OUP user menu

Risk of congenital anomalies detected during antenatal serum screening in women with pregestational diabetes

J.G. Ray , M.J. Vermeulen , C. Meier , P.R. Wyatt
DOI: http://dx.doi.org/10.1093/qjmed/hch107 651-653 First published online: 14 September 2004


Background: Most studies comparing women with and without pregestational diabetes mellitus have not systematically screened for fetal anomalies in early pregnancy, potentially leading to selection bias.

Aim: To evaluate the risk for certain congenital anomalies in women participating in an antenatal maternal screening program.

Design: Retrospective cohort study.

Methods: We studied all women who underwent antenatal maternal serum screening in Ontario from 1994 to 2000. Fetal anomalies were documented antenatally by ultrasonography or at autopsy, and postnatally diagnosed birth defects were recorded after 20 weeks gestational age for all live- and stillborn affected infants. We compared the risk of open neural tube defects and urinary tract defects among women with and without pregestational diabetes.

Results: Of 413 219 women screened during pregnancy, 2069 (0.5%) had diabetes. Compared to non-diabetic women, the adjusted odds ratios (95%CI) for neural tube and urinary tract defects among women with diabetes were 2.5 (0.9–6.8) and 2.6 (1.4–4.9), respectively.

Discussion: Among women who undergo second trimester maternal serum screening, pregestational diabetes is associated with an increased risk of having a fetus with an open neural tube defect or urinary tract disorder.


The offspring of women with pregestational diabetes mellitus (DM) are known to be at increased risk for certain congenital anomalies, especially in the presence of poor periconceptional glycemic control.1 However, most completed epidemiological studies have not systematically screened for fetal anomalies antenatally among women with and without DM,2–6 creating a potential selection bias. For example, many pregnancies affected by an open neural tube defect (NTD) end in termination,7 suggesting that live- and stillbirth data alone are not sufficiently sensitive.7–9 We studied the association between pregestational DM and open NTD and urinary tract anomalies in the offspring of women participating in an antenatal maternal screening (MSS) program.


This retrospective cohort study included all women in Ontario who underwent antenatal MSS from January 1994 to August 2000. Under the universal provincial health insurance plan, level II fetal ultrasonography, MSS and genetic counselling were available to all pregnant women residing in Ontario. Since 1993, standardized MSS has been made accessible at 15 to 20 weeks' gestation through any physician or midwife, with a mean rate of uptake of >70%. Maternal date of birth, weight and the presence of pregestational DM are recorded in a standardized fashion on the MSS requisition sheet, completed at the time of screening, and usually based on maternal reporting.

MSS measures alpha-fetoprotein, human chorionic gonadotropin and unconjugated oestriol. A positive screening for open NTD is defined as a gestation-specific concentration of serum alpha-fetoprotein ≥2.2× the mean. All women with a positive MSS are referred for counselling at one of 17 genetics centres across Ontario. Each centre contributes follow-up data to the Ontario Maternal Serum Screening Database funded by the Ministry of Health and Longterm Care of Ontario, based in Toronto, Ontario.

Congenital anomalies were detected by ultrasonography during the antenatal period, or by fetal autopsy following therapeutic termination, and were based on the follow-up records collected by each genetics centre. A postnatally diagnosed anomaly comprised all live- and still-born affected infants after 20 weeks' gestation identified through data linkage of the mother's Ontario health insurance number with that of her infant during the delivery hospitalization, through the Canadian Institute for Health Information Discharge Abstract Database.

We assessed for open NTD and urinary tract disorders (Table 1) because they may be related to the presence of maternal DM,1 and each has a different mode of detection. Specifically, while the majority of open NTD are detected in utero,9 and end in therapeutic termination,7 urinary tract defects may be detected before birth, but often indirectly so, due to the presence of oligohydraminios.10

View this table:
Table 1

Risk of congenital anomalies among offspring of women with and without pregestational diabetes mellitus

DiabetesOpen neural tube defect* (n = 285)Urinary tract defect** (n = 792)
n (rate per 1000)Crude OR (95%CI)Adjusted OR*** (95%CI)n (rate per 1000)Crude OR (95%CI)Adjusted OR*** (95%CI)
Absent281 (0.7)1.0 (ref)1.0 (ref)781 (1.9)1.0 (ref)1.0 (ref)
Present4 (1.9)2.8 (1.0–7.6)2.5 (0.9–6.8)11 (5.3)2.8 (1.5–5.1)2.6 (1.4–4.9)
  • *Isolated anencephalus, spina bifida with hydrocephalus and isolated spina bifida (ICD-9 740.0, 741.0 or 741.9). **Ureteral duplication, cystic kidney, renal dysgenesis or hydronephrosis (ICD-9 753, 753.1 or 753.2). ***Adjusted for maternal age (1-year increments), weight (1-kg increments) and year of screening. ICD-9, International Classification of Diseases, ninth revision, diagnostic code; OR, odds ratio.

Mean maternal age and weight were compared according to diabetes status by t-test. The risk of each anomaly, in association with DM, was expressed as a crude odds ratio (OR) and 95%CI. Logistic regression analysis was used to adjust the OR for maternal age, weight and year of screening. All statistical analyses were performed using SAS Version 8.0. Statistical significance was set at a 2-sided p < 0.05. There were no participant identifiers in the data set analysed, and permission to conduct this study was obtained through a research protocol approved through the Ministry of Health and Longterm Care in Ontario.


Over the seven-year study period, 413 219 women underwent MSS, of whom 2069 (0.5%) had a diagnosis of pregestational DM. The mean (SD) ages of those with and without DM were 30.1 (5.1) years and 30.1 (5.0) years, respectively (p = 0.60). The mean respective weights differed: 74.9 (17.0) vs. 66.8 (14.3) kg (p < 0.001).

The number of cases of NTD and urinary tract disorders are listed in Table 1, of which 68.8% and 3.0% of cases, respectively, were identified antenatally. The presence of DM increased the risk of NTD (crude OR 2.8, 95%CI 1.0–7.6) but was no longer significant in the multivariate analysis (OR 2.5, 95%CI 0.9–6.8). The risk of urinary tract disorders was significantly higher among women with DM in the univariate (OR 2.8, 95%CI: 1.4–5.1) and multivariate (OR 2.6, 95%CI 1.4–4.9) analyses (Table 1).


Among women who underwent second trimester MSS, we observed a higher risk of open NTD and urinary tract disorders in association with the presence of pregestational DM.

We did not include women with a spontaneous pregnancy loss before 15 weeks, which is more common in women with DM who have poor periconceptional glycaemic control.4,,11 Further, women who choose to undergo MSS may differ from those who do not. While the combined use of antenatal and postnatal evaluation may have improved the accuracy of our risk estimates, especially NTD, we may have missed some anomalies.9,,10 This may have reduced the accuracy of the estimated risk of NTD and urinary tract disorders in the entire population. Our study could not distinguish women with type 1 from those with type 2 DM, their glycaemic control, or their use of folic acid periconceptionally.4 While a self-reported diagnosis of DM appears to be quite accurate, in both the pregnant and non-pregnant state, it may underestimate the true prevalence of DM.12,,13

These data offer clinicians a contemporary estimate of the rates of occurrence of open NTD and urinary tract defects among women with pregestational DM. Our estimated rates were somewhat lower than expected.1,3,5,,6 One explanation is that all women in Ontario receive publicly funded health care and pre-pregnancy counselling, which would be relevant in the era after the Diabetes Control And Complications Trial was published,14 which showed that good DM care, with tight glycaemic control, improves pregnancy outcome.15 The best available evidence suggests that women with DM should be encouraged to plan their pregnancy, attempt to normalize their blood glucose concentrations, and take a daily folic acid supplement.1,,4

Future research should attempt to include, at the earliest gestational age possible, all women with DM who have a confirmed pregnancy. Thereafter, these women can be surveyed for spontaneous abortion and its cause, and their offspring can be assessed for structural anomalies, detected both in utero and at birth.7 Given that some anomalies (e.g. urinary tract defects) may not be detected in the antenatal or neonatal periods, epidemiological surveillance in early childhood is also recommended.


We thank both the Ontario provincial laboratories and genetics clinics for contributing data to the Ontario MSS Database, and the women of Ontario for supporting the maternal serum screening program. Both the Spina Bifida and Hydrocephalus Association of Canada and the physicians of Ontario, through the Physicians' Services Incorporated Foundation, financially supported this study.


View Abstract