Risks of specific congenital anomalies in offspring of women with diabetes: A systematic review and meta-analysis of population-based studies including over 80 million births

Abstract

Background

Pre-gestational diabetes mellitus (PGDM) has been known to be a risk factor for congenital heart defects (CHDs) for decades. However, the associations between maternal PGDM and gestational diabetes mellitus (GDM) and the risk of specific types of CHDs and congenital anomalies (CAs) in other systems remain under debate. We aimed to investigate type-specific CAs in offspring of women with diabetes and to examine the extent to which types of maternal diabetes are associated with increased risk of CAs in offspring.

Methods and findings

We searched PubMed and Embase from database inception to 15 October 2021 for population-based studies reporting on type-specific CAs in offspring born to women with PGDM (combined type 1 and 2) or GDM, with no limitation on language. Reviewers extracted data for relevant outcomes and performed random effects meta-analyses, subgroup analyses, and multivariable meta-regression. Risk of bias appraisal was performed using the Cochrane Risk of Bias Tool. This study was registered in PROSPERO (CRD42021229217). Primary outcomes were overall CAs and CHDs. Secondary outcomes were type-specific CAs. Overall, 59 population-based studies published from 1990 to 2021 with 80,437,056 participants met the inclusion criteria. Of the participants, 2,407,862 (3.0%) women had PGDM and 2,353,205 (2.9%) women had GDM. The meta-analyses showed increased risks of overall CAs/CHDs in offspring born to women with PGDM (for overall CAs, relative risk [RR] = 1.99, 95% CI 1.82 to 2.17, P < 0.001; for CHDs, RR = 3.46, 95% CI 2.77 to 4.32, P < 0.001) or GDM (for overall CAs, RR = 1.18, 95% CI 1.13 to 1.23, P < 0.001; for CHDs, RR = 1.50, 95% CI 1.38 to 1.64, P < 0.001). The results of the meta-regression analyses showed significant differences in RRs of CAs/CHDs in PGDM versus GDM (all P < 0.001). Of the 23 CA categories, excluding CHD-related categories, in offspring, maternal PGDM was associated with a significantly increased risk of CAs in 21 categories; the corresponding RRs ranged from 1.57 (for hypospadias, 95% CI 1.22 to 2.02) to 18.18 (for holoprosencephaly, 95% CI 4.03 to 82.06). Maternal GDM was associated with a small but significant increase in the risk of CAs in 9 categories; the corresponding RRs ranged from 1.14 (for limb reduction, 95% CI 1.06 to 1.23) to 5.70 (for heterotaxia, 95% CI 1.09 to 29.92). The main limitation of our analysis is that some high significant heterogeneity still persisted in both subgroup and sensitivity analyses.

Conclusions

In this study, we observed an increased rate of CAs in offspring of women with diabetes and noted the differences for PGDM versus GDM. The RRs of overall CAs and CHDs in offspring of women with PGDM were higher than those in offspring of women with GDM. Screening for diabetes in pregnant women may enable better glycemic control, and may enable identification of offspring at risk for CAs.

Introduction

Currently, the global prevalence of diabetes is increasing among women of reproductive age [1,2]. A diabetic intrauterine environment can cause placental dysfunction and hormone alterations, which could lead to various congenital anomalies (CAs) in offspring of women with diabetes [1]. Notably, pre-gestational diabetes mellitus (PGDM, which includes type 1 and 2 diabetes) has been known to be a risk factor for congenital heart defects (CHDs) for decades [3]. However, there is controversy among current research regarding the association between maternal PGDM and the risk of specific types of CHDs and other CAs of the nervous, digestive, genitourinary, and musculoskeletal systems [4–7]. Further studies are thus needed for clarification of this risk.

Gestational diabetes mellitus (GDM), which is defined as any degree of glucose intolerance with onset or first recognition during pregnancy, is one of the most common complications of pregnancy and affects up to 9%–26% of the obstetric population [8,9]. Similar to PGDM, GDM also has a considerable impact on the health outcomes of the mother and infant during pregnancy, delivery, and beyond. Recently, an increasing number of studies have concentrated on evaluating the risks of specific types of CAs in offspring born to women with GDM [4,5,10–12]. The early period of organogenesis, which occurs during the third to eighth week of gestation, is an important time for organ development. However, hyperglycemia associated with GDM occurs after this critical early window for organogenesis. Therefore, the question as to whether there is an association between GDM and the risk of specific types of CAs in offspring remains.

Previous meta-analyses have mainly focused on the associations between maternal diabetes and CHDs in offspring, and little is known about the influence of maternal diabetes on other specific types of CAs [13,14]. Additionally, new data from population-based studies of more than 36 million births have provided solid estimates of the risk of CHDs in offspring of women with diabetes [4,10–12]. This considerable amount of data could also be used to explore the association between maternal diabetes and other types of CAs. Currently, a quantitative summary of population-based studies on the associations between maternal diabetes (pre-gestational or gestational) and type-specific CAs in offspring is lacking. Comprehensive estimates of the risk of specific CAs associated with maternal diabetes are needed to counsel patients and for public health purposes. Moreover, it is essential that estimates are provided according to different types of maternal diabetes, given the diversity in etiology, treatment, and prognosis.

We performed a detailed systematic review and large-scale meta-analysis to summarize and quantify the existing population-based data on type-specific CAs in offspring of women with diabetes. Furthermore, we examined the extent to which specific types of maternal diabetes (i.e., pre-gestational or gestational) are associated with increased risk of CAs in offspring.

Methods

We performed a literature search in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [15] and Meta-analysis Of Observational Studies in Epidemiology (MOOSE) guidelines [16] (see S1 Text). Before study selection, the protocol for this review was registered in PROSPERO, registration number CRD42021229217 (S1 Protocol).

Search strategy and inclusion criteria

We searched PubMed and Embase from database inception to 15 October 2021. The search strategy combined Medical Subject Headings (MeSH) and Embase subject heading (Emtree) terms with other unindexed or free text terms, with no limitation on language. Details of the full search strategy are provided in S1 Text. Reference lists of retrieved articles and previous systematic and narrative reviews were searched manually to retrieve all relevant documents. Duplicate citations were removed.

Population-based cross-sectional, case–control, and cohort studies that reported original data were eligible for inclusion if they (1) reported any CAs in offspring born to women with diabetes (i.e., pre-gestational [combined type 1 and 2] or gestational diabetes), (2) had a comparison group that included mothers without diabetes, and (3) provided sufficient data from which a risk estimate could be calculated if a risk estimate was not reported. All conference abstracts, guidelines, case reports, case series, commentaries, letters, and animal studies were excluded.

Two independent authors (S-YG and T-NZ) reviewed the titles and abstracts to identify any relevant studies. The full texts of potentially eligible studies that appeared to meet the inclusion criteria were then obtained and independently evaluated by the 2 reviewers. Any disagreement was settled by consensus among all authors. If multiple studies were derived from the same dataset and reported the same associated outcome, the study with the most complete findings or the greatest number of participants was included for analysis. The literature review and study selection process referenced the PRISMA flowchart (Fig 1). When information needed for inclusion in the analyses was missing, the Library of Shengjing Hospital of China Medical University helped us get full access of the relevant data.

Fig 1 Flowchart of selection of studies included in the meta-analysis. (Figure not reproduced here; see the original article.)

Data extraction

A standardized, pre-designed spreadsheet was used for extracting data from the included studies. Study quality and synthesis of evidence were assessed. The following data were recorded in the spreadsheet: first author, publication year, geographic location, study period, study design, data source, type of diabetes, sample size, types of birth, ascertainment of exposure, definition of outcome, outcome risk estimates and 95% confidence intervals (CIs), and adjusted confounders.

Primary outcomes were the rates of overall CAs and type-specific CHDs (heterotaxia, conotruncal defects, atrioventricular septal defect, anomalous pulmonary venous return, left ventricular outflow tract defect, right ventricular outflow tract defect, septal defects, and single ventricle). Secondary outcomes were the rates of other type-specific CAs (involving the nervous system; eye, ear, face, and neck; orofacial clefts; digestive system; abdominal wall; genitourinary system; and musculoskeletal system). S1 Table shows the definitions of these outcomes.

Two authors (T-NZ and S-YG) independently performed data extraction according to the Cochrane Handbook guidelines [17]. Findings were reported according to PRISMA [15] and MOOSE guidance [16]. Any disagreement was settled by consensus among all authors. For studies that did not report any adjusted effect sizes, the crude risk estimate was used. If an included study reported several risk estimates, we extracted the fully adjusted effect sizes. For studies that reported the risk estimates of CAs stratified by isolated and multiple statuses, we used the effect sizes of the isolated CAs. Because odds ratios, prevalence rate ratios, and hazard ratios are excellent approximations of risk ratios in the case of rare outcomes [18], all risk estimates are referred to and reported as relative risks (RRs) for simplicity. If an included study lacked required data, we asked for help from the Library of Shengjing Hospital of China Medical University to get the missing information.

Risk of bias and study quality

The risk of bias assessment was conducted by T-NZ and S-YG using the Risk of Bias in Non-randomized Studies–of Interventions (ROBINS-I) tool [19]. This tool comprises 7 domains: bias due to confounding, bias in the selection of participants, bias in the classification of interventions, bias due to deviations from intended interventions, bias due to missing data, bias in measurement of outcomes, and bias in the selection of the reported result. We rated the possible risk of bias in each of the 7 domains as low risk, moderate risk, serious risk, critical risk, or no information for each available outcome of each included study.

Statistical analysis

For studies that reported effect sizes separately, the results were pooled using a fixed effects model to obtain an overall estimate and then included in the pooled effect size in the meta-analysis. The effective count method proposed by Hamling et al. [20] was used to recalculate the effect sizes. If a selected study did not include an effect size, the unadjusted risk estimate and 95% CI were calculated from the raw data for simplicity using EpiCalc 2000 (https://en.freedownloadmanager.org/Windows-PC/EpiCalc-2000-FREE.html). Estimates were pooled using the DerSimonian and Laird random effects model to calculate summarized RRs and 95% CIs [21], in which I² values were calculated as indicators of heterogeneity. I² values of ≤25%, 26%–50%, 51%–74%, and ≥75% were considered to indicate no, low, moderate, and high heterogeneity between the included studies, respectively [22]. For the primary outcomes of the study, subgroup analyses were undertaken to explore causes of heterogeneity: by region (Europe, North/South America, or Asia-Pacific), year of enrollment (categorized using the median as the cutoff value: before 1997 or in or after 1997), number of participants (categorized using the median as the cutoff value: <282,260 or ≥282,260), and adjustment for confounders (i.e., maternal age, race/ethnicity, body mass index, education, smoking/alcohol consumption, parity, and pregnancy complications). Heterogeneity between subgroups was evaluated by meta-regression analysis if data were reported in more than 10 studies, following to the Cochrane guidelines [23]. Meta-regression analyses were also used to examine the extent to which the types of maternal diabetes (i.e., pre-gestational or gestational) are associated with increased risk of overall CAs/CHDs in offspring. Publication bias was examined by inspecting funnel plots for the outcomes and was further evaluated with Begg’s test [24] and Egger’s test [25] if sufficient studies existed (n ≥ 10) [17]. A sensitivity analysis was undertaken to explore the association of each study with the overall pooled estimate. Statistical analyses were conducted using Stata version 13.0 (StataCorp, College Station, Texas). A 2-tailed P value less than 0.05 was considered statistically significant.

Results

Search results and study characteristics

We identified 24,989 potentially eligible articles in PubMed and Embase through the search strategy plus 1 additional article through hand searching. Of these, 2,331 records were duplicates (Fig 1). In total, 103 articles qualified for full-text review based on title and abstract screening. Of these, an additional 44 articles were excluded for the following reasons: 3 studies enrolled mothers with any type of diabetes and did not distinguish pre-gestational and gestational diabetes, 1 study did not have a comparison group that included mothers without diabetes, 4 studies were derived from the same dataset and reported the same associated outcome as another included study, 3 studies enrolled mothers with CAs, 6 studies included data that could not be extracted or calculated, and 27 studies were not population-based (S2 Table). Finally, 59 population-based studies (published from 1990 to 2021) that met all eligibility criteria contributed to the quantitative synthesis and included a total of 80,437,056 participants (range of participants per study: 155 to 29,211,974) for analysis. Of these, 2,407,862 (3.0%) women had PGDM and 2,353,205 (2.9%) women had GDM; 879,156 cases of overall CAs and 350,051 cases of CHDs were observed. S3 Table gives the method of ascertainment of maternal diabetes of the included studies. Of the 59 studies included, there were 27 studies from the European region (United Kingdom [26–30], Finland [31,32], Sweden [10,33–37], Denmark [10,28,35,38,39], Norway [6,28,40,41], Hungary [7,42–45], Germany [28,46], Netherlands [28], Belgium [28], Wales [28,30], Ireland [28,30], Switzerland [28], France [28,47], Italy [28,48,49], Spain [28], Portugal [28], and Malta [28]), 25 studies from North/South America (United States [4,5,11,12,14,50–57], Canada [50,58–68], and Brazil [69]), and 7 studies from the Asian-Pacific region (China [70–72], Russia [73], Australia [74], and Qatar [75,76]). Table 1 summarizes the characteristics of the included studies, and a more detailed breakdown can be found in S4 Table.

Bias assessment

We assessed the risk of bias for 34 of 59 included studies using ROBINS-I. The assessments are summarized for primary outcomes in Figs A–D in S1 Fig. None of the included studies were rated with a low risk of bias in all domains. The main causes of serious or critical bias risk according to ROBINS-I were weaknesses in the confounding bias domain, selection of participant bias domain, and missing data bias domain.

Exposure to PGDM/GDM and overall CAs in offspring

We first explored whether there was an association between maternal diabetes and overall CAs (not including CHDs) in offspring. Nineteen studies investigated the relationship between maternal PGDM and overall CAs in offspring [4,11,12,27,29,38,44,46,49,56,58,61–63,65,67,70,74,76], and 15 studies investigated the relationship between maternal GDM and overall CAs in offspring [4,11,12,33,39,44,48,56,58,61,62,65,67,69,76]. Our results suggested that maternal PGDM was associated with overall CAs in offspring (RR = 1.99, 95% CI 1.82 to 2.17, I² = 90.0%, P < 0.001; Fig 2), with no evidence of publication bias (Begg’s P = 0.88, Egger’s P = 0.30; Fig E in S1 Fig). A similar association was observed for overall CAs in offspring of women with type 1 diabetes (RR = 2.03, 95% CI 1.66 to 2.48, I² = 82.5%, P < 0.001; Fig K1 in S1 Fig) and in offspring of women with GDM (RR = 1.18, 95% CI 1.13 to 1.23, I² = 76.0%, P < 0.001; Fig 3), with no evidence of publication bias (Begg’s P = 0.39, Egger’s P = 0.32; Fig F in S1 Fig). However, there was no statistically significant association of the risk of overall CAs in offspring of women with type 2 diabetes (RR = 1.31, 95% CI 0.80 to 2.15, I² = 98.2%, P < 0.001; Fig L1 in S1 Fig).

Fig 2 Forest plot of the RRs in population-based studies for maternal pre-gestational diabetes and the risk of overall congenital anomalies (RR = 1.99, 95% CI 1.82 to 2.17, I2 = 90.0%, P < 0.001).Analytical weights are from random effects meta-analysis. Grey boxes represent study estimates; their size is proportional to the respective analytical weight. Lines through the boxes represent the 95% CIs around the study estimates. The diamond represents the mean estimate and its 95% CI. The vertical red dashed line indicates the mean estimate. CI, confidence interval; DL, DerSimonian and Laird random effects model; RR, relative risk. (Figure not reproduced here; see the original article.)

Fig 3 Forest plot of the RRs in population-based studies for maternal gestational diabetes and the risk of overall congenital anomalies (RR = 1.18, 95% CI 1.13 to 1.23, I2 = 76.0%, P < 0.001).Analytical weights are from random effects meta-analysis. Grey boxes represent study estimates; their size is proportional to the respective analytical weight. Lines through the boxes represent the 95% CIs around the study estimates. The diamond represents the mean estimate and its 95% CI. The vertical red dashed line indicates the mean estimate. CI, confidence interval; DL, DerSimonian and Laird random effects model; RR, relative risk. (Figure not reproduced here; see the original article.)

Exposure to PGDM and CHDs in offspring

A total of 18 studies reported on the association between maternal PGDM and CHDs in offspring [4,6,10–12,28–30,38,44,47,49,52,54,56,63,68,74]. Our results suggested that there is a statistically significant increase in risk of CHDs in offspring of women with PGDM (RR = 3.46, 95% CI 2.77 to 4.32, I² = 98.2%, P < 0.001; Fig 4), with no evidence of publication bias (Begg’s P = 0.60, Egger’s P = 0.85; Fig E in S1 Fig). Similarly, maternal type 1 and type 2 diabetes were associated with increased risk of CHDs in offspring (type 1: RR = 3.75, 95% CI 1.86 to 7.57, I² = 99.1%, P < 0.001; Fig K2 in S1 Fig; type 2: RR = 3.15, 95% CI 1.72 to 5.78, I² = 93.6%, P < 0.001; Fig J2 in S1 Fig). Notably, we found that maternal PGDM was associated with increased risk of all specific types of CHDs available for examination in the present study. The RRs of specific types of CHDs ranged from 2.23 (for hypoplastic left heart, 95% CI 1.07 to 4.64, I² = 64.0%, P = 0.040) to 12.16 (for truncus arteriosus, 95% CI 7.52 to 19.68, I² = 0%, P = 0.866) (Table 2; Figs G1–G18 in S1 Fig).

Fig 4 Forest plot of the RRs in population-based studies for maternal pre-gestational diabetes and the risk of congenital heart defects (RR = 3.46, 95% CI 2.77 to 4.32, I2 = 98.2%, P < 0.001).Analytical weights are from random effects meta-analysis. Grey boxes represent study estimates; their size is proportional to the respective analytical weight. Lines through the boxes represent the 95% CIs around the study estimates. The diamond represents the mean estimate and its 95% CI. The vertical red dashed line indicates the mean estimate. CI, confidence interval; DL, DerSimonian and Laird random effects model; RR, relative risk. (Figure not reproduced here; see the original article.)

Exposure to GDM and CHDs in offspring

Eleven studies explored the relationship between GDM and CHDs in offspring [4,6,11,12,38,44,47,51,52,56,72]. Our results suggested that maternal GDM is associated with CHDs (RR = 1.50, 95% CI 1.38 to 1.64, I² = 81.2%, P < 0.001; Fig 5), with no evidence of publication bias (Begg’s P = 0.837, Egger’s P = 0.885; Fig F in S1 Fig). Regarding specific types of CHDs, we found that offspring of women with GDM had an increased risk of heterotaxia (RR = 5.70, 95% CI 1.09 to 29.92, I² = 85.7%, P = 0.008), tetralogy of Fallot (RR = 1.41, 95% CI 1.20 to 1.66, I² = 0%, P = 0.600), left ventricular outflow tract defect (RR = 1.67, 95% CI 1.15 to 2.41, I² = 50.0%, P = 0.112), coarctation of aorta (RR = 1.50, 95% CI 1.23 to 1.83, I² = 35.4%, P = 0.213), right ventricular outflow tract defect (RR = 1.25, 95% CI 1.03 to 1.53, I² = 0%, P = 0.739), ventricular septal defect (RR = 1.31, 95% CI 1.24 to 1.38, I² = 0%, P = 0.960), and atrial septal defect (RR = 1.45, 95% CI 1.40 to 1.50, I² = 0%, P = 0.426) (Table 2; Figs I1–I15 in S1 Fig).

Fig 5 Forest plot of the RRs in population-based studies for maternal gestational diabetes and the risk of congenital heart defects (RR = 1.50, 95% CI 1.38 to 1.64, I2 = 81.2%, P < 0.001).Analytical weights are from random effects meta-analysis. Grey boxes represent study estimates; their size is proportional to the respective analytical weight. Lines through the boxes represent the 95% CIs around the study estimates. The diamond represents the mean estimate and its 95% CI. The vertical red dashed line indicates the mean estimate. CI, confidence interval; DL, DerSimonian and Laird random effects model; RR, relative risk. (Figure not reproduced here; see the original article.)

Exposure to PGDM and other type-specific CAs in offspring

We examined the associations between maternal PGDM and other type-specific CAs in offspring. Our results suggested that offspring of women with PGDM had an increased risk of CAs of the nervous system (RR = 2.54, 95% CI 1.73 to 3.73, I² = 94.8%, P < 0.001); eye, ear, face, and neck (RR = 3.14, 95% CI 2.90 to 3.39, I² = 0%, P = 0.444); digestive system (RR = 2.02, 95% CI 1.24 to 3.28, I² = 92.3%, P < 0.001); genitourinary system (RR = 1.73, 95% CI 1.35 to 2.21, I² = 89.2%, P < 0.001); and musculoskeletal system (RR = 1.98, 95% CI 1.45 to 2.72, I² = 94.4%, P < 0.001), as well as an increased risk of multiple CAs (RR = 3.06, 95% CI 2.36 to 3.96, I² = 39.6%, P = 0.158). The associations were statistically significant in 14 of 16 type-specific CA categories. The corresponding RRs ranged from 1.57 (for hypospadias, 95% CI 1.22 to 2.02, I² = 74.1%, P < 0.001) to 18.18 (for holoprosencephaly, 95% CI 4.03 to 82.06, I² = 66.3%, P = 0.085) (Table 3; Figs H1–H25 in S1 Fig).

Exposure to GDM and other type-specific CAs in offspring

Maternal GDM was associated with an increased risk of CAs of the eye, ear, face, and neck (RR = 1.15, 95% CI 1.09 to 1.22, I² = 0%, P = 0.355) and musculoskeletal system (RR = 1.18, 95% CI 1.15 to 1.22, I² = 0%, P = 0.424) in offspring. In addition, maternal GDM also contributed to an increased risk of specific types of CAs in offspring, including hydrocephaly (RR = 1.34, 95% CI 1.16 to 1.54, I² = 0%, P = 0.960), holoprosencephaly (RR = 1.87, 95% CI 1.09 to 3.22, I² = 0%, P = 0.558), cleft lip with or without cleft palate (RR = 1.26, 95% CI 1.19 to 1.34, I² = 0%, P = 0.547), diaphragmatic hernia (RR = 1.21, 95% CI 1.08 to 1.37, I² = 0%, P = 0.779), omphalocele (RR = 1.21, 95% CI 1.05 to 1.40, I² = 0%, P = 0.743), and hypospadias (RR = 1.29, 95% CI 1.16 to 1.44, I² = 45.9%, P = 0.100) (Table 3; Figs J1–J22 in S1 Fig).

Subgroup, meta-regression, and sensitivity analyses

The sensitivity analysis evaluated the effect of omitting 1 study at a time from each analysis. In the sensitivity analysis, we observed that the high I² value of 81.2% shown in the results for CHDs in offspring of mothers with GDM reduced to a moderate I² value of 55.0% when excluding the study by Billionnet et al. [47]. Although the increased risk association remained robust across scenarios, some moderate to significant heterogeneity still persisted and could not be reduced in sensitivity analyses. To explore the source of heterogeneity, we performed subgroup and meta-regression analyses in the predefined subgroups of study location, year of enrollment, study sample size, and adjustment for confounders (Tables 4 and 5). The findings of increased overall CA/CHD risk associated with maternal diabetes were consistently observed in most of the subgroup analyses. The results of the subgroup analyses suggested that differences in study sample size, population region, year of enrollment, and adjustment for confounders were major sources of heterogeneity. We observed that the high I² value of 81.2% observed in the results for CHDs in offspring of mothers with GDM was reduced to no (I² = 0%) or low (I² = 47.7%) heterogeneity after adjustment for race/ethnicity, body mass index, education, smoking/alcohol consumption, parity, and pregnancy complications (Table 5). In addition, the results of meta-regression analyses showed statistically significant differences in the RRs of CAs/CHDs in PGDM versus GDM (all P_{meta-regression} < 0.001) (Fig 6).

Fig 6 Risks of overall congenital anomalies and congenital heart defects in offspring according to different types of maternal diabetes.Relative risks (RRs) and 95% confidence intervals (CIs) are presented to show the risk of overall congenital anomalies and congenital heart defects in offspring born to women with different types of maternal diabetes compared with the risk among offspring born to women without diabetes. Pmeta-regression values were <0.001 for the comparison within congenital anomalies and congenital heart defects between gestational diabetes and pre-gestational diabetes. (Figure not reproduced here; see the original article.)

Discussion

To the best of our knowledge, the present study is the first comprehensive systematic review and meta-analysis of population-based studies of over 80 million participants that shows an increased risk of type-specific CAs, especially CHDs, in offspring of women with pre-gestational or gestational diabetes. The study findings suggested that maternal PGDM was associated with a significant increase in the risk of CAs in offspring in 38 of 45 categories, while maternal GDM was associated with a small but significant increase in the risk of CAs in 16 of 45 categories. The corresponding RRs of overall CAs/CHDs in offspring of women with PGDM were higher than those in offspring of women with GDM, with no overlap in the 95% CIs.

Although the exact pathophysiology of the relationship between maternal diabetes and CAs in offspring remains unclear, metabolic changes in women with diabetes could play a critical role in the development of CAs in their offspring. Sustained hyperglycemia is the main characteristic of diabetes; this activates multiple metabolic pathways that play a role in the formation of CAs [77]. Notably, a common mechanism behind diabetic complications is mitochondrial superoxide radical production [77]. The production of reactive oxygen species (ROS) is induced by hyperglycemia, which is the crucial process in diabetes mellitus pathogenesis, and oxidative stress (OS) is known to affect embryonic development [77–79]. Although OS does not cause a direct genotoxic effect, a previous study showed that OS affects the expression of some genes involved in the various stages of embryonic development, and each gene affected might have a specific sensitivity to hyperglycemic conditions and changes in the cellular redox state, thus mediating the formation of CAs in offspring [77]. Studies have suggested that the formation of CAs involves AMP-activated protein kinase (AMPK), an enzyme with kinase activity that is activated in response to an increase in adenosine monophosphate nucleoside levels [78,80]. AMPK could play a key role in the formation of CAs; it participates in the regulation of energy metabolism and, once activated, moves into the cell nucleus and phosphorylates multiple proteins, including hypoxia-inducible factor 1α, which could mediate the development of CAs [77]. However, further study is needed to discern the exact mechanisms involved in the activation of AMPK and the induction of CAs by ROS.

An important finding from our meta-regression analyses is the statistically significant difference in the risk of overall CAs in offspring of women with PGDM versus offspring of women with GDM. That is, the risk of CAs in offspring was higher in women with PGDM than in those with GDM. Pregnancy begins at fertilization, and organogenesis begins during the third to eighth week post-conception and continues until birth. Therefore, the first trimester of pregnancy is the most critical period for organogenesis. In women with PGDM, there can be a lengthy period of sustained hyperglycemia before and during pregnancy, which can significantly impact organogenesis and contribute to CAs in offspring. This differs from GDM, which is usually diagnosed during the 24th to 28th week of gestation [8]. Therefore, in a woman with GDM, blood glucose levels could be normal or just slightly elevated during the first trimester, leading to minimal influence on organogenesis. This could partly explain why offspring of women with PGDM were at greater risk for CAs compared with offspring of women with GDM. However, women who develop GDM during pregnancy usually have evidence of metabolic dysfunction before pregnancy, such as pancreatic β-cell defects and increased insulin resistance [81,82], which may contribute to the development of hyperglycemia and thus increase the rate of malformations in infants, although further studies are needed to elucidate the potential mechanisms involved. Another key finding from our meta-regression analysis was the similar result observed regarding CHDs in offspring of women with PGDM and offspring of women with GDM. The heart is the first functional organ to develop and starts to beat and pump blood at around 22 days after fertilization [83]. The septum, including the interatrial septum, begins to form at 4 to 7 weeks of gestation.

Hyperglycemia could have a more critical influence on heart development in the early stage of pregnancy than in the late stage of pregnancy. Therefore, screening for diabetes in pregnant women will enable better glycemic control, which might reduce the rate of malformations, especially during organogenesis. However, the exact mechanisms underlying the influence of diabetes on organogenesis in different stages remain unclear and require further study.

In the present study, the results regarding specific types of CHDs in offspring with maternal PGDM were consistent with 2 previous meta-analyses [13,14]. A recent systematic review and meta-analysis conducted by Chen and colleagues involved a pooled analysis of 24 population-based studies and 18 hospital-based studies; the findings suggested that maternal GDM was significantly associated with the risk of most phenotypes of CHDs [13]. New data from population-based studies of more than 36 million births provided solid estimates of the associations between maternal GDM and specific types of CHDs in offspring [4,10–12]. However, these studies mainly focused on the association between different types of maternal diabetes and CHDs. Little is known about the association between maternal diabetes and other specific types of CAs in offspring or the extent to which types of maternal diabetes are associated with the increased risk of CAs.

One recent meta-analysis by Parimi and colleagues explored the association between maternal diabetes and the risk in offspring of CAKUT, which refers to a range of structural and functional anomalies of the kidney, collecting system, bladder, and urethra [84]. Our findings were in line with the results from Parimi et al. [84] that showed that offspring of mothers with PGDM had an almost 2-fold increased risk of CAKUT; however, results regarding the association between maternal GDM and the risk of CAKUT were inconsistent. Our findings demonstrated associations between maternal diabetes and 23 CA categories (excluding CHD-related categories) in offspring and suggested that offspring of women with PGDM had an increased risk of 21 specific types of CAs, while increased risks of 9 specific types of CAs were observed in offspring of women with GDM.

Strengths and limitations

Our study has several strengths. The first strength is the large sample size of over 80 million births from population-based data, which provides robust evidence regarding the risk of CAs in offspring of women with diabetes and are widely generalizable. Second, our study examined the associations between maternal diabetes and various types of CAs across multiple categories of maternal diabetes. Unlike previous studies [13,14,84] that only assessed the risk of CHDs or CAKUT in maternal diabetes, the present study systematically and quantitatively summarized the associations between maternal diabetes and 45 type-specific CAs in offspring. Third, consistent results of the pooled RRs supported the robustness of the findings of our study. Finally, the current study examined the extent to which types of maternal diabetes (i.e., pre-gestational and gestational) are associated with increased risk of CAs in offspring. The relative consistency of associations observed appears to support the hypothesis that maternal diabetes, especially PGDM, increases the likelihood of type-specific CAs in offspring.

However, several limitations should be noted. Although the increased risk association remained robust across various scenarios, some high levels of statistical heterogeneity generally persisted and could not be reduced in subgroup and sensitivity analyses. There were some causes of heterogeneity in the included studies. First, there is lack of consensus and uniformity in the screening standards and diagnostic criteria for GDM. Also, pre-pregnancy diabetes is sometimes unrecognized and discovered only during pregnancy as GDM, which could lead to overestimation of RRs associated with GDM. Second, the ascertainment of some CAs may vary substantially between studies. Some CAs are easy to ascertain (e.g., anencephaly), while some may not be recognized immediately after birth and may be discovered only later in infancy (e.g., milder atrial septal defects). This also contributes to the heterogeneity of the results. Third, most studies included live births only; the lack of information on stillbirths and terminations of pregnancy for fetal anomaly could introduce selection bias and lead to underestimation of the strength of the associations between maternal diabetes and risk of CAs in offspring. Fourth, there may be other unmeasured confounding factors in addition to those adjusted for in each study. In this regard, further study could be performed to reduce the aforementioned causes of heterogeneity in a more in-depth analysis. An additional limitation was that although we summarized and quantified the existing population-based data on overall CAs/CHD observed under maternal type 1 or type 2 diabetes, data on other type-specific CAs in offspring associated with maternal type 1 or type 2 diabetes are limited. Additional studies are needed to address this issue. Furthermore, information on treatment (e.g., insulin use) or how well-controlled blood glucose levels were in the study participants was not available in most of the studies included in the current study. Further work should strive to address this lack of information. Finally, we observed a negative association between maternal GDM and risk of gastroschisis. The reasons for why maternal GDM was inversely associated with the risk of gastroschisis are currently unknown; this finding warrants confirmation and further investigation in future studies. Residual confounding may contribute to the inverse association, but further confirmation is still needed.

Conclusion

In the present study, we observed an increased rate of CAs in the offspring of women with maternal diabetes and noted the differences between PGDM and GDM. Considering the substantial rise in the prevalence of maternal diabetes over recent decades, the expectation that this prevalence will continue to increase, the number of pregnancies worldwide, and the significant individual and global burdens associated with CAs in offspring, screening for diabetes in pregnant women may enable better glycemic control, and may enable identification of offspring at risk for CAs.

Supporting information

Acknowledgements

We thank Professor Qi-Jun Wu, who kindly provided suggestions for this meta-analysis.

References

1. McCanceDR, MareshM, SacksDA. A practical manual of diabetes in pregnancy. 2nd edition. Hoboken (NJ): Wiley; 2017.
2. FerraraA. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care. 2007;30(Suppl 2):S141–6. doi: 10.2337/dc07-s206 17596462 doi:10.2337/dc07-s206 PubMed
3. JenkinsKJ, CorreaA, FeinsteinJA, BottoL, BrittAE, DanielsSR, et al. Noninherited risk factors and congenital cardiovascular defects: current knowledge: a scientific statement from the American Heart Association Council on Cardiovascular Disease in the Young: endorsed by the American Academy of Pediatrics. Circulation. 2007;115:2995–3014. doi: 10.1161/CIRCULATIONAHA.106.183216 17519397 doi:10.1161/CIRCULATIONAHA.106.183216 PubMed
4. SchrawJM, LangloisPH, LupoPJ. Comprehensive assessment of the associations between maternal diabetes and structural birth defects in offspring: a phenome-wide association study. Ann Epidemiol. 2021;53:14–20.e8. doi: 10.1016/j.annepidem.2020.08.006 32920098 doi:10.1016/j.annepidem.2020.08.006 PubMed
5. TinkerSC, GilboaSM, MooreCA, WallerDK, SimeoneRM, KimSY, et al. Specific birth defects in pregnancies of women with diabetes: National Birth Defects Prevention Study, 1997–2011. Am J Obstet Gynecol. 2020;222:176.e1–11. doi: 10.1016/j.ajog.2019.08.028 31454511PMC7186569 doi:10.1016/j.ajog.2019.08.028 PubMed
6. LeirgulE, BrodwallK, GreveG, VollsetSE, HolmstrømH, TellGS, et al. Maternal diabetes, birth weight, and neonatal risk of congenital heart defects in Norway, 1994–2009. Obstet Gynecol. 2016;128:1116–25. doi: 10.1097/AOG.0000000000001694 27741197 doi:10.1097/AOG.0000000000001694 PubMed
7. VereczkeyA, GerencsérB, CzeizelAE, SzabóI. Association of certain chronic maternal diseases with the risk of specific congenital heart defects: a population-based study. Eur J Obstet Gynecol Reprod Biol. 2014;182:1–6. doi: 10.1016/j.ejogrb.2014.08.022 25216447 doi:10.1016/j.ejogrb.2014.08.022 PubMed
8. American Diabetes Association. 2. Classification and diagnosis of diabetes: standards of medical care in diabetes—2019. Diabetes Care. 2019;42(Suppl 1):S13–28. doi: 10.2337/dc19-S002 30559228 doi:10.2337/dc19-S002 PubMed
9. SacksDA, HaddenDR, MareshM, DeerochanawongC, DyerAR, MetzgerBE, et al. Frequency of gestational diabetes mellitus at collaborating centers based on IADPSG consensus panel-recommended criteria: the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study. Diabetes Care. 2012;35:526–28. doi: 10.2337/dc11-1641 22355019PMC3322716 doi:10.2337/dc11-1641 PubMed
10. ArendtLH, PedersenLH, PedersenL, OvesenPG, HenriksenTB, LindhardMS, et al. Glycemic control in pregnancies complicated by pre-existing diabetes mellitus and congenital malformations: a Danish population-based study. Clin Epidemiol. 2021;13:615–26. doi: 10.2147/CLEP.S298748 34345185PMC8325058 doi:10.2147/CLEP.S298748 PubMed
11. WuY, LiuB, SunY, DuY, SantillanMK, SantillanDA, et al. association of maternal prepregnancy diabetes and gestational diabetes mellitus with congenital anomalies of the newborn. Diabetes Care. 2020;43:2983–90. doi: 10.2337/dc20-0261 33087319PMC7770264 doi:10.2337/dc20-0261 PubMed
12. YangG-R, DyeTD, LiD. Effects of pre-gestational diabetes mellitus and gestational diabetes mellitus on macrosomia and birth defects in upstate New York. Diabetes Res Clin Pract. 2019;155:107811. doi: 10.1016/j.diabres.2019.107811 31401151PMC8783133 doi:10.1016/j.diabres.2019.107811 PubMed
13. ChenL, YangT, ChenL, WangL, WangT, ZhaoL, et al. Risk of congenital heart defects in offspring exposed to maternal diabetes mellitus: an updated systematic review and meta-analysis. Arch Gynecol Obstet. 2019;300:1491–506. doi: 10.1007/s00404-019-05376-6 31713644 doi:10.1007/s00404-019-05376-6 PubMed
14. HoangTT, MarengoLK, MitchellLE, CanfieldMA, AgopianAJ. Original findings and updated meta-analysis for the association between maternal diabetes and risk for congenital heart disease phenotypes. Am J Epidemiol. 2017;186:118–28. doi: 10.1093/aje/kwx033 28505225PMC5860024 doi:10.1093/aje/kwx033 PubMed
15. PageMJ, McKenzieJE, BossuytPM, BoutronI, HoffmannTC, MulrowCD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLoS Med. 2021;18:e1003583. doi: 10.1371/journal.pmed.1003583 33780438PMC8007028 doi:10.1371/journal.pmed.1003583 PubMed
16. StroupDF, BerlinJA, MortonSC, OlkinI, WilliamsonGD, RennieD, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–12. doi: 10.1001/jama.283.15.2008 10789670 doi:10.1001/jama.283.15.2008 PubMed
17. HigginsJPT, GreenS, editors. Cochrane handbook for systematic reviews of interventions. Cochrane Collaboration; 2011.
18. RothmanKJ, GreenlandS, LashTL. Modern epidemiology. 3rd edition. Philadelphia: Lippincott Williams & Wilkins; 2014. doi: 10.1097/EDE.0000000000000062 doi:10.1097/EDE.0000000000000062
19. SterneJA, HernánMA, ReevesBC, SavovićJ, BerkmanND, ViswanathanM, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. doi: 10.1136/bmj.i4919 27733354PMC5062054 doi:10.1136/bmj.i4919 PubMed
20. HamlingJ, LeeP, WeitkunatR, AmbühlM. Facilitating meta-analyses by deriving relative effect and precision estimates for alternative comparisons from a set of estimates presented by exposure level or disease category. Stat Med. 2008;27:954–70. doi: 10.1002/sim.3013 17676579 doi:10.1002/sim.3013 PubMed
21. DerSimonianR, LairdN. Meta-analysis in clinical trials. Control Clin Trials. 1986;7:177–88. doi: 10.1016/0197-2456(86)90046-2 3802833 doi:10.1016/0197-2456(86)90046-2 PubMed
22. HigginsJPT, ThompsonSG, DeeksJJ, AltmanDG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–60. doi: 10.1136/bmj.327.7414.557 12958120PMC192859 doi:10.1136/bmj.327.7414.557 PubMed
23. DeeksJJ, HigginsJPT, AltmanDG. Analysing data and undertaking meta-analyses. In: HigginsJPT, ThomasJ, ChandlerJ, CumpstonM, LiT, PageMJ, et al, editors. Cochrane handbook for systematic reviews of interventions. Version 6.0. Cochrane Collaboration; 2019 [cited 2021 Mar 21]. Available from: https://training.cochrane.org/handbook/archive/v6/chapter-10.
24. BeggCB, MazumdarM. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50:1088–101. 7786990 PubMed
25. EggerM, Davey SmithG, SchneiderM, MinderC. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315:629–34. doi: 10.1136/bmj.315.7109.629 9310563PMC2127453 doi:10.1136/bmj.315.7109.629 PubMed
26. AllenLA, Cannings-JohnRL, EvansA, ThayerDS, FrenchR, ParanjothyS, et al. Pregnancy in teenagers diagnosed with type 1 diabetes mellitus in childhood: a national population-based e-cohort study. Diabetologia. 2020;63:799–810. doi: 10.1007/s00125-019-05063-w 31863141PMC7054376 doi:10.1007/s00125-019-05063-w PubMed
27. DarkeJ, GlinianaiaSV, MarsdenP, BellR. Pregestational diabetes is associated with adverse outcomes in twin pregnancies: a regional register-based study. Acta Obstet Gynecol Scand. 2016;95:339–46. doi: 10.1111/aogs.12825 26600048 doi:10.1111/aogs.12825 PubMed
28. GarneE, LoaneM, DolkH, BarisicI, AddorMC, ArriolaL, et al. Spectrum of congenital anomalies in pregnancies with pregestational diabetes. Birth Defects Res A Clin Mol Teratol. 2012;94:134–40. doi: 10.1002/bdra.22886 22371321 doi:10.1002/bdra.22886 PubMed
29. BellR, GlinianaiaSV, TennantPWG, BilousRW, RankinJ. Peri-conception hyperglycaemia and nephropathy are associated with risk of congenital anomaly in women with pre-existing diabetes: a population-based cohort study. Diabetologia. 2012;55:936–47. doi: 10.1007/s00125-012-2455-y 22314812 doi:10.1007/s00125-012-2455-y PubMed
30. MacintoshMC, FlemingKM, BaileyJA, DoyleP, ModderJ, AcoletD, et al. Perinatal mortality and congenital anomalies in babies of women with type 1 or type 2 diabetes in England, Wales, and Northern Ireland: population based study. BMJ. 2006;333:177. doi: 10.1136/bmj.38856.692986.AE 16782722PMC1513435 doi:10.1136/bmj.38856.692986.AE PubMed
31. RaitioA, TauriainenA, LeinonenMK, SyvänenJ, KemppainenT, LöyttyniemiE, et al. Maternal risk factors for gastroschisis: a population-based case-control study. Birth Defects Res. 2020;112:989–95. doi: 10.1002/bdr2.1703 32406607 doi:10.1002/bdr2.1703 PubMed
32. MowlaS, GisslerM, RäisänenS, KancherlaV. Association between maternal pregestational diabetes mellitus and spina bifida: a population-based case-control study, Finland, 2000–2014. Birth Defects Res. 2020;112:186–95. doi: 10.1002/bdr2.1624 31774241 doi:10.1002/bdr2.1624 PubMed
33. HildénK, HansonU, PerssonM, MagnusonA, SimmonsD, FadlH. Gestational diabetes and adiposity are independent risk factors for perinatal outcomes: a population based cohort study in Sweden. Diabet Med. 2019;36:151–7. doi: 10.1111/dme.13843 30698864PMC6590111 doi:10.1111/dme.13843 PubMed
34. LudvigssonJF, NeoviusM, SöderlingJ, GudbjörnsdottirS, SvenssonAM, FranzénS, et al. Periconception glycaemic control in women with type 1 diabetes and risk of major birth defects: population based cohort study in Sweden. BMJ. 2018;362:k2638. doi: 10.1136/bmj.k2638 29976596PMC6031927 doi:10.1136/bmj.k2638 PubMed
35. ArendtLH, LindhardMS, HenriksenTB, OlsenJ, CnattingiusS, PeterssonG, et al. Maternal diabetes mellitus and genital anomalies in male offspring: a nationwide cohort study in 2 Nordic countries. Epidemiology. 2018;29:280–9. doi: 10.1097/EDE.0000000000000781 29112520 doi:10.1097/EDE.0000000000000781 PubMed
36. PerssonM, FadlH. Perinatal outcome in relation to fetal sex in offspring to mothers with pre-gestational and gestational diabetes—a population-based study. Diabet Med. 2014;31:1047–54. doi: 10.1111/dme.12479 24773081 doi:10.1111/dme.12479 PubMed
37. FadlHE, OstlundIKM, MagnusonAFK, HansonUSB. Maternal and neonatal outcomes and time trends of gestational diabetes mellitus in Sweden from 1991 to 2003. Diabet Med. 2010;27:436–41. doi: 10.1111/j.1464-5491.2010.02978.x 20536516 doi:10.1111/j.1464-5491.2010.02978.x PubMed
38. ØyenN, DiazLJ, LeirgulE, BoydHA, PriestJ, MathiesenER, et al. Prepregnancy diabetes and offspring risk of congenital heart disease: a nationwide cohort study. Circulation. 2016;133:2243–53. doi: 10.1161/CIRCULATIONAHA.115.017465 27166384PMC4890838 doi:10.1161/CIRCULATIONAHA.115.017465 PubMed
39. WuCS, NohrEA, BechBH, VestergaardM, OlsenJ. Long-term health outcomes in children born to mothers with diabetes: a population-based cohort study. PLoS ONE. 2012;7:e36727. doi: 10.1371/journal.pone.0036727 22649497PMC3359312 doi:10.1371/journal.pone.0036727 PubMed
40. KlungsøyrK, NordtveitTI, KaastadTS, SolbergS, SlettenIN, VikAK. Epidemiology of limb reduction defects as registered in the Medical Birth Registry of Norway, 1970–2016: population based study. PLoS ONE. 2019;14(7):e0219930. doi: 10.1371/journal.pone.0219930 31314783PMC6636750 doi:10.1371/journal.pone.0219930 PubMed
41. EidemI, SteneLC, HenriksenT, HanssenKF, VangenS, VollsetSE, et al. Congenital anomalies in newborns of women with type 1 diabetes: nationwide population-based study in Norway, 1999–2004. Acta Obstet Gynecol Scand. 2010;89:1403–11 doi: 10.3109/00016349.2010.518594 20929418 doi:10.3109/00016349.2010.518594 PubMed
42. MavrogenisS, UrbanR, CzeizelAE. Pregnancy complications in the mothers who delivered boys with isolated hypospadias—a population-based case-control study. J Matern Fetal Neonatal Med. 2015;28:489–93. doi: 10.3109/14767058.2014.921902 24809222 doi:10.3109/14767058.2014.921902 PubMed
43. Csáky-SzunyoghM, VereczkeyA, KósaZ, GerencsérB, CzeizelAE. Risk factors in the origin of congenital left-ventricular outflow-tract obstruction defects of the heart: a population-based case-control study. Pediatr Cardiol. 2014;35:108–20. doi: 10.1007/s00246-013-0749-6 23843102 doi:10.1007/s00246-013-0749-6 PubMed
44. BánhidyF, AcsN, PuhóEH, CzeizelAE. Congenital abnormalities in the offspring of pregnant women with type 1, type 2 and gestational diabetes mellitus: a population-based case-control study. Congenit Anom (Kyoto). 2010;50:115–21. doi: 10.1111/j.1741-4520.2010.00275.x 20184644 doi:10.1111/j.1741-4520.2010.00275.x PubMed
45. NielsenGL, NørgardB, PuhoE, RothmanKJ, SørensenHT, CzeizelAE. Risk of specific congenital abnormalities in offspring of women with diabetes. Diabet Med. 2005;22:693–96. doi: 10.1111/j.1464-5491.2005.01477.x 15910618 doi:10.1111/j.1464-5491.2005.01477.x PubMed
46. von KriesR, KimmerleR, SchmidtJE, HachmeisterA, BöhmO, WolfHG. Pregnancy outcomes in mothers with pregestational diabetes: a population-based study in North Rhine (Germany) from 1988 to 1993. Eur J Pediatr. 1997;156:963–7. doi: 10.1007/s004310050752 9453382 doi:10.1007/s004310050752 PubMed
47. BillionnetC, MitanchezD, WeillA, NizardJ, AllaF, HartemannA, et al. Gestational diabetes and adverse perinatal outcomes from 716,152 births in France in 2012. Diabetologia. 2017;60:636–44. doi: 10.1007/s00125-017-4206-6 28197657PMC6518373 doi:10.1007/s00125-017-4206-6 PubMed
48. SeghieriG, Di CianniG, SeghieriM, LacariaE, CorsiE, LencioniC, et al. Risk and adverse outcomes of gestational diabetes in migrants: a population cohort study. Diabetes Res Clin Pract. 2020;163:108128. doi: 10.1016/j.diabres.2020.108128 32259610 doi:10.1016/j.diabres.2020.108128 PubMed
49. VincetiM, MalagoliC, RothmanKJ, RodolfiR, AstolfiG, CalzolariE, et al. Risk of birth defects associated with maternal pregestational diabetes. Eur J Epidemiol. 2014;29:411–8. doi: 10.1007/s10654-014-9913-4 24861339 doi:10.1007/s10654-014-9913-4 PubMed
50. ParkerSE, YazdyMM, TinkerSC, MitchellAA, WerlerMM. The impact of folic acid intake on the association among diabetes mellitus, obesity, and spina bifida. Am J Obstet Gynecol. 2013;209:239.e1–8. doi: 10.1016/j.ajog.2013.05.047 23711668PMC3989866 doi:10.1016/j.ajog.2013.05.047 PubMed
51. AlversonCJ, StricklandMJ, GilboaSM, CorreaA. Maternal smoking and congenital heart defects in the Baltimore-Washington Infant Study. Pediatrics. 2011;127:e647–53. doi: 10.1542/peds.2010-1399 21357347 doi:10.1542/peds.2010-1399 PubMed
52. CorreaA, GilboaSM, BesserLM, BottoLD, MooreCA, HobbsCA, et al. Diabetes mellitus and birth defects. Am J Obstet Gynecol. 2008;199:237.e1–9. doi: 10.1016/j.ajog.2008.06.028 18674752PMC4916956 doi:10.1016/j.ajog.2008.06.028 PubMed
53. AndersonJL, WallerDK, CanfieldMA, ShawGM, WatkinsML, WerlerMM. Maternal obesity, gestational diabetes, and central nervous system birth defects. Epidemiology. 2005;16:87–92. doi: 10.1097/01.ede.0000147122.97061.bb 15613950 doi:10.1097/01.ede.0000147122.97061.bb PubMed
54. LoffredoCA, WilsonPD, FerenczC. Maternal diabetes: an independent risk factor for major cardiovascular malformations with increased mortality of affected infants. Teratology. 2001;64:98–106. doi: 10.1002/tera.1051 11460261 doi:10.1002/tera.1051 PubMed
55. CroenLA, ShawGM, LammerEJ. Risk factors for cytogenetically normal holoprosencephaly in California: a population-based case-control study. Am J Med Genet. 2000;90:320–5. 10710231 PubMed
56. JanssenPA, RothmanI, SchwartzSM. Congenital malformations in newborns of women with established and gestational diabetes in Washington State, 1984–91. Paediatr Perinat Epidemiol. 1996;10:52–63. doi: 10.1111/j.1365-3016.1996.tb00026.x 8746431 doi:10.1111/j.1365-3016.1996.tb00026.x PubMed
57. BecerraJE, KhouryMJ, CorderoJF, EricksonJD. Diabetes mellitus during pregnancy and the risks for specific birth defects: a population-based case-control study. Pediatrics. 1990;85:1–9. 2404255 PubMed
58. LeeD, BoothGL, RayJG, LingV, FeigDS. Undiagnosed type 2 diabetes during pregnancy is associated with increased perinatal mortality: a large population-based cohort study in Ontario, Canada. Diabet Med. 2020;37:1696–704. doi: 10.1111/dme.14250 31994233 doi:10.1111/dme.14250 PubMed
59. DyckRF, KarunanayakeC, PahwaP, StangM, EricksonRL, OsgoodND. Congenital anomalies of the kidney and urinary tract (CAKUT): an emerging relationship with pregestational diabetes mellitus among First Nations and non-First Nations people in Saskatchewan—results from the DIP: ORRIIGENSS Project. Can J Diabetes. 2020:S1499–2671. doi: 10.1016/j.jcjd.2020.10.004 33308984 doi:10.1016/j.jcjd.2020.10.004 PubMed
60. LiuS, EvansJ, MacFarlaneAJ, AnanthCV, LittleJ, KramerMS, et al. Association of maternal risk factors with the recent rise of neural tube defects in Canada. Paediatr Perinat Epidemiol. 2019;33:145–53. doi: 10.1111/ppe.12543 30920008 doi:10.1111/ppe.12543 PubMed
61. ChenL, WangWJ, AugerN, XiaoL, TorrieJ, McHughNG, et al. Diabetes in pregnancy in associations with perinatal and postneonatal mortality in First Nations and non-Indigenous populations in Quebec, Canada: population-based linked birth cohort study. BMJ Open. 2019;9:e025084. doi: 10.1136/bmjopen-2018-025084 30992290PMC6500205 doi:10.1136/bmjopen-2018-025084 PubMed
62. LaiFY, JohnsonJA, DoverD, KaulP. Outcomes of singleton and twin pregnancies complicated by pre-existing diabetes and gestational diabetes: a population-based study in Alberta, Canada, 2005–11. J Diabetes. 2016;8:45–55. doi: 10.1111/1753-0407.12255 25496644 doi:10.1111/1753-0407.12255 PubMed
63. LiuS, RouleauJ, LeónJA, SauveR, JosephKS, RayJG. Impact of pre-pregnancy diabetes mellitus on congenital anomalies, Canada, 2002–2012. Health Promot Chronic Dis Prev Can. 2015;35:79–84. doi: 10.24095/hpcdp.35.5.01 26186019PMC4910455 doi:10.24095/hpcdp.35.5.01 PubMed
64. DartAB, RuthCA, SellersEA, AuW, DeanHJ. Maternal diabetes mellitus and congenital anomalies of the kidney and urinary tract (CAKUT) in the child. Am J Kidney Dis. 2015;65:684–91. doi: 10.1053/j.ajkd.2014.11.017 25595566 doi:10.1053/j.ajkd.2014.11.017 PubMed
65. FeigDS, HweeJ, ShahBR, BoothGL, BiermanAS, LipscombeLL. Trends in incidence of diabetes in pregnancy and serious perinatal outcomes: a large, population-based study in Ontario, Canada, 1996–2010. Diabetes Care. 2014;37:1590–6. doi: 10.2337/dc13-2717 24705609 doi:10.2337/dc13-2717 PubMed
66. LiuS, JosephKS, LisonkovaS, RouleauJ, Van den HofM, SauveR, et al. Association between maternal chronic conditions and congenital heart defects: a population-based cohort study. Circulation. 2013;128:583–9. doi: 10.1161/CIRCULATIONAHA.112.001054 23812182 doi:10.1161/CIRCULATIONAHA.112.001054 PubMed
67. PeticcaP, KeelyEJ, WalkerMC, YangQ, BottomleyJ. Pregnancy outcomes in diabetes subtypes: how do they compare? A province-based study of Ontario, 2005–2006. J Obstet Gynaecol Can. 2009;31:487–96. doi: 10.1016/S1701-2163(16)34210-4 19646313 doi:10.1016/S1701-2163(16)34210-4 PubMed
68. YangJ, CummingsEA, O’ConnellC, JangaardK. Fetal and neonatal outcomes of diabetic pregnancies. Obstet Gynecol. 2006;108:644–50. doi: 10.1097/01.AOG.0000231688.08263.47 16946226 doi:10.1097/01.AOG.0000231688.08263.47 PubMed
69. Oliveira-BrancatiCIF, FerrareseVCC, CostaAR, Fett-ConteAC. Birth defects in Brazil: outcomes of a population-based study. Genet Mol Biol. 2020;43:e20180186. doi: 10.1590/1678-4685-GMB-2018-0186 31429860PMC7198030 doi:10.1590/1678-4685-GMB-2018-0186 PubMed
70. WeiY, XuQ, YangH, YangY, WangL, ChenH, et al. Preconception diabetes mellitus and adverse pregnancy outcomes in over 6.4 million women: a population-based cohort study in China. PLoS Med. 2019;16:e1002926. doi: 10.1371/journal.pmed.1002926 31574092PMC6771981 doi:10.1371/journal.pmed.1002926 PubMed
71. TainY-L, LuhH, LinC-Y, HsuC-N. Incidence and risks of congenital anomalies of kidney and urinary tract in newborns: a population-based case-control study in Taiwan. Medicine. 2016;95:e2659. doi: 10.1097/MD.0000000000002659 26844492PMC4748909 doi:10.1097/MD.0000000000002659 PubMed
72. LiuX, LiuG, WangP, HuangY, LiuE, LiD, et al. Prevalence of congenital heart disease and its related risk indicators among 90,796 Chinese infants aged less than 6 months in Tianjin. Int J Epidemiol. 2015;44:884–93. doi: 10.1093/ije/dyv107 26071138 doi:10.1093/ije/dyv107 PubMed
73. KovalenkoAA, AndaEE, OdlandJØ, NieboerE, BrennT, KrettekA. Risk factors for ventricular septal defects in Murmansk County, Russia: a registry-based study. Int J Env Res Public Health. 2018;15:1320. doi: 10.3390/ijerph15071320 29937526PMC6069126 doi:10.3390/ijerph15071320 PubMed
74. SharpePB, ChanA, HaanEA, HillerJE. Maternal diabetes and congenital anomalies in South Australia 1986–2000: a population-based cohort study. Birth Defects Res A Clin Mol Teratol. 2005;73:605–11. doi: 10.1002/bdra.20172 16007590 doi:10.1002/bdra.20172 PubMed
75. BayoumiMAA, MasriRM, MataniNYS, HendausMA, MasriMM, ChandraP, et al. Maternal and neonatal outcomes in mothers with diabetes mellitus in Qatari population. BMC Pregnancy Childbirth. 2021;21(1):651. doi: 10.1186/s12884-021-04124-6 34560839PMC8464105 doi:10.1186/s12884-021-04124-6 PubMed
76. SolimanA, SalamaH, Al RifaiH, De SanctisV, Al-ObaidlyS, Al QubasiM, et al. The effect of different forms of dysglycemia during pregnancy on maternal and fetal outcomes in treated women and comparison with large cohort studies. Acta Biomed. 2018;89:11–21. doi: 10.23750/abm.v89iS4.7356 30049927PMC6179089 doi:10.23750/abm.v89iS4.7356 PubMed
77. ClapésS, FernándezT, SuárezG. Oxidative stress and birth defects in infants of women with pregestational diabetes. MEDICC Rev. 2013;15:37–40. doi: 10.37757/MR2013V15.N1.9 23396241 doi:10.37757/MR2013V15.N1.9 PubMed
78. WuY, VianaM, ThirumangalathuS, LoekenMR. AMP-activated protein kinase mediates effects of oxidative stress on embryo gene expression in a mouse model of diabetic embryopathy. Diabetologia. 2012;55:245–54. doi: 10.1007/s00125-011-2326-y 21993711PMC3342033 doi:10.1007/s00125-011-2326-y PubMed
79. InzucchiSE. Clinical practice. Diagnosis of diabetes. N Engl J Med. 2012;367:542–50. doi: 10.1056/NEJMcp1103643 22873534 doi:10.1056/NEJMcp1103643 PubMed
80. LoekenMR. Advances in understanding the molecular causes of diabetes-induced birth defects. J Soc Gynecol Investig. 2006;13:2–10. doi: 10.1016/j.jsgi.2005.09.007 16303321 doi:10.1016/j.jsgi.2005.09.007 PubMed
81. McIntyreHD, CatalanoP, ZhangC, DesoyeG, MathiesenER, DammP. Gestational diabetes mellitus. Nat Rev Dis Primers. 2019;5(1):47. doi: 10.1038/s41572-019-0098-8 31296866 doi:10.1038/s41572-019-0098-8 PubMed
82. BuchananTA, XiangAH, PageKA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol. 2012;8(11):639–49. doi: 10.1038/nrendo.2012.96 22751341PMC4404707 doi:10.1038/nrendo.2012.96 PubMed
83. BettsJ, DesaixP, JohnsonE, JodyE, KorolO, KruseD, et al. Anatomy and physiology. OpenStax. 2013 Apr 25.
84. ParimiM, NitschD. A systematic review and meta-analysis of diabetes during pregnancy and congenital genitourinary abnormalities. Kidney Int Rep. 2020;5:678–93. doi: 10.1016/j.ekir.2020.02.1027 32405589PMC7210707 doi:10.1016/j.ekir.2020.02.1027 PubMed