result from these cannot be concluded from this study, as no preconceptional data are available. However, body composition and fat distribution are largely influenced by genetic factors33,34 and therefore it is likely that differences were present before pregnancy. Differences in body composition and fat distribution may be a possible explanation for the distinct clinical manifestations observed in women with pre-eclampsia and SGA offspring during pregnancy. Pregnancy in obese women is associated with endothelial impairment and an exaggerated proinflammatory status when compared with pregnancy in lean women.35

Several other factors may influence fat distribution like age, hormonal status, and lifestyle and may therefore potentially confound our analyses.34 Total and abdominal fat mass increase with age34,36 and postmenopausal status is associated with abdominal obesity.37 As we controlled for age in our analyses and as the far majority of women were premenopausal, we expect no confounding effects of these factors. Moreover, women with a history of pre-eclampsia were younger than controls implicating an underestimation rather than overestimation of the observed difference. Additionally, we adjusted for smoking status since smoking is associated with lower BMI and induces abdominal obesity.38 A limitation of this study is that we could not control for other relevant lifestyle factors such as dietary habits and physical activity, as these data were not available. Another limitation is that we have not controlled for parity, while women with succeeding pregnancies have a tendency to abdominal obesity.39 However, we speculate that parity is higher in women with a history of uncomplicated pregnancies than in women who encountered problems during pregnancy. Therefore, adjusting for parity may even magnify the observed difference.

Our study was conducted in a genetically isolated population. It has been previously demonstrated for this population that individuals are genetically more homogeneous than individuals of outbred population.19 Findings in this population may therefore reflect a common underlying genetic predisposition. It can be questioned whether these findings can be entirely generalised to the population at large. However, because our population is of more recent isolation, the genetic makeup may more closely resemble that of the general population.40 Furthermore, our simulation studies based on the genealogy have shown that this potential problem concerns primarily rare variants. For common genetic variants, our simulation studies show that no substantial differences between isolate and the general population are expected.40 Finally, our findings are in line with previous studies in outbred populations supporting the generalisability of our findings.22,32

In summary, the present study demonstrates that women after pregnancies complicated by pre-eclampsia or women with SGA offspring have a preferential accumulation of fat in the abdominal over the hip region. As abdominal obesity is an independent risk factor of cardiovascular disease,10 it may be an important contributor to the increased cardiovascular risk observed in these women.18

The findings of our study may help to design individually tailored optimal preventive strategies with regard to future cardiovascular disease. Especially strategies targeting accumulation of excessive abdominal fat, including diet and physical exercise41 might be relevant not only for women with previous pre-eclampsia who are more commonly overweight but also for women who had SGA offspring, despite their tendency towards lower BMI.

Disclosure of interest

None.

Contribution to authorship

A.L.B. was responsible for collection, analysis and interpretation of the data and for drafting the manuscript. M.C.Z. and F.R. contributed significantly to the interpretation of the data and to the manuscript revisions. C.J.d.G., B.A.O., C.M.v.D. and E.A.P.S. designed the study and contributed significantly to the interpretation of the data and to the manuscript revisions. All authors gave final approval of the version to be published.

Details of ethics approval

The Medical Ethics Committee of the University Medical Centre Rotterdam approved the scientific protocol of the study. Date of approval: 17 June 2004, reference number: 2004-031.

Funding

This study was sponsored by Erasmus Medical Centre Grant and Centre for Medical Systems Biology (CMSB).

Acknowledgments

The authors thank the participants of the study. Wilma Keller is acknowledged for her contribution to the data collection.

Commentary on ‘Body composition by dual energy-X-ray absorptiometry in women with previous pre-eclampsia or small-for-gestational-age offspring’

The intrauterine environment is a powerful determinant of fetal wellbeing and offspring health in adult life. How a woman’s health and biological characteristics shape this unique milieu remains poorly understood. In this issue of the Journal, Berends et al.1 show that when studied several years post-pregnancy, women with previous pre-eclampsia or small-for-gestational age offspring have preferential accumulation of abdominal fat.1,2 In the group with previous pre-eclampsia abdominal adiposity was indicated by compartmental DEXA as well as by anthropometric measurements, namely waist and hip circumferences; in the small-for-gestational age group the strength of the association was weaker and only indicated by a raised waist-to-hip ratio. This report is one of a series from a large research programme that aims to identify susceptibility genes for complex disorders. The authors suggest that their findings might help design preventive stratagems for cardiovascular disease in women. Although conducted in a genetically isolated population, which, in many respects, is a strength of the study, before this undoubtedly important goal can be realised it will be necessary to replicate the findings in other populations in order to be certain that the association does not represent a genotype-specific interaction between pre-eclampsia and abdominal adiposity. The conclusions must also be viewed with caution, as there appear to have been no a priori hypotheses relating to body composition, and many adjustments were included in what must be regarded as exploratory analyses. The authors offer only passing discussion of possible biological determinants. They have previously reported an increase in components of the metabolic syndrome in women with pre-eclampsia or small-for-gestational age offspring.1 Abdominal adiposity is well recognised as being strongly associated with the metabolic syndrome; perhaps pre-eclampsia is an additional manifestation of this phenotype. In this present study the prevalence of hypertension was increased in the complicated pregnancy groups, but there were no differences in the use of lipid-lowering medications or diabetes. A curious observation that is largely not discussed was the finding that lean body mass was also increased, in addition to fat mass, in the pre-eclampsia group. These authors use the terms small-for-gestational age and intrauterine growth restriction interchangeably. Although to mention this may seem mere quibbling, attention to the pattern of fetal growth might offer insight into the stage of pregnancy most affected and point to possible biological mediators. As would be anticipated, infant birthweight in the pre-eclampsia group was also significantly lower than in the control group. If a common biological pathway to pre-eclampsia and compromised fetal growth is being postulated, it might have been advantageous, and would have improved the power of the study, to have treated infant birthweight as a continuous rather than a categorical variable and as an outcome rather than an index of maternal disease. In their previous related report these authors describe intergenerational similarities in indices of cardiovascular risk in women with previous pre-eclampsia or small-for-gestational age infants and their parents.1 It is therefore relevant to ask about infant outcomes.

Newborn body composition may be a biomarker for later disease risk3,4 and information about infant anthropometry, body composition, insulin sensitivity and cardiovascular development would be of great interest. This is an intriguing and important hypothesis-generating study that should pave the way for a series of future-focused investigations.

N Modi
Section of Neonatal Medicine, Division of Medicine, Chelsea & Westminster campus, Imperial College London, London, UK

  • 1Berends A et al. Hypertension 2008;51:993–4.
  • 2Berends AL, Zillikens MC, de Groot CJM, Rivadeneira F, Oostra BA, van Duijn CM, Steegers EAP. Body composition by dual energy-X-ray absorptiometry in women with previous pre-eclampsia or small for gestational age offspring. BJOG 2008;116:442–450.
  • 3Modi N, Thomas EL, Harrington TA, Uthaya S, Dore CJ, Bell JD. Determinants of adiposity during preweaning postnatal growth in appropriately grown and growth-restricted term infants. Pediatr Res 2006;60:345–8.
  • 4Uthaya S, Thomas EL, Hamilton G, Dore CJ, Bell JD, Modi N. Altered adiposity after extremely preterm birth. Pediatr Res 2005;57:211–15.

Ancillary