ABSTRACT
Objective
To investigate associations between maternal periconceptional (three months prior through the third pregnancy month) myo‐inositol intake and the odds of selected congenital heart defects in offspring.
Design
A population‐based case–control study using the National Birth Defects Prevention Study (NBDPS) database.
Setting
United States.
Population or Sample
Women with singleton live births without major birth defects (controls) and women with singleton live births, stillbirths, or terminations with selected nonsyndromic congenital heart defects (CHD; cases).
Methods
Descriptive analyses, logistic regression models, ascertainment of myo‐inositol intake from supplements and food using a shortened food frequency questionnaire and survey.
Main Outcome Measures
Odds of CHD.
Results
11 752 cases and 11 415 controls were included. Compared to women not taking myo‐inositol supplements, women with any supplemental intake were less likely to have a pregnancy with the selected congenital heart defects as a group (adjusted odds ratio [aOR] = 0.79; 95% confidence interval [CI] 0.66–0.94) or with septal defects alone (aOR = 0.61; 95% CI 0.46–0.81). Compared to women with low total myo‐inositol intake from food or supplements, women with high total myo‐inositol intake (≥ 500 mg/day) were less likely to have a pregnancy with the selected CHD as a group (aOR = 0.88; 95% CI 0.84–0.93) or conotruncal defects (aOR = 0.87; 95% CI 0.79–0.96); left ventricular outflow tract defects (aOR = 0.87; 95% CI 0.78–0.96); right ventricular outflow tract defects (aOR = 0.85; 95% CI 0.77–0.95); or atrial septal defects (aOR = 0.91; 95% CI 0.83–0.99).
Conclusions
An inverse association was observed between maternal myo‐inositol intake during the periconceptional period and the odds of selected CHDs in offspring.
Keywords: congenital heart defects, folic acid, myo‐inositol
1. Introduction
Congenital heart defects (CHD) are the most common group of birth defects, affecting 1% of deliveries in the United States [1, 2]. Of the 134 million births each year globally, there are 1.3 million born with CHD [1, 2, 3, 4, 5]. Congenital heart defects are associated with an increased risk of mortality, neurocognitive disorders, comorbidities and poor academic achievement [6, 7, 8, 9, 10, 11, 12]. Risk factors for CHD are poorly understood, with only a minority estimated to be caused by known genetic factors, and 2% associated with known non‐genetic factors [13, 14, 15, 16]. Although maternal intake of folic acid and one‐carbon‐rich dietary patterns have been associated with a reduced risk of CHD in offspring, exploring other nutritional factors for CHD reduction would be informative [12, 15, 17, 18].
Myo‐inositol is a carbocyclic sugar important for intracellular signal transduction and lipid transport [19, 20]. Myo‐inositol can be consumed from fruits, beans, nuts, grains and supplements [20]. It has been investigated for its potential to reduce neural tube defects (NTDs) risk [21, 22, 23]. Because NTDs and some CHD share a common developmental pathway that involves neural crest cells [24, 25, 26], it is plausible that myo‐inositol also affects CHD risk. Animal studies demonstrated that myo‐inositol might reduce CHD risk via downregulating the activity of the Wnt/β‐catenin pathway during early pregnancy [27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38]. We used data from a large population‐based case–control study to investigate associations between women's periconceptional intake of myo‐inositol and odds of selected CHD in their offspring.
2. Methods
2.1. Data Source
The National Birth Defects Prevention Study (NBDPS) was a large population‐based case–control study of risk factors for major structural birth defects. The NBDPS protocol was detailed previously. Briefly, NBDPS identified live births, stillbirths and terminations with major birth defects with estimated dates of delivery from October 1, 1997, to December 31, 2011 from 10 state birth defects surveillance programs using active surveillance methods. Abstractors at each programme identified potentially eligible cases from medical records in hospitals, birthing centres, and other medical facilities [39]. The abstracted medical information was reviewed by board‐certified clinical geneticists to confirm eligibility using standard case definitions. Cases that could be explained by chromosomal anomalies or variants in a single gene were excluded [39, 40].
Live births without structural birth defects were randomly identified from birth certificates or hospital records in the same catchment area and periods as cases and were included as control children [39]. Potential control children were identified from vital records (Arkansas [2000–2011], Georgia [2001–2011], Iowa, Massachusetts, North Carolina, New Jersey, Utah) or birth hospital records (Arkansas [1997–2000], California, Georgia [1998–2001], New York, Texas) [39].
Women who delivered cases or controls were eligible for participation in an interview if they spoke English or Spanish, had not previously participated in NBDPS, were not incarcerated, and had legal custody of their child at the time of the interview. Each potential participant was asked to provide informed consent [39]. Participants who provided informed consent completed a 1‐h computer‐assisted telephone interview between 6 weeks and 24 months after their estimated date of delivery. The interviewer covered maternal and paternal demographics, behaviours, and medical history from 3 months before pregnancy to the end of the pregnancy [39]. About 67% of cases and 65% of controls were interviewed [39]. The NBDPS received institutional review board approval from the Centres for Disease Control and Prevention and each participating site [39]. The Institutional Review Board at the University of Arkansas for Medical Sciences approved the NBDPS protocol on October 21, 1997, and continuous approval has been maintained with Record #4812.
2.2. Primary Outcome Measures‐Congenital Heart Defects
Paediatric cardiologists classified CHD using abstracted information from medical records [41]. For this study, ‘any CHD’ (i.e., diagnosed with any eligible CHD subtype) and specific CHD were assessed as outcomes.
2.3. Primary Exposure Measures‐Myo‐Inositol
Maternal myo‐inositol intake from diet, supplements, or both was calculated as exposure variables. During the interview, NBDPS participants completed a food frequency questionnaire (FFQ), a 58‐item shortened modification of the Willett FFQ. This was the only record of dietary intake, which captured usual food intake for the year prior to pregnancy [42, 43].
Women were also asked to provide their intake of cereal and food supplements such as energy drinks during each month from 3 months before pregnancy to the end of pregnancy. Responses to these questions were used to calculate maternal dietary myo‐inositol intake as an approximation of the intake during the periconceptional period (3 months before pregnancy through the 3rd month of pregnancy; B3P3) [39]. Details on the calculation of dietary myo‐inositol intake are provided in Appendix S1. Women were also asked to report their intake of dietary supplements (including vitamins and minerals) during each month from 3 months before pregnancy through the last month of pregnancy, including names of vitamins, duration and frequency. We defined ‘ever use’ as any use of supplements containing myo‐inositol versus ‘never use’ as no use of supplements containing myo‐inositol during B3P3. This information was used to calculate myo‐inositol supplement intake during B3P3, with a detailed description provided in Appendix S2.
Total myo‐inositol intake included dietary and supplementary consumption. The dietary and total maternal myo‐inositol intake was categorised based on the minimum dose suggested for NTDs prevention (500 mg/day) [22, 23, 44, 45, 46]. The dietary and total myo‐inositol intake was also classified by its quartile cut points among cases with the lowest quartile as the reference, to assess the potential dose–response relationships.
2.4. Covariates
Potential covariates were identified from an a priori literature review [47, 48]. Covariates included maternal age at delivery (< 20, 20–29, 30–39 and ≥ 40 years), self‐reported maternal education (< 12, 12 years/high school/GED, 1–3 years of college and ≥ 4 years of college), self‐identified maternal race/ethnicity (NH [Non‐Hispanic]‐White, NH‐Black and other), history of maternal epilepsy or its treatment (yes or no), maternal type 2 diabetes diagnosed at any time (yes or no), and pre‐pregnancy maternal body mass index (BMI) (underweight [< 18.5 kg/m2], normal weight [18.5–24.9 kg/m2], overweight [25–29.9 kg/m2] and obese [≥ 30 kg/m2]) [39].
2.5. Covariates for Sensitivity Analyses
2.5.1. Maternal Folic Acid and Folate Intake
Self‐reported intake of food and supplements was used to calculate folic acid and folate intake during B3P3 using methods similar to myo‐inositol intake calculation. This total intake value was converted into Dietary Folate Equivalents (DFE), adjusting for the higher bioavailability of folic acid, and then categorised by the recommended allowances (600 micrograms/day) for pregnant women [49].
2.5.2. Maternal Diet Quality Index‐Pregnancy
We divided participants into quartiles of total myo‐inositol intake and diet quality index‐pregnancy (DQI‐P), and cross‐classified these quartiles to form 16 groups [50]. We calculated the maternal DQI‐P for participants according to Bodnar's method [50]. Participants in the lowest quartiles of maternal DQI‐P and total myo‐inositol intake were used as the reference group. The remaining participants were categorised into a corresponding group based on their quartile of DQI‐P and total myo‐inositol intake.
Maternal type 1 diabetes diagnosed at any time (yes or no), gestational diabetes (yes or no), maternal intake of alcohol, angiotensin‐converting enzyme inhibitors, and statins during B3P3 (yes or no) were also self‐reported and used in sensitivity analyses.
2.6. Statistical Analyses
Covariates were summarised using mean (standard deviation [SD]) and frequency (proportion) and were compared using Chi‐square tests, Fisher's exact tests and t‐tests. The odds of CHD or no major birth defects were compared in a logistic regression model to assess its association with maternal myo‐inositol intake after adjusting for covariates. A 10% change‐in‐estimate criterion was used to further select covariates [51, 52]. Statistical parameters, such as Hosmer‐Lemeshow goodness‐of‐fit statistic, were used to assess the goodness of fit of the regression models. We identified potential outliers of maternal myo‐inositol intake and covariates using statistical parameters such as inter‐quartile range [53]. We did not find any outliers. Effect modification between myo‐inositol and folic acid/folate was assessed using stratified analysis in logistic regression models.
We observed missing values for exposure variables (1.97%–2.40% of participants) and confounders such as maternal education levels (0%–4.99% of participants) and used multiple imputations to impute missing data. As a sensitivity analysis, we re‐ran the models with the new dataset with imputed data [54, 55]. Factors such as maternal age, folic acid‐containing supplements intake, maternal race/ethnicity, total household income levels, study site, maternal type 2 diabetes diagnosed at any time, and year of due date were used to impute the missing values using fully conditional specification [56]. Log‐transformed ORs from the five imputed datasets were used to calculate a pooled estimation [56].
To account for potential confounding effects, we also included the following variables in the sensitivity analyses: DQI‐P, maternal use of alcohol, angiotensin‐converting enzyme inhibitors and statins during B3P3, maternal type 1 diabetes, gestational diabetes, and maternal intake of folic acid/dietary folate during B3P3. All statistical analyses were conducted using SAS (version 9.4, SAS Institute Inc) [57].
3. Results
There were 12 584 CHD cases and 11 792 liveborn control children with a completed maternal interview. Of the 12 584 CHD cases, those who had an isolated ventricular septal defect, no other CHD, and an estimated date of delivery beginning with January 1, 2006 (due to a change in inclusion criteria) (n = 6, 0.05%); were part of a multiple birth (n = 819, 6.5%); had an unknown pregnancy outcome (n = 1, 0.01%); were missing gestational age (therefore could not determine exposure period); or were livebirths or stillbirths with a gestational age < 20 weeks (n = 6, 0.05%) were excluded. Of the 11 792 liveborn control children, those who were part of a multiple birth (n = 374, 3.2%) or were missing gestational age (n = 3, 0.03%) were excluded. After exclusions, we included 11 752 cases and 11 415 controls (Figure S1).
The distribution of dietary myo‐inositol intake during the year before pregnancy was not significantly different between cases and controls (Table 1); however, supplemental use of myo‐inositol was significantly different, with a slightly higher proportion of ever users among controls (2.1% vs. 2.6%). Compared with control participants, case participants were more likely to be ≥ 40 years old at delivery (3.1% vs. 2.3%), to be obese (20.8% vs. 17.5%), to have type 2 diabetes (2.1% vs. 0.4%), epilepsy or epilepsy treatment (1.0% vs. 0.6%), and gestational diabetes (5.5% vs. 4.4%). Case mothers were less likely to have ≥ 4 years of college education (27.4% vs. 31.6%; Table 1). Over one‐third of cases and controls did not consume 600 mcg/day of DFE (Table 1).
TABLE 1.
Characteristics of women with pregnancy outcomes of live births, stillbirths, and terminations with congenital heart defects (cases) and women with live births without major birth defects (controls), National Birth Defects Prevention Study, 1997–2011.
| Maternal characteristics | Cases (N = 11 752) a | Controls (N = 11 415) b | p c |
|---|---|---|---|
| Maternal intake of dietary myo‐inositol e (mg/day) mean (SD) | 539.3 (2.4) | 551.2 (2.5) | 0.066 |
| Maternal intake of supplements containing myo‐inositol f | |||
| Yes | 241 (2.1%) | 300 (2.6%) | 0.004 |
| No | 11 300 (96.2%) | 10 902 (95.5%) | |
| Missing | 211 (1.8%) | 213 (1.9%) | |
| Maternal age at delivery | |||
| < 20 years | 1076 (9.2%) | 1150 (10.1%) | < 0.001 |
| 20–29 years | 5944 (50.6%) | 5774 (50.6%) | |
| 30–39 years | 4373 (37.2%) | 4232 (37.1%) | |
| ≥ 40 years | 359 (3.1%) | 259 (2.3%) | |
| Maternal race/ethnicity | |||
| Non‐Hispanic ‐White | 6755 (57.5%) | 6557 (57.4%) | 0.750 |
| Non‐Hispanic ‐Black | 1331 (11.3%) | 1260 (11.0%) | |
| Other | 3664 (31.2%) | 3591 (31.5%) | |
| Missing | 2 (0.0%) | 7 (0.1%) | |
| Maternal education | |||
| < 12 years | 1998 (17.0%) | 1859 (16.3%) | < 0.001 |
| 12 years/high school/GED | 2965 (25.2%) | 2646 (23.2%) | |
| 1–3 years of college | 3238 (27.6%) | 2972 (26.0%) | |
| ≥ 4 years of college | 3220 (27.4%) | 3602 (31.6%) | |
| Missing | 331 (2.8%) | 336 (2.9%) | |
| Pre‐pregnancy body mass index (BMI) | |||
| Underweight (BMI < 18.5 kg/m2) | 607 (5.2%) | 580 (5.1%) | < 0.001 |
| Normal weight (BMI ≥ 18.5 and < 25 kg/m2) | 5485 (46.7%) | 5830 (51.1%) | |
| Overweight (25 ≤ BMI < 30 kg/m2) | 2657 (22.6%) | 2469 (21.6%) | |
| Obese (≥ 30 kg/m2) | 2439 (20.8%) | 1998 (17.5%) | |
| Missing | 564 (4.8%) | 538 (4.7%) | |
| Maternal type 2 diabetes d | |||
| Yes | 250 (2.1%) | 48 (0.4%) | < 0.001 |
| No | 11 427 (97.2%) | 11 289 (98.9%) | |
| Missing | 75 (0.6%) | 78 (0.7%) | |
| Maternal epilepsy or epilepsy treatment | |||
| Yes | 116 (1.0%) | 70 (0.6%) | 0.002 |
| No | 11 563 (98.4%) | 11 262 (98.7%) | |
| Missing | 73 (0.6%) | 83 (0.7%) | |
| Maternal intake of dietary folate equivalents (mcg/day) mean (SD) | 507.6 (380.6) | 527.6 (395.4) | |
| Higher or equal to 600 mcg/day | 6655 (56.6%) | 6972 (61.1%) | < 0.001 |
| Lower than 600 mcg/day | 4839 (41.2%) | 4187 (36.7%) | |
| Missing | 258 (2.2%) | 256 (2.2%) | |
| Maternal intake of folic acid from supplements | |||
| Yes | 5997 (51.0%) | 6349 (55.6%) | |
| No | 5517 (47.0%) | 4842 (42.4%) | < 0.001 |
| Missing | 238 (2.0%) | 224 (2.0%) | |
| Gestational diabetes during the index‐pregnancy | |||
| Yes | 644 (5.5%) | 505 (4.4%) | 0.0002 |
| No | 11 033 (93.9%) | 10 832 (94.9%) | |
| Missing | 75 (0.6%) | 78 (0.7%) |
Note: Statistically significant results were highlighted in bold font.
Live births, stillbirths, or terminations with any heart defects were considered cases in this table.
Live births without any major birth defects were controls in this table.
P values obtained based on comparisons between combined cases and controls using χ2 tests for categorical variables and t‐tests for continuous variables.
Diagnosed at any time.
One year before pregnancy.
Three months before pregnancy through the third month of pregnancy.
Ever users of supplemental myo‐inositol were more likely to be NH White (73.9% vs. 57.4%), to have ≥ 4 years of college education (47.1% vs. 29.6%), and to have a normal weight (56.8% vs. 48.9%). They were less likely to be younger than 30 years old (46.4% vs. 60.4%) and to have gestational diabetes (3.1% vs. 5.0%; Table S1). Women with dietary myo‐inositol consumption of 500 mg/day or more were less likely to be NH White (50.6% vs. 68.4%) and to have ≥ 4 years of college education (28.3% vs. 32.1%; Table S2). The distribution of dietary and total myo‐inositol on baseline characteristics was very similar (Table S3).
A lower occurrence of any CHD (adjusted odds ratio or aOR = 0.79; 95% confidence interval or CI 0.66–0.94), septal heart defects (aOR = 0.61; 95% CI 0.46–0.81), and ventricular septal defects (aOR = 0.59; 95% CI 0.40–0.86) were associated with ever use of supplements containing myo‐inositol during the periconceptional period (Table 2). Higher maternal myo‐inositol intake from food (≥ 500 mg/day) during the year before pregnancy was associated with lower odds of any CHD (aOR = 0.88; 95% CI 0.84–0.94), conotruncal defects (aOR = 0.87; 95% CI 0.80–0.96), anomalous pulmonary venous return (aOR = 0.79; 95% CI 0.63–0.99), left ventricular outflow tract defects (aOR = 0.87; 95% CI 0.78–0.96), right ventricular outflow tract defects (aOR = 0.85; 95% CI 0.77–0.95), and atrial septal defects (aOR = 0.91; 95% CI 0.83–0.99) compared with lower dietary myo‐inositol intake (< 500 mg/day; Table 3). Specific CHD phenotypes including d‐transposition of the great arteries (aOR = 0.78; 95% CI 0.66–0.91), aortic stenosis (aOR = 0.81; 95% CI 0.66–0.98), and pulmonary valve stenosis (aOR = 0.85; 95% CI 0.75–0.95) were also associated with higher dietary myo‐inositol intake (Table 3). We observed a similar association between higher maternal periconceptional total myo‐inositol intake (from food and supplements) during B3P3 and decreased odds of CHD (Table S4). With increased maternal consumption of dietary or total myo‐inositol, the odds of left ventricular outflow tract defects, aortic stenosis, right ventricular outflow tract defects, and pulmonary valve stenosis decreased in a monotonic pattern (Table 4; Tables S5 and S6).
TABLE 2.
Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptional a maternal myo‐inositol intake from supplements and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
| CHD | Periconceptional use of supplements with myo‐inositol a | ||
|---|---|---|---|
| Frequency | Adjusted OR (95% CI b ) | ||
| Ever use | Never use (referent) | ||
| Controls | 300 | 10 902 | referent |
| Any CHD | 241 | 11 300 | 0.79 (0.66, 0.94) |
| Conotruncal defects | 55 | 2426 | 0.81 (0.60, 1.09) |
| Truncus arteriosus | 3 | 126 | 0.94 (0.29, 3.01) |
| Tetralogy of fallot | 26 | 1109 | 0.88 (0.59, 1.33) |
| d‐Transposition of the great arteries | 16 | 725 | 0.74 (0.45, 1.24) |
| Double outlet right ventricle c | 3 | 297 | 0.30 (0.07, 1.21) |
| VSD conoventricular d | 4 | 114 | 1.44 (0.52, 3.98) |
| Anomalous pulmonary venous return | 7 | 359 | 0.77 (0.36, 1.64) |
| Left ventricular outflow tract defects | 60 | 2033 | 0.97 (0.72, 1.29) |
| Hypoplastic left heart syndrome | 19 | 608 | 1.16 (0.72, 1.87) |
| Coarctation of the aorta | 28 | 1060 | 0.81 (0.54, 1.23) |
| Aortic stenosis | 18 | 461 | 1.25 (0.76, 2.07) |
| Right ventricular outflow tract defects | 45 | 1895 | 0.88 (0.64, 1.22) |
| Pulmonary atresia | 4 | 240 | 0.65 (0.24, 1.78) |
| Pulmonary valve stenosis e | 36 | 1388 | 0.98 (0.69, 1.41) |
| Ebstein anomaly | 2 | 168 | 0.45 (0.11, 1.84) |
| Tricuspid atresia | 3 | 168 | 0.53 (0.13, 2.16) |
| Septal defects | 62 | 4120 | 0.61 (0.46, 0.81) |
| Ventricular septal defect f | 30 | 1949 | 0.59 (0.40, 0.86) |
| Atrial septal defect g | 53 | 2780 | 0.81 (0.60, 1.10) |
| Atrioventricular septal defects | 11 | 351 | 1.19 (0.64, 2.21) |
| Single ventricle/complex | 4 | 314 | 0.52 (0.19, 1.42) |
| Heterotaxia with CHD | 6 | 331 | 0.81 (0.35, 1.83) |
Note: Statistically significant results were highlighted in bold font.
Three months before pregnancy through the third month of pregnancy.
Model adjusted for maternal age at delivery, maternal race/ethnicity, maternal body mass index, study site, maternal education level, maternal type 2 diabetes diagnosed at any time, maternal epilepsy or epilepsy treatment. ORs did not change after adding gestational diabetes to the regression model.
Double Outlet Right Ventricle (DORV)‐Transposition of the Great Arteries and DORV‐Other.
VSD = Ventricular Septal Defect.
For the calculation of ORs for pulmonary valve stenosis, the number of controls used for ever users, never users and participants with missing information about periconceptional use of supplements with myo‐inositol was 291 (2.7%), 10 448 (95.4%) and 209 (1.9%), respectively.
Includes VSD perimembranous, VSD muscular, VSD not otherwise specified (NOS) and Multiple VSDs.
Includes Atrial Septal Defect (ASD) secundum and ASD NOS.
TABLE 3.
Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptional a maternal dietary myo‐inositol intake and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
| CHD | Dietary myo‐inositol intake a | ||
|---|---|---|---|
| Frequency | Adjusted OR (95% CI) b | ||
| High intake c | Low intake d (referent) | ||
| Controls | 6796 | 4383 | referent |
| Any CHD | 6689 | 4852 | 0.88 (0.84, 0.94) |
| Conotruncal Defects | 1447 | 1037 | 0.87 (0.80, 0.96) |
| Truncus Arteriosus | 65 | 62 | 0.69 (0.48, 1.00) |
| Tetralogy of Fallot | 680 | 456 | 0.93 (0.82, 1.06) |
| d‐Transposition of the great arteries | 405 | 338 | 0.78 (0.66, 0.91) |
| Double Outlet Right Ventricle e | 184 | 119 | 0.87 (0.68, 1.11) |
| VSD conoventricular f | 78 | 40 | 1.30 (0.86, 1.95) |
| Anomalous Pulmonary Venous Return | 209 | 155 | 0.79 (0.63, 0.99) |
| Left Ventricular Outflow Tract Defects | 1175 | 922 | 0.87 (0.78, 0.96) |
| Hypoplastic Left Heart Syndrome | 355 | 269 | 0.88 (0.74, 1.05) |
| Coarctation of the Aorta | 626 | 466 | 0.90 (0.79, 1.03) |
| Aortic Stenosis | 253 | 228 | 0.81 (0.66, 0.98) |
| Right Ventricular Outflow Tract Defects | 1097 | 833 | 0.85 (0.77, 0.95) |
| Pulmonary Atresia | 143 | 98 | 0.85 (0.64, 1.11) |
| Pulmonary Valve Stenosis g | 790 | 627 | 0.85 (0.75, 0.95) |
| Ebstein Anomaly | 94 | 76 | 0.76 (0.55, 1.05) |
| Tricuspid Atresia | 113 | 57 | 1.11 (0.78, 1.56) |
| Septal defects | 2493 | 1694 | 0.94 (0.87, 1.01) |
| Ventricular Septal Defect h | 1216 | 768 | 1.01 (0.91, 1.12) |
| Atrial Septal Defect i | 1670 | 1166 | 0.91 (0.83, 0.99) |
| Atrioventricular Septal Defects | 195 | 167 | 0.83 (0.67, 1.04) |
| Single Ventricle/Complex | 186 | 132 | 0.80 (0.63, 1.02) |
| Heterotaxia with CHD | 201 | 136 | 0.85 (0.67, 1.07) |
Note: Statistically significant results were highlighted in bold font.
Three months before pregnancy through the third month of pregnancy.
Model adjusted for maternal age at delivery, maternal race/ethnicity, maternal body mass index, study site, maternal education level, maternal type 2 diabetes diagnosed at any time, and maternal epilepsy or epilepsy treatment.
Participants with dietary myo‐inositol intake ≥ 500 mg/day during the year before pregnancy were classified as having high dietary myo‐inositol intake.
Participants with dietary myo‐inositol intake < 500 mg/day during the year before pregnancy were classified as having low dietary myo‐inositol intake.
Double Outlet Right Ventricle (DORV) includes DORV‐TGA and DORV‐Other.
VSD = Ventricular Septal Defect.
For the calculation of ORs for Pulmonary Valve Stenosis, the number of controls among women with high dietary myo‐inositol, women with low dietary myo‐inositol intake, and women with missing information about dietary myo‐inositol was 6459, 4256 and 233.
Ventricular Septal Defect (VSD) includes VSD perimembranous, VSD muscular, VSD not otherwise specified (NOS) and multiple VSDs.
Atrial Septal Defect (ASD) includes ASD secundum and ASD NOS.
TABLE 4.
Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptional a maternal dietary myo‐inositol intake by quartile and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
| CHD | Dietary myo‐inositol intake a | |||
|---|---|---|---|---|
| Adjusted OR (95% CI) b | p c | |||
| 2nd quartile d | 3rd quartile d | 4th quartile c | ||
| Any CHD | 0.94 (0.87, 1.01) | 0.87 (0.80, 0.94) | 0.87 (0.80, 0.94) | < 0.001 |
| Conotruncal Defects | 0.85 (0.75, 0.97) | 0.83 (0.73, 0.94) | 0.87 (0.76, 0.99) | 0.028 |
| Truncus Arteriosus | 0.73 (0.45, 1.19) | 0.51 (0.30, 0.86) | 0.71 (0.43, 1.19) | 0.081 |
| Tetralogy of Fallot | 0.83 (0.69, 0.99) | 0.89 (0.75, 1.06) | 0.89 (0.74, 1.08) | 0.370 |
| d‐Transposition of the great arteries | 0.90 (0.73, 1.11) | 0.77 (0.62, 0.96) | 0.85 (0.68, 1.07) | 0.067 |
| Double Outlet Right Ventricle e | 0.81 (0.57, 1.14) | 0.69 (0.49, 0.98) | 0.87 (0.62, 1.22) | 0.326 |
| VSD conoventricular f | 1.07 (0.59, 1.93) | 1.49 (0.86, 2.57) | 1.16 (0.65, 2.09) | 0.391 |
| Anomalous Pulmonary Venous Return | 1.09 (0.81, 1.48) | 0.86 (0.63, 1.17) | 0.78 (0.56, 1.08) | 0.055 |
| Left Ventricular Outflow Tract Defects | 0.89 (0.78, 1.02) | 0.86 (0.75, 0.98) | 0.76 (0.65, 0.88) | 0.000 |
| Hypoplastic Left Heart Syndrome | 0.74 (0.58, 0.93) | 0.75 (0.60, 0.95) | 0.78 (0.61, 0.99) | 0.048 |
| Coarctation of the Aorta | 0.93 (0.78, 1.11) | 0.96 (0.80, 1.14) | 0.77 (0.63, 0.93) | 0.021 |
| Aortic Stenosis | 0.97 (0.75, 1.25) | 0.81 (0.62, 1.06) | 0.74 (0.55, 0.99) | 0.021 |
| Right Ventricular Outflow Tract Defects | 0.93 (0.81, 1.07) | 0.81 (0.70, 0.93) | 0.76 (0.65, 0.88) | < 0.001 |
| Pulmonary Atresia | 0.84 (0.58, 1.21) | 0.67 (0.46, 0.98) | 0.77 (0.52, 1.12) | 0.094 |
| Pulmonary Valve Stenosis | 0.92 (0.79, 1.08) | 0.79 (0.67, 0.93) | 0.74 (0.63, 0.88) | 0.000 |
| Ebstein Anomaly | 1.13 (0.73, 1.76) | 0.99 (0.64, 1.55) | 0.78 (0.48, 1.27) | 0.261 |
| Tricuspid Atresia | 0.83 (0.51, 1.34) | 0.90 (0.57, 1.44) | 1.02 (0.64, 1.61) | 0.828 |
| Septal defects | 0.92 (0.83, 1.03) | 0.92 (0.83, 1.02) | 0.98 (0.88, 1.09) | 0.701 |
| Ventricular Septal Defect g | 0.91 (0.78, 1.05) | 0.98 (0.85, 1.13) | 1.04 (0.90, 1.20) | 0.437 |
| Atrial Septal Defect h | 0.95 (0.84, 1.07) | 0.89 (0.78, 1.01) | 0.95 (0.83, 1.08) | 0.272 |
| Atrioventricular Septal Defects | 1.07 (0.80, 1.43) | 0.82 (0.60, 1.11) | 0.78 (0.56, 1.08) | 0.050 |
| Single Ventricle/Complex | 1.02 (0.74, 1.41) | 0.79 (0.57, 1.11) | 0.77 (0.55, 1.09) | 0.064 |
| Heterotaxia with CHD | 0.94 (0.67, 1.31) | 0.99 (0.72, 1.37) | 0.85 (0.61, 1.18) | 0.401 |
Note: Statistically significant results were highlighted in bold font.
During the year before pregnancy.
Model adjusted for maternal age at delivery, maternal race/ethnicity, maternal body mass index, study site, maternal education level, maternal type 2 diabetes diagnosed at any time, and maternal epilepsy or epilepsy treatment.
P values were calculated to assess the trend of dose–response.
The dietary myo‐inositol intake during the year before pregnancy was used to classify participants into 4 groups according to quartile levels among cases with the following cut points: 354, 574 and 893 mg/day. The lowest quartile (dietary myo‐inositol intake < 354 mg/day) was set as the reference group.
Double Outlet Right Ventricle (DORV) includes DORV‐TGA and DORV‐Other.
VSD = Ventricular Septal Defect.
Ventricular Septal Defect (VSD) includes VSD perimembranous, VSD muscular, VSD not otherwise specified (NOS) and Multiple VSDs.
Atrial Septal Defect (ASD) includes ASD secundum and ASD NOS.
In sensitivity analyses, adjusted ORs were reasonably consistent across DQI‐P categories except for truncus arteriosus and single ventricle (Tables S7.1–4), indicating DQI does not substantially affect the association between myo‐inositol intake and odds of CHD in offspring. Similar inverse associations were observed in the low folic acid/folate group; however, this association became null in the high folic acid/folate group (Table S8). Strong inverse associations with truncus arteriosus and Ebstein anomaly were observed in the high folic acid/folate group (Table S8). We observed similar inverse associations between myo‐inositol intake and the odds of CHD in the remaining sensitivity analyses (Table S9; multiple imputation results not displayed).
4. Discussion
4.1. Main Findings
In this population‐based case–control study, we observed a modest inverse association between maternal myo‐inositol intake during the periconceptional period and odds of CHD in offspring. Dietary and supplemental myo‐inositol intake were associated with any CHD and specific CHD subtypes. These findings were robust in sensitivity analyses after considering the diet quality index, missing values and other confounders. We observed a small but promising dose–response relationship, with higher dietary myo‐inositol intake associated with lower odds of several CHD subtypes. Lack of statistical significance in association analyses may be due to very small sample sizes (n = 3) in subgroups. The differences in the distributions in the variables that we observed may be due to the large sample size and may not be meaningful differences.
Folic acid/folate and myo‐inositol are both accessible from food and supplements. Because folic acid/folate consumption reduced CHD risk [58, 59, 60], we adjusted for folic acid/folate and found an inverse association between myo‐inositol intake and the odds of any CHD in offspring. Similar inverse associations were observed among women with low folic acid/folate intake. However, this association became null among women with high folic acid/folate intake. The effect of folic acid/folate might be so strong that myo‐inositol is no longer noticeable. Despite this, the odds of Truncus Arteriosus and Ebstein Anomaly were lower in the high myo‐inositol group, suggesting the important role of myo‐inositol in the prevention of selected critical CHD.
4.2. Strengths and Limitations
This study has several strengths. First, NBDPS is one of the largest population‐based case–control studies in the United States and includes participants from 10 states with diverse dietary patterns and demographic characteristics [61]. Active case ascertainment was used to recruit NBDPS participants, which increases the quality and completeness of the data [39]. Also, paediatric cardiologists systematically classified CHD, improving the quality and homogeneity of the outcome and reducing the misclassification of the outcome [62, 63, 64, 65]. In addition, trained interviewers collected information and increased the quality of exposure variables and covariates [66].
Our study has some potential limitations. Dietary and supplemental myo‐inositol was assessed via FFQ and recall up to 24 months postpartum. This could lead to potential recall bias that we cannot completely rule out with the limited information in NBDPS data. Secondly, the shortened Willett FFQ might not include all important food items containing myo‐inositol and therefore could lead to biases. We used a published paper as the source of food composition data because there is no standard food composition table for myo‐inositol in the United States Department of Agriculture database or any other validated databases. Despite our efforts, the quantity of myo‐inositol intake may not be fully and accurately captured and could have led to misclassification of the exposure [20, 67]. However, the potential misclassification of the exposure would be similar in cases and controls and would bias the result towards the null [68, 69, 70]. Additionally, we relied on self‐reported myo‐inositol intake because we did not have myo‐inositol measurements at the tissue level. This could potentially introduce errors, particularly given that myo‐inositol is synthesised de novo in humans. Furthermore, analyses were conducted using NH‐White, NH‐Black and Other due to small sample sizes among the more granular races/ethnicities. We used NH‐White, NH‐Black, Hispanic, Asian, Native American and Other and found similar odds ratios. Moreover, the use of myo‐inositol supplements was rare in this study; thus, the results should be interpreted with caution. Also, multiple sensitivity analyses were performed in this study. Although they were hypothesis‐driven and generally aligned with the primary findings, we acknowledge the potential for false positives in our study. Lastly, there might be residual confounding regarding nutritional, socioeconomic, or behavioural factors in our findings.
4.3. Interpretation
Due to the lack of published human studies on myo‐inositol and CHD, our results cannot be directly compared to prior studies; the only studies of myo‐inositol in humans focusing on congenital anomalies were those on NTDs [23, 71]. Greene et al. observed one case of NTD in the placebo plus folic acid group and did not observe any NTD cases in the group taking myo‐inositol plus folic acid [23]. In contrast, Shaw and colleagues reported no association between low dietary myo‐inositol intake and NTDs [71]. The different outcomes (NTDs vs. CHD), as well as different study methods (e.g., different food frequency questionnaires, different timing of intake information, etc.) could explain the differences between the current study and the two previous studies. Furthermore, the biological mechanisms between myo‐inositol and CHD and myo‐inositol and NTDs might not be identical [72, 73].
Myo‐inositol might reduce CHD risk through the Wnt/β‐catenin pathway. Myo‐inositol reduces the formation of inositol pentakisphosphate (IP5) [29]. Because IP5 inhibits the activity of GSK3, and GSK3 is an important intermediary in the Wnt/β‐catenin pathway, excess myo‐inositol could inhibit the Wnt/β‐catenin pathway [29]. This hypothesis is supported by animal studies that found myo‐inositol depletion decreases the activity of GSK3, and supplementing myo‐inositol increases the activity of GSK3 [30, 37]. The Wnt/β‐catenin pathway should be suppressed during the very early stage of heart development, the specification of cardiac progenitor cells and terminal differentiation (before 3 weeks of gestational age), to have normal cardiac development [27, 28]. Potentiation of Wnt signalling during these stages could affect the development of the heart overall and lead to a high risk of cardiac abnormalities in the primary and secondary heart fields [28, 29, 30, 35, 36], which is supported by animal studies [28, 29, 74]. Animal studies also illustrated the protective effect of myo‐inositol on CHD risk via the Wnt pathway [29, 38]. Because of the wide‐ranging role of Wnt in early cardiogenesis, it is conceivable that if indeed a higher intake of myo‐inositol has a protective effect, such effect could be seen in a broad range of CHD [29]. More studies regarding factors in the Wnt pathway are needed.
5. Conclusion
This study suggests a potential modest inverse association between maternal myo‐inositol intake between three months before pregnancy and the third month of pregnancy and the odds of selected CHD in offspring, as the first examination of this relationship in humans. Further replication is required to assess whether such inverse associations could inform dietary strategies for the prevention of at least some CHD, reducing the public health burden of CHD.
Author Contributions
R.C., W.N.N., L.J.S., J.Y., M.S.O. and E.H.B. conceived the study. R.C. accessed the NBDPS data and performed data extraction, variable creation, analysis, and results interpretation. W.N.N., L.J.S., J.Y., M.S.O., and E.H.B. provided important methodological suggestions such as measuring key variables and selecting the appropriate tests and regression models. R.C. led the writing of the manuscript. W.N.N., L.J,S., J.Y., M.S.O., E.H.B., X.L., L.M.A., L.D.B., M.L.B., R.H.F., M.M.J. E.N., A.F.O., P.A.R., and G.M.S. provided crucial feedback on refining the study's methodologies, recommending additional statistical tests, and helping to enhance the logical flow and coherence of the narrative. They also contributed significantly to the interpretation of the key findings, providing insights that clarified the implications of the results.
Disclosure
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centres for Disease Control and Prevention.
Conflicts of Interest
Dr. Finnell formerly held a leadership position with the now‐defunct TeratOmic Consulting and formerly received travel funds to attend editorial board meetings of the Journal of Reproductive and Developmental Medicine. All other authors report no conflicts of interest.
Supporting information
Figure S1: Flow diagram of study participants, National Birth Defects Prevention Study, 1997–2011
Table S1: Distribution of maternal characteristics among participants with different intake of supplements containing myo‐inositol, National Birth Defects Prevention Study, 1997–2011.
Table S2: Distribution of maternal characteristics among participants with different intake of dietary myo‐inositol, National Birth Defects Prevention Study, 1997–2011.
Table S3: Distribution of maternal characteristics among participants with different intake of total myo‐inositol from diet and supplements, National Birth Defects Prevention Study, 1997–2011.
Table S4: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between total maternal periconceptionala myo‐inositol intake and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S5: Distribution of congenital heart defect (CHD) cases and controls across quartiles of periconceptionala maternal dietary myo‐inositol intake, National Birth Defects Prevention Study, 1997–2011.
Table S6: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptionala total maternal myo‐inositol intake by quartile and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐1 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐2 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐3 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐4 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S8: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between total maternal periconceptionala myo‐inositol intake and congenital heart defects (CHDs) stratified by folic acid/folate intake from food and supplements, National Birth Defects Prevention Study, 1997–2011.
Table S9: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptionala maternal dietary myo‐inositol intakeb and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Supplemental Methods
Acknowledgements
We are sincerely grateful to John Farrell, Ph.D. at Boston University, who used the proprietary nutrient analysis program and calculated the dietary myo‐inositol intake for NBDPS participants. We appreciate his collaboration and his support.
Funding: This study was supported by a cooperative agreement from the Centers for Disease Control and Prevention. This project was supported through Centers for Disease Control and Prevention (CDC) cooperative agreements under PA #96043, PA #02081, FOA #DD09‐001, FOA #DD13‐003, NOFO #DD18‐001 and NOFO #DD23‐001 to the Centers for Birth Defects Research and Prevention participating in the National Birth Defects Prevention Study (NBDPS) and/or the Birth Defects Study To Evaluate Pregnancy exposures (BD‐STEPS). This work was supported by grant no. DK56350 from the Nutrition Epidemiology Core of the University of North Carolina Clinical Nutrition Research Center. Coding of drug information in the National Birth Defects Prevention Study used the Slone Drug Dictionary under licence from the Slone Epidemiology Center of Boston University.
Main findings have been presented at the 62nd Annual Meeting of the Society for Birth Defects Research and Prevention, British Columbia, Canada, June 25–29, 2022, and the 48th International Clearinghouse for Birth Defects Surveillance and Research Annual Meeting, Sep 18–21, 2022, Bologna BO, Italia.
Data Availability Statement
The data that support the findings of this study are available from Centers for Disease Control and Prevention. Restrictions apply to the availability of these data, which were used under licence for this study. Data are available from the author(s) with the permission of Centers for Disease Control and Prevention.
References
- 1. Reller M. D., Strickland M. J., Riehle‐Colarusso T., Mahle W. T., and Correa A., “Prevalence of Congenital Heart Defects in Metropolitan Atlanta, 1998‐2005,” Journal of Pediatrics 153 (2008): 807–813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Canfield M., Honein M., Yuskiv N., et al., “National Estimates and Race/Ethnic‐Specific Variation of Selected Birth Defects in the United States, 1999‐2001,” Birth Defects Research. Part A, Clinical and Molecular Teratology 76 (2006): 747–756. [DOI] [PubMed] [Google Scholar]
- 3. van der Linde D., Konings E. E., Slager M. A., et al., “Birth Prevalence of Congenital Heart Disease Worldwide: A Systematic Review and Meta‐Analysis,” Journal of the American College of Cardiology 58 (2011): 2241–2247. [DOI] [PubMed] [Google Scholar]
- 4. Liu Y., Chen S., Zühlke L., et al., “Global Birth Prevalence of Congenital Heart Defects 1970‐2017: Updated Systematic Review and Meta‐Analysis of 260 Studies,” International Journal of Epidemiology 48 (2019): 455–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. UNICEF , How Many Babies Are Born a Year? (UNICEF, 2023). [Google Scholar]
- 6. O'Leary J. M., Ferranti S., Siddiqi O., Landzberg M., and Opotowsky A., “Adult and Pediatric Congenital Heart Disease Hospitalizations in the United States,” Journal of the American College of Cardiology 61 (2013): E541. [Google Scholar]
- 7. Simeone R. M., Oster M. E., Cassell C. H., Armour B. S., Gray D. T., and Honein M. A., “Pediatric Inpatient Hospital Resource Use for Congenital Heart Defects,” Birth Defects Research. Part A, Clinical and Molecular Teratology 100 (2014): 934–943. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jaiyesimi O. and Ruberu D. K., “Noncardiac Defects in Children With Congenital Heart Disease,” Annals of Saudi Medicine 14 (1994): 183–186. [DOI] [PubMed] [Google Scholar]
- 9. Massin M. M., Astadicko I., and Dessy H., “Noncardiac Comorbidities of Congenital Heart Disease in Children,” Acta Paediatrica 96 (2007): 753–755. [DOI] [PubMed] [Google Scholar]
- 10. Im Y. M., Lee S., Yun T. J., and Choi J. Y., “School‐Related Adjustment in Children and Adolescents With Chd,” Cardiology in the Young 27 (2017): 1349–1355. [DOI] [PubMed] [Google Scholar]
- 11. Farr S. L., Downing K. F., Riehle‐Colarusso T., and Abarbanell G., “Functional Limitations and Educational Needs Among Children and Adolescents With Heart Disease,” Congenital Heart Disease 13 (2018): 633–639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Virani S. S., Alonso A., Benjamin E. J., et al., “Heart Disease and Stroke Statistics‐2020 Update: A Report From the American Heart Association,” Circulation 141 (2020): e139–e596. [DOI] [PubMed] [Google Scholar]
- 13. van der Bom T., Zomer A. C., Zwinderman A. H., Meijboom F. J., Bouma B. J., and Mulder B. J., “The Changing Epidemiology of Congenital Heart Disease,” Nature Reviews. Cardiology 8 (2011): 50–60. [DOI] [PubMed] [Google Scholar]
- 14. Centers for disease control and prevention , “Data and Statistics on Congenital Heart Defects” (2022).
- 15. Mullen M., Zhang A., Lui G. K., Romfh A. W., Rhee J. W., and Wu J. C., “Race and Genetics in Congenital Heart Disease: Application of Ipscs, Omics, and Machine Learning Technologies,” Frontiers in Cardiovascular Medicine 8 (2021): 635280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Russell M. W., Chung W. K., Kaltman J. R., and Miller T. A., “Advances in the Understanding of the Genetic Determinants of Congenital Heart Disease and Their Impact on Clinical Outcomes,” Journal of the American Heart Association 7 (2018): 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Chen H., Zhang Y., Wang D., et al., “Periconception Red Blood Cell Folate and Offspring Congenital Heart Disease: Nested Case‐Control and Mendelian Randomization Studies,” Annals of Internal Medicine 175 (2022): 1212–1220. [DOI] [PubMed] [Google Scholar]
- 18. Obermann‐Borst S. A., Vujkovic M., de Vries J. H., et al., “A Maternal Dietary Pattern Characterised by Fish and Seafood in Association With the Risk of Congenital Heart Defects in the Offspring,” BJOG: An International Journal of Obstetrics and Gynaecology 118 (2011): 1205–1215. [DOI] [PubMed] [Google Scholar]
- 19. Wentzel P., “Can We Prevent Diabetic Birth Defects With Micronutrients?,” Diabetes, Obesity & Metabolism 11 (2009): 770–778. [DOI] [PubMed] [Google Scholar]
- 20. R. S. Clements, Jr. and Darnell B., “Myo‐Inositol Content of Common Foods: Development of a High‐Myo‐Inositol Diet,” American Journal of Clinical Nutrition 33 (1980): 1954–1967. [DOI] [PubMed] [Google Scholar]
- 21. Greene N. D., Leung K. Y., and Copp A. J., “Inositol, Neural Tube Closure and the Prevention of Neural Tube Defects,” Birth Defects Research 109 (2017): 68–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Cavalli P. and Ronda E., “Myoinositol: The Bridge (Ponti) to Reach a Healthy Pregnancy,” International Journal of Endocrinology 2017 (2017): 5846286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Greene N. D., Leung K. Y., Gay V., et al., “Inositol for the Prevention of Neural Tube Defects: A Pilot Randomised Controlled Trial,” British Journal of Nutrition 115 (2016): 974–983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Herion N. J., Salbaum J. M., and Kappen C., “Traffic Jam in the Primitive Streak: The Role of Defective Mesoderm Migration in Birth Defects,” Birth Defects Research. Part A, Clinical and Molecular Teratology 100 (2014): 608–622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Boot M. J., Steegers‐Theunissen R. P., Poelmann R. E., Van Iperen L., Lindemans J., and Gittenberger‐de Groot A. C., “Folic Acid and Homocysteine Affect Neural Crest and Neuroepithelial Cell Outgrowth and Differentiation In Vitro,” Developmental Dynamics 227 (2003): 301–308. [DOI] [PubMed] [Google Scholar]
- 26. Keyte A. and Hutson M. R., “The Neural Crest in Cardiac Congenital Anomalies,” Differentiation 84 (2012): 25–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Hill M. A., “Timeline Human Development,” 2020;2024.
- 28. Gessert S. and Kuhl M., “The Multiple Phases and Faces of Wnt Signaling During Cardiac Differentiation and Development,” Circulation Research 107 (2010): 186–199. [DOI] [PubMed] [Google Scholar]
- 29. Han M., Serrano M. C., Lastra‐Vicente R., et al., “Folate Rescues Lithium‐, Homocysteine‐ and wnt3a‐Induced Vertebrate Cardiac Anomalies,” Disease Models & Mechanisms 2 (2009): 467–478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Linask K. K. and Huhta J., “Folate Protection From Congenital Heart Defects Linked With Canonical Wnt Signaling and Epigenetics,” Current Opinion in Pediatrics 22 (2010): 561–566. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Eldar‐Finkelman H. and Martinez A., “Gsk‐3 Inhibitors: Preclinical and Clinical Focus on Cns,” Frontiers in Molecular Neuroscience 4 (2011): 32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Zhuo J. M., Portugal G. S., Kruger W. D., Wang H., Gould T. J., and Pratico D., “Diet‐Induced Hyperhomocysteinemia Increases Amyloid‐Beta Formation and Deposition in a Mouse Model of Alzheimer's Disease,” Current Alzheimer Research 7 (2010): 140–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ilouz R., Kaidanovich O., Gurwitz D., and Eldar‐Finkelman H., “Inhibition of Glycogen Synthase Kinase‐3beta by Bivalent Zinc Ions: Insight Into the Insulin‐Mimetic Action of Zinc,” Biochemical and Biophysical Research Communications 295 (2002): 102–106. [DOI] [PubMed] [Google Scholar]
- 34. Zhao J., Han J., Jiang J., et al., “The Downregulation of Wnt/Beta‐Catenin Signaling Pathway Is Associated With Zinc Deficiency‐Induced Proliferative Deficit of c17.2 Neural Stem Cells,” Brain Research 1615 (2015): 61–70. [DOI] [PubMed] [Google Scholar]
- 35. Giles J. J. and Bannigan J. G., “Teratogenic and Developmental Effects of Lithium,” Current Pharmaceutical Design 12 (2006): 1531–1541. [DOI] [PubMed] [Google Scholar]
- 36. Tian Y., Cohen E. D., and Morrisey E. E., “The Importance of Wnt Signaling in Cardiovascular Development,” Pediatric Cardiology 31 (2010): 342–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Hedgepeth C. M., Conrad L. J., Zhang J., Huang H. C., Lee V. M., and Klein P. S., “Activation of the Wnt Signaling Pathway: A Molecular Mechanism for Lithium Action,” Developmental Biology 185 (1997): 82–91. [DOI] [PubMed] [Google Scholar]
- 38. Qureshi W. M., Latif M. L., Parker T. L., and Pratten M. K., “Lithium Carbonate Teratogenic Effects in Chick Cardiomyocyte Micromass System and Mouse Embryonic Stem Cell Derived Cardiomyocyte–Possible Protective Role of Myo‐Inositol,” Reproductive Toxicology 46 (2014): 106–114. [DOI] [PubMed] [Google Scholar]
- 39. Reefhuis J., Gilboa S. M., Anderka M., et al., “The National Birth Defects Prevention Study: A Review of the Methods,” Birth Defects Research. Part A, Clinical and Molecular Teratology 103 (2015): 656–669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Rasmussen S. A., Olney R. S., Holmes L. B., Lin A. E., Keppler‐Noreuil K. M., and Moore C. A., “Guidelines for Case Classification for the National Birth Defects Prevention Study,” Birth Defects Research Part A: Clinical and Molecular Teratology 67 (2003): 193–201. [DOI] [PubMed] [Google Scholar]
- 41. Botto L. D., Lin A. E., Riehle‐Colarusso T., Malik S., and Correa A., “Seeking Causes: Classifying and Evaluating Congenital Heart Defects in Etiologic Studies,” Birth Defects Research. Part A, Clinical and Molecular Teratology 79 (2007): 714–727. [DOI] [PubMed] [Google Scholar]
- 42. Willett W. C., Reynolds R. D., Cottrell‐Hoehner S., Sampson L., and Browne M. L., “Validation of a Semi‐Quantitative Food Frequency Questionnaire: Comparison With a 1‐Year Diet Record,” Journal of the American Dietetic Association 87 (1987): 43–47. [PubMed] [Google Scholar]
- 43. Willett W. C., Sampson L., Stampfer M. J., et al., “Reproducibility and Validity of a Semiquantitative Food Frequency Questionnaire,” American Journal of Epidemiology 122 (1985): 51–65. [DOI] [PubMed] [Google Scholar]
- 44. Cavalli P. and Copp A. J., “Inositol and Folate Resistant Neural Tube Defects,” Journal of Medical Genetics 39 (2002): E5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Dell'Edera D., Sarlo F., Allegretti A., et al., “Prevention of Neural Tube Defects and Maternal Gestational Diabetes Through the Inositol Supplementation: Preliminary Results,” European Review for Medical and Pharmacological Sciences 21 (2017): 3305–3311. [PubMed] [Google Scholar]
- 46. Cavalli P., Tonni G., Grosso E., and Poggiani C., “Effects of Inositol Supplementation in a Cohort of Mothers at Risk of Producing an Ntd Pregnancy,” Birth Defects Research. Part A, Clinical and Molecular Teratology 91 (2011): 962–965. [DOI] [PubMed] [Google Scholar]
- 47. Greenland S., Pearl J., and Robins J. M., “Causal Diagrams for Epidemiologic Research,” Epidemiology 10 (1999): 37–48. [PubMed] [Google Scholar]
- 48. Textor J., van der Zander B., Gilthorpe M. S., Liskiewicz M., and Ellison G. T., “Robust Causal Inference Using Directed Acyclic Graphs: The r Package ‘dagitty’,” International Journal of Epidemiology 45 (2016): 1887–1894. [DOI] [PubMed] [Google Scholar]
- 49. NIH , Folate (NIH, 2024). [Google Scholar]
- 50. Bodnar L. M. and Siega‐Riz A. M., “A Diet Quality Index for Pregnancy Detects Variation in Diet and Differences by Sociodemographic Factors,” Public Health Nutrition 5 (2002): 801–809. [DOI] [PubMed] [Google Scholar]
- 51. Lee P. H. and Burstyn I., “Identification of Confounder in Epidemiologic Data Contaminated by Measurement Error in Covariates,” BMC Medical Research Methodology 16 (2016): 54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Lee P. H., “Is a Cutoff of 10% Appropriate for the Change‐In‐Estimate Criterion of Confounder Identification?,” Journal of Epidemiology 24 (2014): 161–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Abellana Sangra R. and Farran Codina A., “The Identification, Impact and Management of Missing Values and Outlier Data in Nutritional Epidemiology,” Nutrición Hospitalaria 31 Suppl 3 (2015): 189–195. [DOI] [PubMed] [Google Scholar]
- 54. Razzaghi H., Tinker S. C., Herring A. H., Howards P. P., Waller D. K., and Johnson C. Y., “Impact of Missing Data for Body Mass Index in an Epidemiologic Study,” Maternal and Child Health Journal 20 (2016): 1497–1505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Little R. J., D'Agostino R., Cohen M. L., et al., “The Prevention and Treatment of Missing Data in Clinical Trials,” New England Journal of Medicine 367 (2012): 1355–1360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Angeles UoCL , “Multiple Imputation in Sas Part 1,” (2024).
- 57. Sas , 2006.
- 58. Feng Y., Wang S., Chen R., Tong X., Wu Z., and Mo X., “Maternal Folic Acid Supplementation and the Risk of Congenital Heart Defects in Offspring: A Meta‐Analysis of Epidemiological Observational Studies,” Scientific Reports 5 (2015): 8506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Mao B., Qiu J., Zhao N., et al., “Maternal Folic Acid Supplementation and Dietary Folate Intake and Congenital Heart Defects,” PLoS One 12 (2017): e0187996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60. Botto L. D., Mulinare J., and Erickson J. D., “Do Multivitamin or Folic Acid Supplements Reduce the Risk for Congenital Heart Defects? Evidence and Gaps,” American Journal of Medical Genetics Part A 121a (2003): 95–101. [DOI] [PubMed] [Google Scholar]
- 61. Moore K. J., Carmichael S. L., Forestieri N. E., et al., “Maternal Diet as a Risk Factor for Primary Congenital Glaucoma and Defects of the Anterior Segment of the Eye in the National Birth Defects Prevention Study,” Birth Defects Research 112 (2020): 503–514. [DOI] [PubMed] [Google Scholar]
- 62. Stingone J. A., Luben T. J., Sheridan S. C., et al., “Associations Between Fine Particulate Matter, Extreme Heat Events, and Congenital Heart Defects,” Environmental Epidemiology (Philadelphia, Pa.) 3 (2019): e071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63. Shinde M. U., Vuong A. M., Brender J. D., et al., “Prenatal Exposure to Nitrosatable Drugs, Vitamin c, and Risk of Selected Birth Defects,” Birth Defects Research. Part A, Clinical and Molecular Teratology 97 (2013): 515–531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 64. Patel J., Nembhard W. N., Politis M. D., et al., “Maternal Occupational Exposure to Polycyclic Aromatic Hydrocarbons and the Risk of Isolated Congenital Heart Defects Among Offspring,” Environmental Research 186 (2020): 109550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65. Nembhard W. N., Tang X., Hu Z., MacLeod S., Stowe Z., and Webber D., “Maternal and Infant Genetic Variants, Maternal Periconceptional Use of Selective Serotonin Reuptake Inhibitors, and Risk of Congenital Heart Defects in Offspring: Population Based Study,” BMJ 356 (2017): j832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66. Choi G., Stingone J. A., Desrosiers T. A., et al., “Maternal Exposure to Outdoor Air Pollution and Congenital Limb Deficiencies in the National Birth Defects Prevention Study,” Environmental Research 179 (2019): 108716. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67. Willett W. C., Sampson L., Browne M. L., et al., “The Use of a Self‐Administered Questionnaire to Assess Diet Four Years in the Past,” American Journal of Epidemiology 127 (1988): 188–199. [DOI] [PubMed] [Google Scholar]
- 68. Stingone J. A., Luben T. J., Carmichael S. L., et al., “Maternal Exposure to Nitrogen Dioxide, Intake of Methyl Nutrients, and Congenital Heart Defects in Offspring,” American Journal of Epidemiology 186 (2017): 719–729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69. Gilboa S. M., Lee K. A., Cogswell M. E., et al., “Maternal Intake of Vitamin e and Birth Defects, National Birth Defects Prevention Study, 1997 to 2005,” Birth Defects Research. Part A, Clinical and Molecular Teratology 100 (2014): 647–657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70. Morgenstern H. and Thomas D., “Principles of Study Design in Environmental Epidemiology,” Environmental Health Perspectives 101, no. Suppl 4 (1993): 23–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71. Shaw G. M., Carmichael S. L., Yang W., and Schaffer D. M., “Periconceptional Dietary Intake of Myo‐Inositol and Neural Tube Defects in Offspring,” Birth Defects Research. Part A, Clinical and Molecular Teratology 73 (2005): 184–187. [DOI] [PubMed] [Google Scholar]
- 72. Cogram P., Hynes A., Dunlevy L. P., Greene N. D., and Copp A. J., “Specific Isoforms of Protein Kinase c Are Essential for Prevention of Folate‐Resistant Neural Tube Defects by Inositol,” Human Molecular Genetics 13 (2004): 7–14. [DOI] [PubMed] [Google Scholar]
- 73. Greene N. D. and Copp A. J., “Inositol Prevents Folate‐Resistant Neural Tube Defects in the Mouse,” Nature Medicine 3 (1997): 60–66. [DOI] [PubMed] [Google Scholar]
- 74. Tzahor E., “Wnt/Beta‐Catenin Signaling and Cardiogenesis: Timing Does Matter,” Developmental Cell 13 (2007): 10–13. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Flow diagram of study participants, National Birth Defects Prevention Study, 1997–2011
Table S1: Distribution of maternal characteristics among participants with different intake of supplements containing myo‐inositol, National Birth Defects Prevention Study, 1997–2011.
Table S2: Distribution of maternal characteristics among participants with different intake of dietary myo‐inositol, National Birth Defects Prevention Study, 1997–2011.
Table S3: Distribution of maternal characteristics among participants with different intake of total myo‐inositol from diet and supplements, National Birth Defects Prevention Study, 1997–2011.
Table S4: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between total maternal periconceptionala myo‐inositol intake and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S5: Distribution of congenital heart defect (CHD) cases and controls across quartiles of periconceptionala maternal dietary myo‐inositol intake, National Birth Defects Prevention Study, 1997–2011.
Table S6: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptionala total maternal myo‐inositol intake by quartile and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐1 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐2 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐3 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S7:‐4 Odds ratios (ORs) and 95% confidence intervals (CIs) of congenital heart defects (CHD) according to cross‐classification of periconceptionala maternal myo‐inositol intake and diet quality index, from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Table S8: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between total maternal periconceptionala myo‐inositol intake and congenital heart defects (CHDs) stratified by folic acid/folate intake from food and supplements, National Birth Defects Prevention Study, 1997–2011.
Table S9: Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between periconceptionala maternal dietary myo‐inositol intakeb and congenital heart defects (CHD) from logistic regression analyses, National Birth Defects Prevention Study, 1997–2011.
Supplemental Methods
Data Availability Statement
The data that support the findings of this study are available from Centers for Disease Control and Prevention. Restrictions apply to the availability of these data, which were used under licence for this study. Data are available from the author(s) with the permission of Centers for Disease Control and Prevention.