In the evaluation that ensues after a stillbirth, congenital anomalies are one of the most commonly identifiable causes.1 However, with the routine use of ultrasound scanning, the diagnosis of a major anomaly often precedes the loss.2,3 There are minimal data with which to counsel patients regarding the ongoing rate of stillbirth among anomalous fetuses after ultrasound diagnosis, especially if the anomaly is isolated and not associated with a genetic syndrome.
Unlike other risk factors for stillbirth,4–6 guidelines for the antenatal management of pregnancies complicated by isolated fetal anomalies are limited. In addition to the risk of stillbirth, fetuses with congenital anomalies are at risk for growth restriction,7–9 and frequently pregnancy management is based on this subsequent diagnosis rather than the anomaly itself. Although fetal growth restriction is a known independent risk factor for stillbirth,10,11 the interaction between growth restriction and fetal anomalies and its effect on stillbirth is largely undefined.
In this study, we sought to estimate the risk of stillbirth in fetuses with isolated congenital anomalies diagnosed during routine prenatal ultrasound evaluation and examine the influence of the incidental finding of growth restriction on the stillbirth risk using a large ultrasound database at a single institution.
MATERIALS AND METHODS
We performed a retrospective cohort study of all consecutive singleton pregnancies presenting for routine anatomic ultrasound examination at Washington University between 1990 and 2009. The study was conducted using an institutional perinatal database that includes ultrasonographic findings as well as demographic information, maternal medical history, pregnancy, and neonatal outcomes.12 Approval for the study was granted by the Washington University School of Medicine human studies review board.
Pregnancies complicated by an isolated major fetal anomaly diagnosed prenatally were compared with pregnancies in which a major fetal anomaly was absent. Major congenital anomalies were defined as structural abnormalities likely to result in significant functional impairment or need for medical or surgical intervention. Decisions regarding which anomalies were considered “major” were guided by criteria used in the European Surveillance of Congenital Anomalies (EUROCAT) network.13 Anomalies included in the study were classified by the organ system affected and are listed in Box 1. Pregnancies were excluded if the fetus had more than one major anomaly or a chromosomal abnormality. Absence of other structural abnormalities was based on prenatal ultrasound findings only, whereas chromosomal abnormalities may have been diagnosed by either prenatal or postnatal genetic testing. Additionally, pregnancies complicated by minor anomalies, which included any structural abnormality not listed in Box 1, were excluded. Examples of minor anomalies that were excluded include minor markers for aneuploidy, polydactyly, and mild pyelectasis. Pregnancies resulting in delivery before 24 weeks of gestation were also not included in this analysis, because documentation regarding elective termination of pregnancy was not well captured within the database and local regulations do not permit elective termination after this gestational age.
Box 1 Major Structural Anomalies Included in the Study, by Organ System Cited Here...
Coarctation of the aorta
Tetralogy of Fallot
Transposition of the great vessels
Double-outlet or double-inlet ventricle
Hypoplastic left or right heart
Thorcic or respiratory (n=63)
Congenital pulmonary adenomatoid malformation
Anorectal atresia or imperforate anus
Large bowel obstruction
Small bowel obstruction
Congenital diaphragmatic hernia
Bladder outlet obstruction or urethral atresia or stenosis
Cloacal persistence or cloacal or bladder extrophy
Posterior urethral valves
Characteristics of pregnancies complicated by a major congenital anomaly and nonanomalous pregnancies were compared. Data including maternal medical and obstetric history, age, parity, race, and body mass index (calculated as weight (kg)/[height (m)]2) were recorded at the time of routine anatomic ultrasound scan and stored in the perinatal database. Pregnancy outcome data included in the database such as gestational age at delivery, neonate birth weight, and diagnosis of complications such as gestational diabetes or preeclampsia were collected by a dedicated pregnancy outcome coordinator in an ongoing manner after delivery from the medical record for women delivering within our hospital system or with use of a questionnaire administered to women who delivered elsewhere. If the questionnaire was not returned, the patient or referring health care provider was contacted by telephone. Pregnancies were considered complicated by growth restriction if the birth weight was less than the 10th percentile using the Alexander chart.14 Statistical comparisons were performed using the χ2 test for categorical variables. The Mann–Whitney U test was used to compare gestational age at delivery and birth weight because these continuous variables were not normally distributed.
The stillbirth rate per 1,000 ongoing pregnancies beyond 23 6/7 weeks of gestation was calculated for pregnancies complicated by isolated major congenital anomaly and those pregnancies without major anomalies. To compare the stillbirth rates in anomalous and nonanomalous pregnancies, we calculated the relative risk of stillbirth with the 95% confidence interval. To determine whether stillbirths occurred early or late in gestation, we performed a stratified analysis based on gestational age at delivery before 32 weeks and 32 weeks or after. Stillbirth rates were calculated per ongoing pregnancies; thus, the denominator in the before 32 weeks of gestation stillbirth analysis included all women in the study, whereas the denominator in the 32 weeks of gestation or after strata only included women who were still pregnant at 32 0/7 weeks of gestation. The effect of incidental growth restriction was also investigated using stratified analysis. Multivariable logistic regression was used to adjust for relevant confounders. All characteristics associated with isolated major congenital anomaly in univariable analysis were included in the initial model. A backward, stepwise approach using the likelihood ratio test to assess the effect of the removal of covariates was used to create the final model, which included black race, maternal obesity (body mass index greater than 30), and pregestational diabetes. Additionally, we performed a sensitivity analysis excluding universally lethal anomalies including anencephaly or acrania and bilateral renal agenesis. We then calculated the rate of stillbirth per 1,000 ongoing pregnancies in each of the six organ system categories and compared these rates with the stillbirth rate in the nonanomalous control group by calculating relative risks and 95% confidence intervals (CIs). All statistical analyses were performed using STATA 10.0 special edition.
Within the perinatal ultrasound database, 76,453 singleton pregnancies were identified. After excluding pregnancies complicated by chromosomal abnormalities, minor anomalies, or multiple major anomalies in the same fetus, 74,424 pregnancies remained. Delivery before 24 weeks of gestation occurred in 1,429 pregnancies (1.9%), of which 333 were among pregnancies complicated by an isolated major congenital anomaly and 1,096 were in nonanomalous pregnancies. In addition, 7,957 pregnancies were lost to follow-up (10.7%); 33 pregnancies were in the anomalous group and 7,924 in the nonanomalous group. The final cohort included 65,308 pregnancies, which was comprised of 873 pregnancies with an isolated major congenital anomaly (1.3%) and 64,165 nonanomalous pregnancies (Fig. 1).
Pregnancies complicated by an isolated major congenital anomaly were more likely to occur in women who were white, nulliparous, and of advanced maternal age. Maternal obesity, gestational diabetes, and chronic hypertension were more common in nonanomalous pregnancies. Median gestational age at delivery was earlier in pregnancies with an isolated anomaly. Overall median birth weight was lower in pregnancies with an isolated congenital anomaly; additionally, 24.4% of anomalous fetuses were also growth-restricted at birth, whereas only 11.5% of nonanomalous pregnancies were complicated by growth restriction (Table 1). The proportion of isolated congenital anomalies detected by ultrasonography was similar from 1990–1999 and 2000–2009 (1.29% compared with 1.39%, P=.27).
Fetuses with an isolated congenital anomaly had a 15-fold increased risk of stillbirth after adjusting for maternal obesity, pregestational diabetes, and black race. The stillbirth rate was highest (127/1,000 pregnancies) among pregnancies complicated by both a congenital anomaly and growth restriction. However, because of the relatively high rate of stillbirth in nonanomalous growth-restricted pregnancies (18/1,000 pregnancies), the risk of stillbirth associated with a major congenital anomaly in growth-restricted pregnancies (adjusted odds ratio [OR] 8.20, 95% CI 5.27–12.74) is lower than risk associated with a major congenital anomaly in nongrowth-restricted pregnancies (adjusted OR 15.01, 95% CI 9.34–24.12). Among pregnancies complicated by an isolated anomaly, growth restriction was associated with a greater risk of stillbirth (adjusted OR 4.88, 95% CI 2.65–8.98). In pregnancies complicated by an isolated major congenital anomaly as well as incidental growth restriction, the stillbirth rate was higher at 32 weeks of gestation or greater than before 32 weeks of gestation. Conversely, a higher rate of stillbirth was found before 32 weeks of gestation rather than 32 weeks of gestation or greater in anomalous pregnancies that were not growth-restricted (Table 2).
Twenty-eight pregnancies were complicated by an anomaly considered always lethal, including anencephaly, acrania, and bilateral renal agenesis. A sensitivity analysis excluding these anomalies from the isolated major congenital anomaly group did not significantly affect the results of the primary analysis. Isolated major anomaly remained significantly associated with an increased risk of stillbirth compared with nonanomalous pregnancies (47/1,000 pregnancies [n=40] compared with 4/1,000 pregnancies [n=254]; adjusted OR 12.95, 95% CI 9.18–18.23). The stillbirth rate was also higher in pregnancies complicated an isolated congenital anomaly compared with nonanomalous pregnancies whether the pregnancy was also complicated by growth restriction (111/1,000 [n=21] compared with 18/1,000 [n=133] pregnancies; adjusted OR 7.17, 95% CI 4.40–11.70) or not (29/1,000 [n=19] compared with 2/1,000 [n=121] pregnancies; adjusted OR 14.49, 95% CI 8.87–23.70). Furthermore, in anomalous pregnancies, growth restriction was associated with an increased risk of stillbirth (111/1,000 [n=21] compared with 29/1,000 [n=19] pregnancies; adjusted OR 4.42, 95% CI 2.29–8.52).
Pregnancies complicated by isolated major congenital anomalies in each of the organ system categories considered were at an increased risk of stillbirth relative to nonanomalous pregnancies. The highest stillbirth rate was found among fetuses with congenital heart disease (143/1,000 pregnancies) (Table 3).
We found that pregnancies complicated by isolated major congenital anomalies are associated with a 15-fold increased risk of stillbirth. Overall, one in every 18 pregnancies complicated by an isolated major anomaly will result in fetal death. Incidental growth restriction was associated with an even higher rate of stillbirth, occurring in approximately one in every eight pregnancies complicated by growth restriction and isolated congenital anomaly.
The results of this study can be used to counsel patients regarding the increased risk of stillbirth associated with isolated major congenital anomalies and develop antepartum management plans. Although growth restriction is a known risk factor for stillbirth,10,11 our data confirm that stillbirth rates are highest in fetuses that are both anomalous and growth-restricted. Furthermore, rates of stillbirth in nongrowth-restricted anomalous fetuses were higher than the stillbirth rate among nonanomalous, growth-restricted pregnancies. Increased fetal surveillance is often instituted for pregnancies complicated by a wide variety of conditions that are associated with increased stillbirth risk.4 However, fetal anomaly, with perhaps the exception of gastroschisis,15 is not considered an indication for testing unless the fetus is also growth-restricted. This management strategy may be misguided given the high risk of stillbirth in anomalous fetuses independent of growth restriction. Nevertheless, initiating antenatal surveillance in pregnancies complicated by an isolated fetal anomaly is a complex decision because the competing risk of neonatal demise increases with decreasing gestational age, particularly in anomalous fetuses.16–19 For specific anomalies, there may be a gestational age at which the risk of stillbirth exceeds the postnatal mortality risk and thus the initiation of antenatal surveillance with its incumbent false-positive rate20 warrants consideration. Unfortunately, we did not collect specific data about fetal surveillance in this study; thus, further research is needed to better define the time point in gestation when the stillbirth rate approximates the neonatal death rate for individual anomalies.
Our finding that there is an association between fetal abnormality and stillbirth is consistent with prior studies.1,13,21,22 However, our study design allowed us to explore the relationship from a different perspective, with the goal of obtaining information with which to counsel women and families who have received the diagnosis of an isolated major fetal anomaly at the time of routine anatomic ultrasound scan and who elect to continue the pregnancy and reach a gestational age at which most nonanomalous fetuses are considered viable. Most other studies that have examined the association between stillbirth and fetal anomalies have done so from the perspective of evaluating causes of stillbirth,1,21,22 which does not provide data regarding the ongoing risk of stillbirth in an anomalous fetus. The EUROCAT study, a large international registry in Europe that has been in existence for more than 30 years, has provided much of the available information regarding risks associated with fetal anomalies.23 However, multiple data sources are used for case ascertainment, which includes registries of infants who are diagnosed postnatally up to age 1 year. The stillbirth risk calculated using data that include postnatal diagnosis would be expected to be lower than the stillbirth risk associated with fetal anomalies that are detected by ultrasound examination prenatally. Although ultrasound examination detects between 40 and 64% of fetal structural abnormalities,2,3,24 those that are detected on ultrasonography are more likely to be severe3 and thus may be associated with a higher risk of intrauterine death.
Most other studies evaluating the association between stillbirth and anomalies have included fetuses with multiple anomalies.13 It is difficult to attribute the risk of stillbirth associated with a single structural abnormality if fetuses with multiple anomalies are included. Additionally, it is more likely that a fetus with multiple anomalies has a genetic syndrome, which itself might be associated with increased mortality.25,26 Our use of only prenatal ultrasound findings to define the absence of other structural malformations but both prenatal and postnatal genetic testing to exclude pregnancies complicated by chromosomal abnormalities may seem incongruent. However, this reflects the stillbirth risk using prenatally available information. Although ultrasonography may not detect all structural abnormalities, prenatal genetic testing is available and offered to all women. Ultimately, data from our study could be used to counsel women about risk of stillbirth if the fetus does not have a chromosomal abnormality and has only a single anomaly detected on ultrasound scan, although there may be additional abnormalities not detectable prenatally.
The American College of Obstetricians and Gynecologists defines stillbirth as fetal death at 20 weeks of gestation or greater or a fetal weight 350 g or greater if the gestational age is unknown.4 We chose to exclude women who delivered before 24 weeks of gestation based on local regulations regarding termination of pregnancy. We acknowledge that there is a selection bias introduced by this approach because we surmise that pregnancies that are terminated are more likely to have had a more severe congenital anomaly. Our approach, however, would likely bias the results toward the null because the more severe congenital anomalies may be associated with higher stillbirth risk.
Overall, both isolated congenital anomaly and stillbirth are rare events. The large size of our single-center ultrasound database gave us the ability to perform this analysis. However, there was less precision of the risk estimates in some of the subgroup analyses as a result of the small numbers. The ultrasound and patient follow-up information in the database is more detailed than is typically recorded in larger national and international registries.27 A limitation of this is that data collected at a single referral center could decrease the generalizability of our findings. Although there was follow-up available on 89.6% of women who underwent ultrasound evaluation at our center, some pregnancies were excluded because of incomplete data. Further investigation found that these women were more likely to be younger, black race, obese, and multiparous compared with women included in the study. It is unclear how the exclusion of these pregnancies would have affected our results. Additionally, chromosomal analysis was only performed in 73.9% of cases of isolated anomalies; thus, some cases of genetically abnormal fetuses could have been misclassified.
The study was conducted over an almost 20-year time period. Changes in ultrasound detection rates over this time period were likely minimal because a similar proportion of all pregnancies was found to be complicated by an isolated congenital anomaly. However, the availability and efficacy of postnatal care of fetuses with congenital anomalies over this time period may have affected our results. Some may argue that defining growth restriction using birth weight is another limitation because obstetric management is based on prenatal diagnosis of growth restriction. However, ultrasound assessment of fetal weight is largely inaccurate28; thus, the use of birth weight provides a more direct approach to examining the true relationship between growth restriction and stillbirth. Furthermore, our finding that the stillbirth risk is high in pregnancies complicated by an isolated congenital anomaly regardless of incidental growth restriction in anomalous fetuses means that reliance on prenatal assessment of fetal growth to guide management is unnecessary.
In summary, we found that pregnancies complicated by an isolated congenital fetal anomaly are at high risk of stillbirth regardless of the incidental diagnosis of growth restriction. Our data could be used to help obstetric care providers counsel patients receiving an antenatal diagnosis of an isolated anomaly. Although antenatal surveillance is frequently initiated in pregnancies at high risk for stillbirth, health care practitioners caring for these patients should weigh the competing risks of postnatal mortality with antenatal death. Critical evaluation of these competing risks, specific to individual anomalies, should be the focus of future studies.
1. Stillbirth Collaborative Research Network Writing Group. Causes of death among stillbirths. JAMA 2011;306:2459–68.
2. Romosan G, Henriksson E, Rylander A, Valentin L. Diagnostic performance of routine ultrasound screening for fetal abnormalities in an unselected Swedish population in 2000–2005. Ultrasound Obstet Gynecol 2009;34:526–33.
3. Grandjean H, Larroque D, Levi S. The performance of routine ultrasonographic screening of pregnancies in the Eurofetus Study. Am J Obstet Gynecol 1999;181:446–54.
4. Management of stillbirth. ACOG Practice Bulletin No. 102. American College of Obstetricians and Gynecologists. Obstet Gynecol 2009;113:748–61.
5. Fetal growth restriction. Practice Bulletin No. 134. American College of Obstetricians and Gynecologists. Obstet Gynecol 2013;121:1122–33.
6. Thyroid disease in pregnancy. ACOG Practice Bulletin No. 37. American College of Obstetricians and Gynecologists. Obstet Gynecol 2002;100:387–96.
7. Wallenstein MB, Harper LM, Odibo AO, Roehl KA, Longman RE, Macones GA, et al.. Fetal congenital heart disease and intrauterine growth restriction: a retrospective cohort study. J Matern Fetal Neonatal Med 2012;25:662–5.
8. Raynor BD, Richards D. Growth retardation in fetuses with gastroschisis. J Ultrasound Med 1997;16:13–6.
9. Khoury MJ, Erickson JD, Cordero JF, McCarthy BJ. Congenital malformations and intrauterine growth retardation: a population study. Pediatrics 1988;82:83–90.
10. Gardosi J, Madurasinghe V, Williams M, Malik A, Francis A. Maternal and fetal risk factors for stillbirth: population based study. BMJ 2013;346:f108.
11. Flenady V, Koopmans L, Middleton P, Froen JF, Smith GC, Gibbons K, et al.. Major risk factors for stillbirth in high-income countries: a systematic review and meta-analysis. Lancet 2011;377:1331–40.
12. Goetzinger KR, Cahill AG, Macones GA, Odibo AO. Echogenic bowel on second-trimester ultrasonography: evaluating the risk of adverse pregnancy outcome. Obstet Gynecol 2011;117:1341–8.
13. Dolk H, Loane M, Garne E. The prevalence of congenital anomalies in Europe. Adv Exp Med Biol 2010;686:349–64.
14. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996;87:163–8.
15. Brantberg A, Blaas HG, Salvesen KA, Haugen SE, Eik-Nes SH. Surveillance and outcome of fetuses with gastroschisis. Ultrasound Obstet Gynecol 2004;23:4–13.
16. Costello JM, Polito A, Brown DW, McElrath TF, Graham DA, Thiagarajan RR, et al.. Birth before 39 weeks' gestation is associated with worse outcomes in neonates with heart disease. Pediatrics 2010;126:277–84.
17. Cnota JF, Gupta R, Michelfelder EC, Ittenbach RF. Congenital heart disease infant death rates decrease as gestational age advances from 34 to 40 weeks. J Pediatr 2011;159:761–5.
18. Colvin J, Bower C, Dickinson JE, Sokol J. Outcomes of congenital diaphragmatic hernia: a population-based study in Western Australia. Pediatrics 2005;116:e356–63.
19. Porter A, Benson CB, Hawley P, Wilkins-Haug L. Outcome of fetuses with a prenatal ultrasound diagnosis of isolated omphalocele. Prenat Diagn 2009;29:668–73.
20. Evertson LR, Gauthier RJ, Schifrin BS, Paul RH. Antepartum fetal heart rate testing. I. Evolution of the nonstress test. Am J Obstet Gynecol 1979;133:29–33.
21. Getahun D, Ananth CV, Kinzler WL. Risk factors for antepartum and intrapartum stillbirth: a population-based study. Am J Obstet Gynecol 2007;196:499–507.
22. Facchinetti F, Alberico S, Benedetto C, Cetin I, Cozzolino S, Di Renzo GC, et al.. A multicenter, case-control study on risk factors for antepartum stillbirth. J Matern Fetal Neonatal Med 2011;24:407–10.
23. Boyd PA, Haeusler M, Barisic I, Loane M, Garne E, Dolk H. Paper 1: The EUROCAT network—organization and processes. Birth Defects Res A Clin Mol Teratol 2011;91(suppl 1):S2–15.
24. Levi S. Ultrasound in prenatal diagnosis: polemics around routine ultrasound screening for second trimester fetal malformations. Prenat Diagn 2002;22:285–95.
25. Wapner RJ. Genetics of stillbirth. Clin Obstet Gynecol 2010;53:628–34.
26. Korteweg FJ, Bouman K, Erwich JJ, Timmer A, Veeger NJ, Ravise JM, et al.. Cytogenetic analysis after evaluation of 750 fetal deaths: proposal for diagnostic workup. Obstet Gynecol 2008;111:865–74.
27. Odibo AO, Francis A, Cahill AG, Macones GA, Crane JP, Gardosi J. Association between pregnancy complications and small-for-gestational-age birth weight defined by customized fetal growth standard versus a population-based standard. J Matern Fetal Neonatal Med 2011;24:411–7.
28. Dudley NJ. A systematic review of the ultrasound estimation of fetal weight. Ultrasound Obstet Gynecol 2005;25:80–9.