SEARCH

SEARCH BY CITATION

Keywords:

  • Body weight;
  • congenital abnormalities;
  • congenital heart disease;
  • prenatal diagnosis;
  • ultrasonography

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Please cite this paper as: Best K, Tennant P, Bell R, Rankin J. Impact of maternal body mass index on the antenatal detection of congenital anomalies. BJOG 2012;119:1503–1511.

Objective  To investigate the association between maternal body mass index (BMI) and antenatal ultrasound detection of congenital anomalies.

Design  Population-based register study.

Setting  North of England (UK).

Population  All pregnancies (n = 3096) associated with a congenital anomaly notified to the Northern Congenital Abnormality Survey (NorCAS) during 2006–2009. Cases with chromosomal and teratogenic anomalies (n = 611) or without information on antenatal scanning (n = 4) were excluded.

Methods  Adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for antenatal detection according to maternal BMI categories were estimated using logistic regression.

Main outcome measures  For all anomalies combined, cases were defined as ‘detected’ if any congenital anomaly was suspected antenatally. Organ system-specific anomalies were defined as detected if an anomaly of the correct system was suspected.

Results  Antenatal detection of any anomaly occurred in 1146 of 2483 (46.2%) cases with normal karyotype. The odds of detection were significantly decreased in obese (BMI ≥ 30 kg/m2) women compared with women of recommended BMI (18.5–24.9 kg/m2; aOR, 0.77; 95% CI, 0.60–0.99; P = 0.046). Cardiovascular system anomalies were suspected antenatally in 109 of 945 (11.5%) cases. The odds of detecting a cardiovascular anomaly were significantly greater in underweight women (BMI < 18.5 kg/m2) than in women of recommended BMI (aOR, 2.95; 95% CI, 1.13–7.70; P = 0.027). There was no association between BMI and detection in any other organ system or between BMI and termination of pregnancy for fetal anomaly.

Conclusions  Antenatal ultrasound detection of a congenital anomaly is decreased in obese pregnant women. This has implications for the scanning and counselling of obese women.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

The proportion of overweight and obese women of childbearing age is increasing, with first-trimester obesity [body mass index (BMI) ≥ 30 kg/m2] more than doubling from 7.6% in 1989 to 15.6% in 2007 in England.1 This has implications for the prevalence of congenital anomalies, some of which occur more frequently in overweight (BMI = 25–29.9 kg/m2) and obese pregnant women.2,3

Antenatal detection of congenital anomalies gives the opportunity to prepare parents for the birth of a child with a congenital anomaly, to plan postnatal management or to consider termination of pregnancy. However, between 2005 and 2009, antenatal diagnosis occurred in only 47% of nonchromosomal cases notified to UK congenital anomaly registers.4

Evidence suggests that the sensitivity of antenatal ultrasound detection of congenital anomalies is further reduced with increasing BMI.5,6 Dashe et al.6 found a decreasing trend in ultrasound detection rates for congenital anomaly as BMI at the first antenatal visit increased. Similarly, Tabor et al.5 found that women with a BMI > 25 kg/m2 had a significantly lower detection rate than women with a BMI ≤ 25 kg/m2. Although both were large cohort studies, they only included 181 and 100 cases of congenital anomaly respectively, and neither investigated trends in specific anomaly groups.

Several studies have found associations between increased maternal BMI and suboptimal visualisation of the fetus.7–12 Visualisation of cardiac structures12,13 and soft tissues12 has been shown to be particularly impaired with increasing BMI. This association may partly explain the increased prevalence of congenital anomalies at birth in overweight and obese women,2 if a lower proportion are detected antenatally and, subsequently, fewer cases result in the termination of pregnancy, but this hypothesis has not been investigated.

This study investigated the association between maternal BMI and the antenatal ultrasound detection of congenital anomalies and between maternal BMI and pregnancy outcome in a population-based case series from the north of England.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Study population

The Northern Congenital Abnormality Survey (NorCAS) is a population-based register of congenital anomalies established in 1985. NorCAS prospectively collects data on congenital anomalies in mothers residing in the north of England (the North East and North Cumbria). This is a stable population with approximately 32 000 births per year. Data are collected on cases that occur in late miscarriages (20–23 weeks of gestation), terminations of pregnancy following antenatal diagnosis of a fetal anomaly (at any gestation), stillbirths (≥24 weeks of gestation) or live births. All anomalies are coded using the World Health Organization (WHO) International Classification of Diseases, Version Ten (ICD-10) and categorised according to the European Surveillance of Congenital Anomalies (EUROCAT) guidelines.14 NorCAS allows up to six anomaly subtypes to be recorded for each case. Minor anomalies, such as syndactyly (between toes two and three) or tongue tie, were excluded from all analyses. Further information on minor anomalies and coding is available in the EUROCAT guidelines (http://www.eurocat-network.eu/content/EUROCAT-Guide-1.3.pdf). To ensure a high case ascertainment, congenital anomalies are notified to the register from a variety of sources, including antenatal ultrasound departments, fetal medicine records, cytogenetic laboratories, the regional cardiology centre, pathology departments and paediatric surgery departments. Cardiovascular anomalies are confirmed by surgery, echocardiography, cardiac catheterisation or autopsy. Anomalies suspected antenatally are followed up after delivery and, if an anomaly is confirmed, both antenatal and final diagnoses are recorded.

All cases with a confirmed congenital anomaly notified to NorCAS, delivered between 1 January 2006 and 31 December 2009, were included in this study. Isolated cases (defined as occurring alone, or if all coexisting anomalies are commonly associated with the main anomaly) were assigned to groups depending on the organ system with which they were associated. Cases with two or more major anomalies from different organ systems were categorised as multiple anomalies.

Pregnancies associated with chromosomal anomalies and teratogenic syndromes were excluded as they are not primarily detected using the routine second-trimester ultrasound scan (generally offered between 18 and 21 weeks of gestation in the UK). Information regarding antenatal suspicion of a congenital anomaly was recorded from the routine second-trimester ultrasound scan, the subsequent anomaly scan or occasionally, the initial dating scan (carried out at around 10 to 12 weeks of gestation). Data may have come from any of these scans, but NorCAS only records gestational age at final antenatal diagnosis, which does not always correspond to the gestational age at first antenatal suspicion.

At the first antenatal visit, self-reported or measured maternal height and weight were documented in the mother’s medical records. These were then reported on the NorCAS notification form and BMI was derived as weight (kg)/[height (m)2]. Denominator data for total births (live and stillbirths) and live births for the same years, used to calculate prevalence, were obtained from the Office for National Statistics.

Information on maternal pre-gestational diabetes status was derived from the Northern Survey of Diabetes in Pregnancy (NorDIP),15 a collaborative survey of all pregnancies in women with diabetes at least six months before pregnancy.

Information on multiple pregnancies was derived from the Northern Survey of Twin and Multiple Pregnancy (NorSTAMP),16 a register of all multiple pregnancies occurring in the region.

Statistical analyses

Maternal BMI (kg/m2) was categorised as underweight (<18.5 kg/m2), recommended BMI (18.5–24.9 kg/m2), overweight (25–29.9 kg/m2) and obese (≥30 kg/m2) according to the WHO guidelines. Maternal age at delivery (years), gestational age at final antenatal diagnosis (weeks) and gestational age at first antenatal visit (weeks) were examined as continuous variables. The Index of Multiple Deprivation (IMD) 2007, determined from the maternal residential postcode at delivery, was used as a proxy measure of individual deprivation. The IMD is an area-level measure of deprivation compiled from data across seven domains: income, employment, health deprivation and disability, education skills and training, barriers to housing and services, crime and living environment, and is the UK government’s preferred area-based measure of deprivation.17 IMD scores were ranked and divided into tertiles and treated as a categorical variable, where the lowest tertile represents the least deprived women and the highest tertile the most deprived women.

For all anomalies combined, cases were defined as ‘detected’ if any EUROCAT-classified congenital anomaly was suspected antenatally. Organ system-specific anomalies were categorised as detected if any anomaly of the correct system was suspected. The Cuzick’s test for trend was used to test for a trend in detection rates across increasing BMI categories. Crude and adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for antenatal detection of congenital anomaly were estimated via maximum likelihood logistic regression. Models representing the adjusted odds of antenatal detection included maternal age, IMD, diabetes, multiple pregnancy and BMI. Gestational age at the first antenatal visit and gestational age at the final antenatal diagnosis were compared across BMI groups using Kruskal–Wallis tests. As approximately 30% of BMI data were missing, these cases were excluded from the regression analysis. Cases with missing data for one or more of the covariates were listwise excluded from the regression analysis. A test of proportions was performed to assess whether there was a difference in detection rates between women with missing and recorded BMI.

Interactions between maternal BMI and diabetes and maternal BMI and multiple pregnancies were investigated through the addition of cross-product terms. Interactions between maternal BMI and other statistically significant variables were also examined. Using termination of pregnancy for fetal anomaly as a binary outcome variable (categorised as Yes/No), a test for trend across BMI categories was performed and aORs were calculated, with adjustment for maternal age, IMD, diabetes, multiple pregnancy and BMI. Statistical analysis was performed using Stata 12 (StataCorp, College Station, TX, USA) and P < 0.05 was considered to be statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

There were 3096 cases of congenital anomaly confirmed postnatally among 132 885 pregnancies during the four-year study period, giving a total prevalence of 23.3 (95% CI: 22.5–24.1) per 1000 live and stillbirths. There were 597 (19.3%) cases associated with chromosomal anomalies and 14 (0.5%) associated with a teratogenic syndrome excluded from further analysis. Two cases that were not antenatally scanned and two with missing information on whether or not they were antenatally scanned were also excluded.

Of the remaining 2483 cases, 40 (1.6%) occurred in women with pre-gestational diabetes, 128 (5.2%) in twin pregnancies and three in separate triplet pregnancies. Other summary statistics are shown in Table 1.

Table 1.   Demographic statistics by antenatally suspected and unsuspected congenital anomalies*
VariableAll cases, n (%)Undetected cases, n (%)Detected cases, n (%) P
  1. *Chromosomal and teratogenic syndromes and women who did not have a scan were excluded. Missing categories were not included in any of the statistical tests.

  2. **Median (interquartile range) and Mann–Whitney U-test.

  3. ***χ2 test of association (Fisher’s exact test where column frequency <5).

  4. ****Cuzick’s test for trend.

  5. *****Gestational age at delivery: 1% missing. Gestational age at booking: 45.7% missing. Gestation at diagnosis of anomaly: 54.3% missing.

BMI (kg/m2)
Underweight (<18.5)67 (2.7)27 (40.3)40 (59.7)0.007μ****
Recommended weight (18.5–24.9)793 (31.9)376 (47.4)417 (52.6)
Overweight (25–29.9)468 (18.9)243 (51.9)225 (48.1)
Obese (≥30)358 (14.4)194 (54.2)164 (45.8)
Missing797 (32.1)497 (61.4)300 (37.6)
Maternal age at delivery (years)
<20267 (10.8)125 (46.8)142 (53.2)<0.001****
20–341820 (73.3)956 (52.5)864 (47.5)
≥35383 (15.4)244 (63.7)139 (36.3)
Missing13 (0.5)12 (92.3)1 (7.69)
Index of Multiple Deprivation
Least deprived773 (31.1)435 (56.3)338 (43.7)0.078****
Moderate844 (34.0)461 (54.6)383 (45.34)
Most deprived863 (34.8)439 (50.9)424 (49.1)
Missing1 (0.1)2 (66.7)1 (33.3)
Pre-gestational diabetes***
Yes40 (1.6)20 (50.0)20 (50.0)0.623
No2443 (98.4)1317 (53.9)1126 (46.1)
Multiple pregnancy***
Yes131 (5.3)53 (40.5)78 (59.5)0.002
No2352 (94.7)1284 (54.6)1068 (45.3)
Birth outcomes***
Late miscarriage30 (1.2)14 (46.7)16 (53.3)<0.001
Termination of pregnancy361 (14.5)0360 (100.0)
Antepartum stillbirth35 (1.4)9 (25.7)26 (74.3)
Intrapartum stillbirth2 (0.1)0 (0)2 (100.0)
Early neonatal death47 (1.9)11 (23.4)36 (76.6)
Late neonatal death26 (1.1)14 (53.9)12 (46.2)
Post-neonatal death38 (1.5)19 (50.0)19 (50.0)
Live birth1940 (78.1)1270 (65.5)670 (34.5)
Missing4 (0.1)04 (100.0)
Sex***
Male1304705 (54.1)599 (45.9)0.470
Female1131628 (55.5)503 (44.5)
Unknown484 (8.3)44 (91.7)
Year of delivery
2006648 (26.1)358 (55.3)290 (44.8)0.718****
2007620 (25.0)337 (54.4)283 (45.7)
2008650 (26.2)350 (53.5)301 (46.5)
2009565 (22.8)297 (52.0)271 (48.0)
Gestational age at delivery** , ***** (weeks) 38 (35–40)39 (37–40)36 (21–39)<0.001
Gestation at booking** , ***** (weeks) 9 (8–12)9 (7–12)10 (8–12)0.201
Gestation at diagnosis of anomaly**,***** (weeks) 20 (18–22)N/A20 (18–22)  N/A

Cardiovascular anomalies were the most common congenital anomaly group notified to NorCAS (945 cases; 38.1%; Table S1), but were the least commonly suspected antenatally (11.5%; Table S1). Urinary anomalies, nervous system anomalies, orofacial clefts and digestive system anomalies were specifically suspected antenatally in 88.4, 84.7, 44.4 and 35.1% of cases, respectively (Table S1).

Excluding women with missing maternal BMI, 67 (4.0%) anomalies occurred in women who were underweight, 793 (47.0%) in women who were of recommended BMI, 468 (27.8%) in women who were overweight and 358 (21.2%) in women who were obese (Table 1). An anomaly of any system was detected antenatally in 40 (59.7%), 417 (52.6%), 225 (48.1%) and 164 (45.8%) cases in women who were underweight, of recommended BMI, overweight and obese, respectively (Figure 1). Detection rates decreased significantly with increasing BMI category (test for trend: P = 0.007). Cases in women with missing BMI were significantly less likely to have been detected antenatally (300/797 = 37.6%) than those in women with a recorded BMI (846/1686 = 50.2%; test of proportions: P < 0.001). There was no evidence of a difference in the distribution of gestational age at final antenatal detection of anomaly or gestational age at first antenatal visit across BMI categories (Kruskal–Wallis test: P = 0.688 and P = 0.430, respectively).

image

Figure 1.  Trends in the percentage of suspected cases and the percentage of cases ending in termination of pregnancy across body mass index (BMI) categories (chromosomal and teratogenic syndromes were excluded). P values calculated using Cuzick’s test for trend.

Download figure to PowerPoint

The odds of detection of an anomaly (in any system) were significantly lower in obese women than in women of recommended BMI (aOR = 0.77; 95% CI: 0.60–0.99; P = 0.046; Table 2). There were no significant differences in the odds of detection in underweight (P = 0.414) or overweight (P = 0.157) women compared with women of recommended BMI. Increasing maternal age was significantly associated with decreasing odds of antenatal detection (aOR = 0.97; 95% CI: 0.96–0.99; P < 0.001). The odds of detection were also increased in women with multiple pregnancies (aOR = 1.55; 95% CI: 1.06–2.26; P = 0.024). There were no significant associations with IMD (P = 0.889 and P = 0.698 for least and most deprived, respectively) or pre-gestational diabetes (P = 0.766).

Table 2.   Association between maternal body mass index (BMI) and the odds of antenatal detection of congenital anomaly (overall and in most common anomaly groups*), as estimated by logistic regression
Postnatal diagnosisBMI category**Total, n (%)Detected antenatally***, n (% in BMI category)UnadjustedAdjusted****
OR (95% CI) P aOR (95% CI) P
  1. CI, confidence interval; OR, odds ratio; –, insufficient data to produce an OR.

  2. *Groups are categorised according to the European Surveillance of Congenital Anomalies (EUROCAT) guidelines.14

  3. **The missing categories were not included in the regression analysis.

  4. ***Ultrasound detection of any EUROCAT congenital anomaly within the same anomaly group as the final diagnosis. For the summary analysis of ‘any congenital anomaly’, the number is for any antenatally detected anomaly.

  5. ****Adjusted for Index of Multiple Deprivation (IMD) tertiles, maternal age at delivery, pre-gestational diabetes, multiple pregnancy.

Any congenital anomalyUnderweight67 (2.7)40 (59.7)1.34 (0.80–2.22)0.2641.24 (0.74–2.06)0.414
Recommended793 (31.9)417 (52.6)1 (Reference)1 (Reference) 
Overweight468 (18.9)225 (48.1)0.83 (0.66–1.05)0.1220.85 (0.67–1.07)0.157
Obese358 (14.4)164 (45.8)0.76 (0.59–0.98)0.0340.77 (0.60–0.99)0.046
Missing797 (32.1)300 (37.6)    
Cardiovascular anomalyUnderweight25 (2.7)7 (28.0)2.56 (1.00–6.56)0.0502.95 (1.13–7.70)0.027
Recommended273 (28.9)36 (13.2)1 (Reference)1 (Reference) 
Overweight177 (18.7)27 (15.3)1.19 (0.69. 2.03)0.5371.18 (0.68–2.04)0.557
Obese144 (15.2)18 (12.5)0.94 (0.51–1.72)0.8430.84 (0.45–1.56)0.575
Missing326 (34.5)21 (6.4)    
Urinary system anomalyUnderweight7 (2.3)7 (100.0) 
Recommended112 (37.1)103 (92.0)1 (Reference)1 (Reference) 
Overweight61 (20.2)52 (85.3)0.57 (0.22–1.48)0.2460.58 (0.22–1.56)0.284
Obese37 (12.3)34 (91.9)0.81 (0.24–2.75)0.7340.84 (0.24–2.96)0.815
Missing84 (28.2)73 (85.9)    
Nervous system anomalyUnderweight8 (3.1)8 (100.0) 
Recommended93 (35.5)85 (91.4)1 (Reference)1 (Reference) 
Overweight51 (19.5)45 (88.2)1.01 (0.35–2.90)0.9911.06 (0.36–3.14)0.914
Obese41 (15.7)38 (92.7)1.24 (0.37–4.15)0.7261.47 (0.43–5.07)0.540
Missing69 (26.3)51 (73.9)    
Orofacial cleftUnderweight3 (2.1)0 (0.0) 
Recommended58 (40.9)30 (51.7)1 (Reference)1 (Reference) 
Overweight32 (22.5)15 (46.9)0.88 (0.37–2.09)0.7770.92 (0.38–2.20)0.846
Obese17 (12.0)9 (52.9)0.89 (0.30–2.62)0.8310.68 (0.21–2.16)0.508
Missing32 (22.5)11 (34.4)    
Digestive system anomalyUnderweight5 (4.5)3 (60.0)1.62 (0.23–11.26)0.6281.95 (0.26–14.86)0.519
Recommended27 (28.8)13 (48.2)1 (Reference)1 (Reference) 
Overweight13 (11.7)4 (30.8)0.48 (0.12–1.94)0.3020.59 (0.13–2.74)0.503
Obese17 (15.3)7 (41.2)0.75 (0.22–2.57)0.6510.67 (0.17–2.61)0.564
Missing49 (44.1)13 (26.5)    

There was no evidence that the influence of BMI on the odds of antenatal detection was significantly different in multiple relative to singleton pregnancies or in women with pre-gestational diabetes relative to those without. One significant interaction was observed between overweight BMI and maternal age (P = 0.017). In women aged 35 years or more, the odds of an anomaly being detected antenatally was significantly lower in overweight women (aOR = 0.46; 95% CI: 0.25–0.84; P = 0.012) than in women of recommended BMI. No such effect was observed in women under the age of 35 years (overweight versus recommended BMI: aOR = 0.95; 95% CI: 0.74–1.22; P = 0.663).

Of the cases that were confirmed postnatally as having isolated cardiovascular anomalies, there was an approximate three-fold increased odds of detecting a cardiovascular anomaly in underweight women than in women of recommended BMI (aOR = 2.95; 95% CI: 1.13–7.70; P = 0.027; Table 2). There were no significant differences in the antenatal detection of cardiovascular anomalies in overweight and obese women compared with women of recommended BMI (Table 2).

There were no significant associations between BMI (of any category) and antenatal detection of congenital anomalies in any other organ system (the most commonly affected organ systems are shown in Table 2).

Overall, 361 (14.5%) of all cases ended in termination of pregnancy, with seven (10.5%), 134 (17.0%), 84 (18.0%) and 56 (15.7%) cases occurring in underweight, recommended BMI, overweight and obese women, respectively (Figure 1). There was no trend in termination of pregnancy rates over increasing BMI categories (test for trend: P = 0.835). Of those cases that were suspected antenatally, 17.5, 32.4, 37.3 and 34.4% ended in a termination of pregnancy in underweight, recommended BMI, overweight and obese women, respectively (test for trend: P = 0.112). After adjusting for maternal age, IMD, diabetes and multiple pregnancies, there was no significant association between termination and BMI (logistic regression: P = 0.075, P = 0.182 and P = 0.431 in underweight, overweight and obese women, respectively). There was no association with IMD (P = 0.335 and P = 0.081 for least and most deprived, respectively) or diabetes (P = 0.408), but there was some evidence that the odds of termination decreased with increasing maternal age (aOR = 0.98; 95% CI: 0.96–1.00; P = 0.051) and in multiple relative to singleton pregnancies (aOR = 0.43; 95% CI: 0.23–0.80; P = 0.008).

Termination of pregnancy occurred in 28 (3.0%) confirmed cases of cardiovascular anomaly. There was no association between BMI categories and termination of pregnancy among cardiovascular cases (P = 0.876 and P = 0.201 for overweight and obese women, respectively).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

This study found a significant decreasing trend in the antenatal ultrasound detection of congenital anomalies across all BMI categories. In addition, there was a significantly increased odds of detecting an anomaly of the cardiovascular system in underweight women relative to women of recommended BMI. There was no significant association between BMI and antenatal detection of congenital anomalies in any other organ system, or between BMI and termination of pregnancy for fetal anomaly.

This study is the largest to examine the effect of maternal BMI on the antenatal detection of congenital anomalies. The primary strength of the study is that information on cases was extracted from a high-quality, population-based congenital anomaly register. Consistent methods of identifying and notifying cases are used to ensure a high case ascertainment. In addition, NorCAS is an active register that follows cases to age 12 years to maximise ascertainment of anomalies, such as those of the cardiovascular system, which may not be diagnosed until well into childhood. NorCAS records data on up to six congenital anomalies per case; thus, we were able to classify each case as isolated, associated or chromosomal, and to examine antenatal detection rates by organ system, which has not been performed previously.

NorCAS is held within the same database as two other high-quality population-based registers (NorDIP and NorSTAMP). As records are linked as cases are notified, we were able to accurately identify cases occurring in women with pre-gestational diabetes or multiple pregnancies. Therefore, we could adjust for these potentially confounding factors, which are both associated with an increased risk of congenital anomaly18,19 and increased antenatal surveillance.20,21

The study has some limitations. The maternal BMI data recorded at the first antenatal visit were mostly derived from self-reported height and weight. It has been suggested that around 20% of women underreport their weight at first attendance for antenatal care, leading to an underestimation of their BMI.2 Moreover, almost one-third of the BMI data were missing. Despite our best efforts to obtain these data, they were not recorded in the study years. Cases in women with missing BMI data were less likely to have been detected antenatally than those in women who had a BMI recorded. If overweight or obese women account for a higher proportion of the missing data, the strength of our association may be an underestimate.

Although we were able to account for pregnancies associated with pre-gestational diabetes, we could not account for those associated with gestational diabetes. Mothers with gestational diabetes are more likely to be overweight or obese22 and, because of the higher risks associated with their pregnancy, may receive additional antenatal scans. Thus, we may have underreported the size of the true association between antenatal detection and obesity.

The second-trimester ultrasound may have been too early to detect certain nervous system anomalies, such as posterior fossa or agenesis of the corpus callosum, which are not usually detectable until a later gestational age. However, a large proportion of nervous system anomalies were suspected antenatally (84.7%), and so this would not have had a major impact on the results.

As the purpose of the study was to investigate the sensitivity of routine antenatal ultrasound scanning to detect congenital anomalies, we excluded chromosomal anomalies. We could not distinguish between those anomalies identified via ultrasound examination and those detected by other means (e.g. genetic testing). Therefore, this study cannot describe the association between BMI and the odds of antenatal detection in this group of anomalies.

In addition, we could not adjust the logistic regression models for gestational age at scanning. Although gestational age at final antenatal diagnosis is recorded, this may correspond to a scan subsequent to that which caused the initial suspicion. Furthermore, this variable has a high proportion of missing values. Evidence suggests that scans occurring at later gestational ages may lead to better visualisation of the fetus,7,13 although not all studies support this finding.5,8 Nevertheless, we examined gestational age at final antenatal diagnosis, and at first antenatal visit, and found no significant difference across BMI categories, which suggests that the association between BMI and the odds of detection is not related to differences in gestational age at the time of the scan.

Considering all congenital anomalies, we found similar antenatal detection rates to those described by EUROCAT in the UK.4 However, we identified significantly lower detection rates of cardiovascular anomalies than did Boyd et al.23 in 2005–2006 in England and Wales. However, the study by Boyd et al.23 only included serious cardiac anomalies (defined as common arterial trunk, discordant ventriculoarterial connection, transposition of the great vessels, tetralogy of Fallot, Ebstein’s anomaly or coarctation of the aorta) that were amenable to detection, whereas we investigated all EUROCAT-defined major anomalies of the cardiovascular system.14 Garne et al.24 reported cardiovascular detection rates of 25% in Europe and 35% in England, which are slightly more consistent with ours, but this was an older study and detection rates may have since improved.

Apart from the cardiovascular system, there were a limited number of anomalies from other individual organ systems. As a result, this study may have been underpowered to detect associations between maternal BMI and the antenatal detection of anomalies within other organ systems. Nonsignificant associations should therefore not be interpreted as evidence of no effect. Further studies are needed with larger sample sizes.

Few studies have investigated the association between maternal BMI and the detection of congenital anomalies, but those that have, show consistent findings to those reported here. In low-risk pregnancies, Dashe et al.6 identified a negative trend in the rates of detection as BMI increased, with rates of 66 and 48% in normal (BMI < 25 kg/m2) and class 1 obese (BMI = 30–34.9 kg/m2) women, respectively. These detection rates are higher than ours, possibly because the authors excluded cases associated with atrial septal defect, a cardiovascular anomaly which is difficult to detect antenatally [e.g. in our study, 2.9% (1/35) of isolated atrial septal defects were suspected antenatally]. In addition, Dashe et al.6 only recorded cases if they were diagnosed before hospital discharge or if they occurred in neonatal deaths, whereas NorCAS can receive notification of a case up to age 12 years. Similarly, Tabor et al.5 identified a difference in detection rates between women with BMI ≤ 25 kg/m2 and BMI > 25 kg/m2 (76.4 and 53.3%, respectively). These detection rates may also appear to be higher than in our study because of a more modest BMI categorisation and because the length of follow-up was shorter than ours (median of 22 months). Neither of these studies performed regression analyses, and so we cannot compare the adjusted odds of detection.5,6 A number of other studies showing greater suboptimal visualisation for increasing BMI7–10,12,25 also complement our findings.

In our study, obese BMI was not significantly associated with antenatal detection of cardiovascular anomalies. This contrasts with the studies by Hendler et al.7 and Khoury et al.,12 who both identified greater suboptimal visualisation of the fetal heart in obese women than in women of recommended BMI. This discrepancy may have occurred because of low study power. Although we found an overall effect in obese women, and cardiovascular anomalies were the largest congenital anomaly group, they also had the lowest rate of detection (11.5%), and therefore very few antenatally detected cases (n = 88). Hence, although our study had sufficient power to detect a large effect size for cardiovascular anomalies (as observed with underweight women), it may not have had sufficient power for a moderate effect, which potentially exists for obese women. To our knowledge, this is the first study to investigate the effect of maternal underweight on the ultrasound detection of congenital anomalies. We found an increased rate of antenatal detection in underweight women, which suggests that ultrasound visibility decreases continuously over all BMI categories.

Although obese women were less likely to have an anomaly suspected antenatally, they were not significantly less likely to have a termination of pregnancy, compared with women of recommended BMI. Termination of pregnancy was a relatively rare outcome and the effect of obesity on antenatal detection was moderate, and so it is possible that we did not have the power to identify a small effect. However, a recent systematic review found no evidence to support this, as the association between congenital anomaly and maternal obesity was similar irrespective of whether terminations of pregnancy for fetal anomaly were included.3 Thus, antenatal scanning for congenital anomalies is less effective as BMI increases, resulting in fewer cases detected antenatally, but this study provides no evidence that this has an impact on termination of pregnancy for fetal anomaly.

We found that cases occurring in a multiple pregnancy were more likely to be detected than cases occurring in a singleton pregnancy. Congenital anomalies may be more frequently suspected in multiple pregnancies because more scans are offered to these women or, potentially, more experienced sonographers might scan these women.21 Further research is required to investigate this association. There was no evidence that the effect of BMI on antenatal detection was different among multiple relative to singleton pregnancies, and so it was feasible to incorporate and adjust for multiple pregnancies in our analysis.

As the amount of abdominal adipose tissue increases, the depth travelled by the ultrasound waves during the mid-trimester scan becomes greater.25 Therefore, more waves are absorbed into the surrounding tissue, which causes the signal to weaken and, as a result, visualisation of the fetus and therefore any congenital anomalies diminishes.25 If BMI is considered as a marker for abdominal adipose tissue, this would explain why detection decreases over all BMI categories, and is not impaired in obese women alone. Furthermore, this might explain why we identified an interaction between overweight BMI and maternal age, if the older overweight mothers had more abdominal adipose tissue than their younger counterparts with the same BMI, as has been reported previously.26,27

The National Institute for Health and Clinical Excellence (NICE) suggests that pregnant women should be informed that antenatal detection rates may vary by maternal BMI.28 Recommendations should be directed to improving ultrasound sensitivity, for example by advising enhanced scanning for overweight and obese women. Paladini25 has described several methods to boost ultrasound image quality and therefore enhance visualisation of the fetal heart, which should be further evaluated in women of increased BMI.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Congenital anomalies, particularly within the cardiovascular system, are difficult to detect antenatally. Our study shows that this is further challenged as maternal BMI increases. However, we found that this had no measurable impact on the proportion of pregnancies resulting in a termination of pregnancy. Women should be informed of the limitations of ultrasound for the detection of congenital anomalies. To maximise visualisation of the fetal heart, recommendations to improve ultrasound scanning sensitivity in women with increased BMI should be evaluated further.

Disclosure of interests

All authors report no conflict of interest.

Contribution to authorship

KEB undertook the analysis and interpretation of the data and drafted the manuscript. JR conceived the project and, with RB and PWGT, participated in the interpretation of the data and the critical review of the manuscript. All authors read and approved the final version of the manuscript before submission.

Details of ethics approval

As part of the British Isles Network of Congenital Anomaly Registers (BINOCAR), NorCAS has exemption from the National Information Governance Board for Health and Social Care from a requirement for consent for inclusion on the register and has ethical approval (09/H0405/48) to undertake studies involving the use of the data. This study was given a favourable ethical opinion from the Northumberland Research Ethics Committee (07/Q0902/2) and Research and Development approval from each of the participating hospitals.

Funding

The NorCAS register and KEB are funded by the Healthcare Quality Improvement Partnership. This project was funded by Newlife foundation for disabled children.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

We are grateful to all the Link Clinicians in the north of England for their continued collaboration and support of NorCAS.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information
  • 1
    Heslehurst N, Rankin J, Wilkinson JR, Summerbell CD. A nationally representative study of maternal obesity in England, UK: trends in incidence and demographic inequalities in 619 323 births, 1989–2007. Int J Obes 2009;34:4208.
  • 2
    Rankin J, Tennant PW, Stothard KJ, Bythell M, Summerbell CD, Bell Rl. Maternal body mass index and congenital anomaly risk: a cohort study. Int J Obes 2010;34:137180.
  • 3
    Stothard KJ, Tennant PWG, Bell R, Rankin J. Maternal overweight and obesity and the risk of congenital anomalies. J Am Med Assoc 2009;301:636.
  • 4
    European Surveillance of Congenital Anomalies (EUROCAT). [http://www.eurocat-network.eu/prenatalscreeninganddiagnosis/prenataldetectionrates]. Accessed 6 January 2011.
  • 5
    Tabor A, Zdravkovic M, Perslev A, Møller LK, Pedersen BL. Screening for congenital malformations by ultrasonography in the general population of pregnant women: factors affecting the efficacy. Acta Obstet Gynecol Scand 2003;82:10928.
  • 6
    Dashe JS, McIntire DD, Twickler DM. Effect of maternal obesity on the ultrasound detection of anomalous fetuses. Obstet Gynecol 2009;113:10017.
  • 7
    Hendler I, Blackwell SC, Bujold E, Treadwell MC, Wolfe HM, Sokol RJ, et al. The impact of maternal obesity on midtrimester sonographic visualization of fetal cardiac and craniospinal structures. Int J Obes 2004;28:160711.
  • 8
    Wolfe HM, Sokol RJ, Martier SM, Zador IE. Maternal obesity: a potential source of error in sonographic prenatal diagnosis. Obstet Gynecol 1990;76:339.
  • 9
    Phatak M, Ramsay J. Impact of maternal obesity on procedure of mid-trimester anomaly scan. J Obstet Gynaecol 2010;30:44750.
  • 10
    Maxwell C, Dunn E, Tomlinson G, Glanc P. How does maternal obesity affect the routine fetal anatomic ultrasound? J Matern Fetal Neonatal Med 2010;23:118792.
  • 11
    Dashe JS, McIntire DD, Twickler DM. Maternal obesity limits the ultrasound evaluation of fetal anatomy. J Ultrasound Med 2009;28:102530.
  • 12
    Khoury FR, Ehrenberg HM, Mercer BM. The impact of maternal obesity on satisfactory detailed anatomic ultrasound image acquisition. J Matern Fetal Neonatal Med 2009;22:33741.
  • 13
    Hendler I, Blackwell SC, Bujold E, Treadwell MC, Mittal P, Sokol RJ, et al. Suboptimal second-trimester ultrasonographic visualization of the fetal heart in obese women: should we repeat the examination? J Ultrasound Med 2005;24:12059.
  • 14
    European Surveillance of Congenital Anomalies (EUROCAT). EUROCAT Guide 1.3 and Reference Documents, 2005 [http://www.eurocat-network.eu/content/EUROCAT-Guide-1.3.pdf]. Accessed 9 March 2012.
  • 15
    Bell R, Bailey K, Cresswell T, Hawthorne G, Critchley J, Lewis-Barned N. Trends in prevalence and outcomes of pregnancy in women with pre-existing type I and type II diabetes. BJOG 2008;115:44552.
  • 16
    Glinianaia SV, Rankin J, Wright C, Sturgiss SN, Renwick M. A multiple pregnancy register in the north of England. Twin Res Hum Genet 2002;5:4369.
  • 17
    Noble MWG, Wright G, Dibben C, Smith GAN, McLennan D, Antitila C, et al. The English Indices of Deprivation 2004 (Revised). London: Neighbourhood Renewal Unit, Office of the Deputy Prime Minister, 2004.
  • 18
    Bell R, Glinianaia S, Tennant P, Bilous R, Rankin J. Peri-conception hyperglycaemia and nephropathy are associated with risk of congenital anomaly in women with pre-existing diabetes: a population-based cohort study. Diabetologia 2011;55:93647.
  • 19
    Glinianaia SV, Rankin J, Wright C. Congenital anomalies in twins: a register-based study. Hum Reprod 2008;23:1306.
  • 20
    National Institute for Health and Clinical Excellence (NICE). Diabetes In Pregnancy: Management of Diabetes and its Complications from Pre-Conception to the Postnatal Period. London: NICE, 2008 [cited 7 March 2012].
  • 21
    National Institute for Health and Clinical Excellence (NICE). The Management of Twin and Triplet Pregnancies in the Antenatal Period. London: NICE, 2011 [cited 7 March 2012].
  • 22
    Chu SY, Callaghan WM, Kim SY, Schmid CH, Lau J, England LJ, et al. Maternal obesity and risk of gestational diabetes mellitus. Diabetes Care 2007;30:2070.
  • 23
    Boyd PA, Tonks AM, Rankin J, Rounding C, Wellesley D, Draper ES. Monitoring the prenatal detection of structural fetal congenital anomalies in England and Wales: register-based study. J Med Screen 2011;18:27.
  • 24
    Garne E, Stoll C, Clementi M, The Euroscangroup. Evaluation of prenatal diagnosis of congenital heart diseases by ultrasound: experience from 20 European registries. Ultrasound Obstet Gynecol 2001;17:38691.
  • 25
    Paladini D. Sonography in obese and overweight pregnant women: clinical, medicolegal and technical issues. Ultrasound Obstet Gynecol 2009;33:7209.
  • 26
    Kotani K, Tokunaga K, Fujioka S, Kobatake T, Keno Y, Yoshida S, et al. Sexual dimorphism of age-related changes in whole-body fat distribution in the obese. Int J Obes 1994;18:207.
  • 27
    Enzi G, Gasparo M, Biondetti PR, Fiore D, Semisa M, Zurlo F. Subcutaneous and visceral fat distribution according to sex, age, and overweight, evaluated by computed tomography. Am J Clin Nutr 1986;44:73946.
  • 28
    National Institute for Health and Clinical Excellence (NICE). Antenatal Care Guidelines. London: NICE, 2008. [http://www.nice.org.uk/nicemedia/pdf/CG062NICEguideline.pdf]. Accessed 28 February 2012.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  10. Supporting Information

Table S1. Number and percentage of cases by congenital anomaly group.

FilenameFormatSizeDescription
BJO_3462_sm_TableS1.pdf43KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.