Maternal, paternal, and neonatal risk factors for neural tube defects: A systematic review and meta-analysis

Journal Pre-proof Maternal, Paternal, and Neonatal Risk Factors for Neural Tube Defects: A Systematic Review and Meta-Analysis Shanshan Jia, Xiaowei Wei, Ling Ma, Yanfu Wang, Hui Gu, Dan Liu, Wei Ma, Zhengwei Yuan

PII:

S0736-5748(19)30151-0

DOI:

https://doi.org/10.1016/j.ijdevneu.2019.09.006

Reference:

DN 2391

To appear in:

International Journal of Developmental Neuroscience

Received Date:

30 May 2019

Revised Date:

22 September 2019

Accepted Date:

24 September 2019

Please cite this article as: Jia S, Wei X, Ma L, Wang Y, Gu H, Liu D, Ma W, Yuan Z, Maternal, Paternal, and Neonatal Risk Factors for Neural Tube Defects: A Systematic Review and Meta-Analysis, International Journal of Developmental Neuroscience (2019), doi: https://doi.org/10.1016/j.ijdevneu.2019.09.006

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Title: Maternal, Paternal, and Neonatal Risk Factors for Neural Tube Defects: A Systematic Review and Meta-Analysis Shanshan Jia1, Xiaowei Wei1, Ling Ma1,2, Yanfu Wang1, Hui Gu1, Dan Liu1, Wei Ma1, Zhengwei Yuan1*. 1

Key Laboratory of Health Ministry for Congenital Malformation, Shengjing Hospital, China Medical University, Shenyang, PR China 2 Department of Pathophysiology, Basic Science College, China Medical University, Shenyang, PR China

-p

ro of

Corresponding author: Zhengwei Yuan, Key Laboratory of Health Ministry for Congenital Malformation, Shengjing Hospital, China Medical University. No. 36, Sanhao Street, Heping District, Shenyang, 110004, PR China. Tel: +86 24 23929903 Fax:+86 24 23929903 E-mail address: [email protected]

re

Highlights:

45 studies are included in the systematic review and meta-analysis.



12 maternal, paternal, and neonatal potential risk factors on neural tube defects.



Contributing to prenatal care education and public health policy.

ur

na

lP



Abstract

Jo

Objective: Neural tube defects (NTDs) are severe congenital anomalies. The etiology of NTDs is not fully known, and studies on the potential risk factors of NTDs present inconsistent results. Thus, we conducted a systematic review and meta-analysis to investigate the maternal, paternal, and neonatal risk factors for NTDs. Study Design:

We systematically reviewed relative original studies published

1

through October 6, 2018 available in Pubmed, Embase and the Cochrane Library without restrictions for language. The selected studies measured maternal, paternal, and neonatal risk factors and examined their associations with NTDs. A metaanalysis, including subgroup analysis and sensitivity analysis, was conducted to estimate the pooled effect measures. Two reviewers independently extracted data using a predesigned data collection form.

ro of

Results: Forty-five studies were eligible for inclusion in the meta-analysis, and

twelve potential risk factors were analyzed. The factors that were associated with NTDs risk included stressful life events [odds ratio (OR),1.61; 95% confidence

-p

interval (CI), 1.242.08; p<0.001; I2 = 59.2%], low maternal education level (OR,

re

1.42; 95% CI, 1.191.70; p<0.001; I2 = 47.7%), pregestational diabetes (OR, 2.24; 95% CI, 1.214.12; p<0.010; I2 =56.3%), low paternal age (OR, 1.41; 95% CI,

lP

1.101.81; p=0.007; I2 =0.0%), low birth weight (OR, 5.53; 95% CI, 1.9515.70;

na

p=0.001; I2 = 98.5%), and neonatal female gender (OR, 1.54; 95% CI, 1.102.14; p=0.012; I2 = 67.8%).

ur

Conclusion: Stressful life events, pregestational diabetes, low birth weight, and neonatal female gender are risk factors associated with NTDs. Low maternal

Jo

education level and low paternal age are factors that are moderately associated with NTDs. Further cohort studies are required to verify the factors associated with NTDs and control the risk of this severe birth defect. Keywords: neural tube defects; risk factors; systematic review; meta-analysis

Introduction 2

Neural tube defects (NTDs) are one of the most common categories of severe congenital anomalies without effective treatments, alongside congenital heart defects and musculoskeletal defects(1). NTDs have a worldwide prevalence of 0.66 cases per 1,000 births(2). Neural tube development occurs in the embryo during week 3–4 after conception, and primary neurulation is responsible for generating the entire neural tube(3). The etiology of NTDs is probably complex and obscure, but both

ro of

genetic and environmental factors are believed to be involved.

A number of population-based studies have focused on specific environmental

exposures as possible risk factors for NTDs. Several studies have proposed that some

-p

maternal (e.g., low dietary folate intake and history of NTDs), paternal (e.g.,

re

occupational exposure), and neonatal (e.g., female gender) factors may increase the risk of NTDs(4,5). However, the overall conclusions of these studies are inconsistent. consumption during early pregnancy

lP

Some studies report that maternal tea

na

increases the risk of NTDs(6-8), whereas another study does not find an association(9). Petrova and Vieira propose that maternal age is correlated with the risk

ur

of infant NTDs(10,11). There is no consensus regarding the effect of paternal age on NTD risk; it remains to be confirmed whether younger or older fathers increase the

Jo

risk of producing an infant with NTDs, or whether this risk differs depending on the type of NTDs.

Although the association between maternal factors and NTDs risk has been well studied, the role of paternal and neonatal factors has received less attention. Previous studies of paternal and neonatal factors have been based primarily on small samples,

3

and most of the meta-analyses explore only a few potential risk factors. Here, we perform a meta-analysis of all relevant available data that were not evaluated in previous studies to determine whether maternal, paternal, and neonatal factors are associated with increased risk for NTDs. Further studies are required to validate the risk factors identified in this study, which can then be used to inform preventative and prophylactic strategies to reduce the incidence of NTDs.

ro of

Materials and Methods STUDY IDENTIFICATION AND SELECTION

The study was conducted according to the Preferred Reporting Items for Systematic

-p

Reviews and Meta-Analyses (PRISMA) criteria(12)(Supplementary File S1).

re

PubMed, Embase, and Cochrane Library databases were searched using a combination of the following keywords and Medical Subject Headings (MeSH):

lP

“maternal” or “pregnancy” or “parental” or “paternal” or “neonatal” or “infant,

na

newborn” or “neural tube defects” and “NTDs” or “spina bifida” or “SB” or “anencephaly” or “encephalocele”, in combination with “cohort study” or “case-

ur

control study”. The snowballing technique was used to pursue references of references and assessed reports of studies not found in the database searches. The last

Jo

search was performed to make sure no related studies were missed. Search strategies for Pubmed, Embase and the Cochrane Library were presented in Supplementary File S2, respectively. All studies were initially screened by examining the titles and abstracts. Two authors (Ling Ma and Yanfu Wang) assessed the full-text articles of candidate studies for inclusion eligibility. Disagreements were discussed until

4

reaching consensus. The following inclusion criteria were applied for study inclusion in our meta-analysis: (1) study evaluated the associations between maternal, paternal, and neonatal factors and risk of NTDs; (2) study was based on a cohort or case-control study design; and (3) study reported odds ratios (ORs) or relative risks (RRs) and corresponding 95% confidence intervals (CIs), or other

available statistical data to calculate odds ratios

ro of

(ORs) or relative risks (RRs) and corresponding 95% confidence intervals (CIs). The following exclusion criteria were used: (1) study did not eliminate the impact of genetic causes on NTDs risk, or (2) study overlapped with another study.

-p

DATA EXTRACTION

re

The following data from the included studies were collected by each of two authors: first author’s name, publication year, country or region where the study was

lP

conducted, numbers of cases and controls, effect size, study design, study quality, and

na

risk factors.

We only included risk factors that were evaluated in two or more studies. For each

ur

risk factor, we extracted the OR/RR and the 95% CI. We assumed similarity between OR and RR because NTDs are rare events(13) . For studies without adjusted

Jo

estimates, we used crude estimates. Data extracted from each study without crude estimates were arranged in 22 tables, and the results were combined to produce pooled unadjusted ORs with 95% CIs according to the Mantel-Haenszel method using random-effects models. Analyses were conducted using the Review Manager (Revman) version 5.3 statistical package.

5

QUALITY ASSESSMENT The methodological quality of each included study was independently assessed by two authors using the Newcastle-Ottawa Scale (NOS) (14) and Non-Randomised Studies of Interventions (ROBINS-I) (15). The quality score of NOS was evaluated based on the following three categories: group selection (four points), comparability between groups (two points), and exposure assessment in case-control studies or

ro of

outcome assessment in cohort studies (three points). A study could be awarded a

maximum of one score for each numbered item, so the maximum total score was nine points. Studies were considered to have good quality if they were scored above the

-p

median (> five points). Non-Randomised Studies of Interventions (ROBINS-I) are

re

scored on seven domains: confounding, participant selection, classification of interventions, deviation from intended intervention, missing data, measurement of

lP

outcomes, and reporting bias. Disagreements between the two authors were discussed

reached.

na

until reaching consensus. The third investigator was consulted if consensus was not

ur

STATISTICAL ANALYSIS

A separate meta-analysis was conducted for each risk factor, and the pooled OR was

Jo

calculated using the fixed-effects model (I2<50%) or the random-effects model (I2≥50%). Heterogeneity among studies was assessed using the chi-square and I2 tests. Data were considered statistically heterogeneous if I2>50% and P-heterogeneity was < 0.1. Publication bias was assessed by conducting tests for funnel plot asymmetry using Begg’s test(16) and Egger’s test(17) (p<0.05). Subgroup analysis was performed

6

to assess the impact of the other variables on study outcomes. Sensitivity analysis was performed to identify heterogeneous studies. All statistical analyses were conducted using Review Manager (RevMan) version 5.3 and Stata version 12.0 statistical packages.

Results STUDY SELECTION AND CHARACTERISTICS

ro of

We assessed a total of 3,426 studies identified in the database searches (2,607 in

PubMed, 13 in Cochrane Library, and 806 in Embase), and an additional 7 studies

(Supplementary File S3) were culled from other sources. There were 759 duplicate

-p

records in the databases searches. After examining the titles and abstracts, 2,623

re

studies were excluded for irrelevance to maternal, paternal, or neonatal factors and NTD risk. After reading the full-text articles, 6 studies (Supplementary File S4) did

lP

not satisfy the inclusion criteria. Finally, 45 studies were selected for inclusion. The

na

meta-analysis evaluated 12 potential risk factors (8 maternal, 1 paternal, and 3 neonatal). The selection process is presented in Figure 1, and the characteristics of

ur

each included study are presented in Tables 13(3,6-9,18-57). QUALITY OF THE INCLUDED STUDIES

Jo

The Newcastle-Ottawa Scale (NOS) was used to evaluate the studies. Each observational study can be awarded a maximum of one star for each numbered item in the selection and exposure/outcome categories. A maximum of two stars can be given for comparability. The methodological quality scores of the included studies ranged from five to eight points. The appraisal details are presented in Supplementary Tables

7

S1 and S2. Thirty-six of the 45 studies were considered high quality with scores greater than five points. The follow-up duration of cohort studies was not adequate to record spontaneous abortion, stillbirth, or termination of pregnancy due to prenatal diagnosis; therefore, the quality scores of neonatal studies were lower with respect to the outcome categories. A few studies in our analysis only provided crude estimates; for these, only one star was given for comparability categories. Supplementary File S5

ro of

presented potential confounders those adjusted OR and 95% CI were calculated in 45 included studies. For these reasons, 20% of the included studies scored poorly in our

meta-analysis. We also used the ROBINS-I to evaluate these non-randomised studies

-p

(Supplementary Table S3).

re

Table 4 presents the following data for the meta-analysis: maternal, paternal, and neonatal risk factors included in the analysis; summary effect estimates [odds

lP

ratio/relative risks (ORs/RRs)] with 95% CI; and the P-value for each risk factor. The

na

following maternal risk factors were evaluated in the meta-analysis: tea drinking (≥1 cup/day) (OR, 1.69 and 95% CI, 0.943.06); history of maternal abortion (OR, 1.31

ur

and 95% CI, 0.951.80); history of spontaneous abortion (OR, 0.93 and 95% CI, 0.551.56); stressful life events (such as maternal antenatal bereavement, major

Jo

injuries, relationship difficulties, legal/financial problems, and job losses) (OR, 1.61 and 95% CI, 1.242.08); low maternal education level (without high school diploma) (OR, 1.42 and 95% CI, 1.191.70); nausea and vomiting during the first trimester of pregnancy (OR, 1.48 and 95% CI, 0.733.00); pregestational diabetes (OR, 2.24 and 95% CI, 1.214.12); and hazardous waste sites (distance of site from residence ranges

8

from 1.6 to 3 km) (OR, 1.10 and 95% CI, 0.891.37). The following paternal and neonatal risk factors were evaluated in the meta-analysis: low paternal age (<20 vs. 2529) (OR, 1.31 and 95% CI, 1.061.62); low birth weight (OR, 5.53 and 95% CI, 1.9515.7); neonatal female gender (OR, 1.54 and 95% CI, 1.102.14); and birth season (OR, 1.11 and 95% CI, 0.991.25). PUBLICATION BIAS

ro of

Publication bias was assessed with Egger’s regression asymmetry test, Begg’s rank

correlation test, and Begg’s funnel plot using Stata version 12.0. Publication bias was assessed for the factors examined in the pooled analysis of more than four studies

-p

(Fig. 2). Significant publication bias (p<0.05) was found only for history of maternal

re

abortion (Begg’s test, p=0.107; Egger’s test, p=0.045). SUBGROUP ANALYSIS

lP

The studies included in our meta-analysis categorized the father’s age into the

na

following standard 5-year age groups: <20, 20-24, 25-29, 30-34, 35-49, and ≥50 years old. The results were presented for each specific NTD group (spina bifida and

ur

anencephaly). We performed subgroup analysis to assess associations of the paternal age subsets with types of NTDs. Advanced paternal age (>45) was not associated with

Jo

risk of NTDs or type of NTDs (spina bifida or anencephaly) in our analysis. By contrast, low paternal age (<20) was associated with a statistically significant risk of spina bifida (OR, 1.41 and 95% CI, 1.10-1.81) (Fig. 3). In our meta-analysis, the relationship between paternal age and spina bifida/anencephaly is based on 7,864,762 subjects from birth registrations data. The included studies in our meta-analysis

9

indicated the subphenotypes of the isolated neural tube defects, classified according to The International Classification of Diseases. There were 8 cohort studies and 37 case-control studies included in our systematic review and meta-analysis. Since the included studies on low birth weight were cohort studies, we performed subgroup analysis on the risk factors of low paternal age and pregestational diabetes. Supplementary Figures 3 and 4 presented the pooled effect

ro of

measures of subgroup analysis on different types of study respectively. SENSITIVITY ANALYSIS

The “leave one out” sensitivity analysis (Fig. 4) indicated that two studies(3,40) on

-p

stressful maternal life events were key contributors to heterogeneity between studies.

re

The study heterogeneity (I2 = 59.2%) was reduced to 9.4% after excluding these two

lP

studies in Supplementary Figures S1 and S2. The pooled effect measures (OR, 1.57; 95% CI, 1.341.84; p<0.001) was presented in forest plot.

na

Discussion

Identifying patients who have a high risk of infant NTDs is crucial for early detection

ur

and intervention, and would benefit both clinical practice and public health. To our knowledge, this is the first systematic review and meta-analysis to determine relevant

Jo

maternal, paternal, and neonatal factors that increase the risk of NTDs. Our metaanalysis indicated that the following factors were associated with increased risk of NTDs: stressful life events, pregestational diabetes, low birth weight, and neonatal female gender. Other significant factors with lower strengths of association (risk estimates < 1.5) were low maternal education level and low paternal age.

10

The studies of Ingstrup(3) and Suarez(40) consider extremely broad windows for stressful life events starting at one year before conception. Other studies consider time frames between three months before conception and the first trimester, because many of the included studies evaluate the occurrence of NTDs during critical developmental periods in early pregnancy. Ingstrup did not report any information about women having abortions or stillbirths due to NTDs. Data from the EUROCAT trial

ro of

(Anomalies EESoC, 2012 - 2016) indicate that approximately 79.89% of children

with NTDs die before birth due to natural death or elective termination of pregnancy after prenatal ultrasound scanning for congenital malformations(58). The results of

-p

our sensitivity analysis do not significantly change the pooled estimates, which

re

strengthens the confidence that can be place in our results. Maternal stressful life events may contribute to maternal physiological and lifestyle changes. Elevated

lP

maternal corticotrophin-releasing hormone(59), corticosteroid levels(60), and

na

homocysteine levels(61), during pregnancy may be the mechanisms by which stressful life events might influence prenatal development. Other possible factors

ur

linking stress with NTDs include negative coping behaviors that lead to hazardous exposures and may contribute to NTDs(26), such as cigarette smoking(62), poor

Jo

sleep(63), illicit drugs such as cannabis(64), and reduced nutrient intake(65). Results of subgroup analysis on the risk factor of pregestational diabetes showed different types of study may cause heterogeneity (I2=56.3%). The heterogeneity of case-control studies and cohort studies were both 0.0%. In addition, whether casecontrol studies or cohort studies indicated pregestational diabetes was a risk factor for

11

NTDs. In 2014, approximately 2.2% of births in the U.S. were to women with pregestational diabetes mellitus(66). Poor glycemic control at the time of conception and during pregnancy can cause severe complications, such as spontaneous abortion and congenital malformations(67). These facts agree with our meta-analysis, which indicates that pregestational diabetes is related to the prevalence of NTDs. Women with lower socioeconomic status, such as those who did not graduate from high

ro of

school, may be at risk for the absence of dietary folate or multivitamin

supplementation, food fortification (e.g., meat, milk, fresh vegetables, and fruits), and prenatal screening(22). After excluding these odds ratio were not adjusted for

-p

confounders, then performing pooled effect measures again, we found lower maternal

re

educational level (OR, 1.40; 95% CI, 1.10-1.84; p=0.015; I2 =46.4%) was still a potential risk factor for NTDs. Our results suggest that access to quality education

lP

may reduce the risk of NTDs.

na

Two studies (18,19) conducted in China reported that severe nausea and vomiting in the first trimester of pregnancy were associated with the occurrence of NTDs. Nausea

ur

and vomiting of pregnancy (NVP) peaks at 712 weeks, and severe NVP may be a consequence, rather than a cause, of NTDs. NVP during the first trimester of

Jo

pregnancy reduces dietary diversity and nutrient intake, and this retards intrauterine growth and can lead to low birth weight (68). Low birth weight, otherwise known as “growth restriction”, is a marker of poor maternal nutritional status that is strongly associated with increased risk of NTDs. The study of Norman(54) considered meningocele, myelomeningocele, and encephalocele were types of NTDs,

12

anencephaly, iniencephaly, and craniorachischisis were excluded from analysis. However, anencephaly, spina bifida, encephalocele, hydrocephalus and microcephalus were eligible as NTD cases in the study of Khoury(57). Then we performed pooled effect measures (OR, 2.89; 95% CI, 2.074.05; p<0.001; I2 =0.0%) on those types of NTDs excluding anencephaly, hydrocephalus and microcephalus. The study heterogeneity (I2 = 98.5%) was reduced to 0.0%. The odds ratios (ORs) of Nili 2006

ro of

and Deb 2013 studied on neonatal female gender were not adjusted for confounders. We performed pooled effect measures (OR, 1.44; 95% CI, 1.13-1.84; p=0.003; I2 =0.0%) with no heterogeneity on the studies which adjusted for confounders.

-p

Given these mutations accumulate in the sperm of older fathers, the probability of

re

NTDs may rise with increasing paternal age according to the “copy error” hypothesis proposed by Penrose (69). However, our study suggested that advanced

lP

paternal age was not associated with an increased prevalence of NTDs and its

na

common subtypes. Maternal periconceptional supplementation with dietary folate and multivitamins may explain the absence of an association between older paternal

ur

age and NTDs. By contrast, younger fathers were associated with NTD risk in our meta-analysis, which might be due to the occurrence of unplanned pregnancies in this

Jo

younger group who may not take folic acid supplementation and may engage in negative coping behaviors(52). Meng et al.(70) reported that passive smoking during pregnancy significantly increased the risk of NTDs. The present explanation for the young paternal age effect is that more unhealthy lifestyle among young couples may

13

be prevalent, such as smoking and alcohol use. Emotional support(40), food fortification(71), and preconception care(66) may reduce the prevalence of NTDs. There were some limitations in our meta-analysis. Although this meta-analysis had not been registered online, we strictly followed the steps of system evaluation. Our study enrolled observational studies that can be subject to confounding, retrospective exposure reports, and non-response of intended participants. First, the case-control

ro of

and retrospective-cohort design of the included articles inevitably leads to recall bias.

Second, several of the included studies only reported crude ORs or only provided raw data and did not adjust for other potential confounders, which might have led to

-p

overestimation and high heterogeneity of the association between risk factors and

re

NTDs. Third, although our meta-analysis included 45 studies enrolling more than 1,911,677 pregnant women (22,262 cases and 1,889,416 controls), a few studies in

lP

our analysis had small sample size and may cause bias. Finally, only published data

na

were included in our meta-analysis, and heterogeneity among the studies was detected for most of the analyzed risk factors.

ur

Conclusions

This systematic review and meta-analysis evaluated maternal, paternal, and

Jo

neonatal risk factors for NTDs. Stressful life events, pregestational diabetes, low birth weight, neonatal female gender, low maternal education level and low paternal age were associated with NTDs. Meanwhile, the neural tube defects among infants of younger fathers warrant additional research, to better understand the underlying mechanisms. Identification of these risk factors contributes to basic research and may

14

improve early prenatal care education and public health policy.

Conflict of interest The authors declare that they have no interests that might be perceived as posing conflict or bias.

Acknowledgments

Jo

ur

na

lP

re

-p

ro of

This study is supported by the National Key Research and Development Program(2016YFC1000505); the National Basic Research Program of China (973 program, No. 2013CB945402); and the National Natural Foundation of China (Grant numbers: 81671469, 81370717).

15

References

Jo

ur

na

lP

re

-p

ro of

1. Simeone RM, Feldkamp ML, Reefhuis J, Mitchell AA, Gilboa SM, Honein MA, et al. CDC Grand Rounds: Understanding the Causes of Major Birth Defects - Steps to Prevention. MMWR Morb Mortal Wkly Rep. 2015;64(39):1104-7. 2. Molloy AM, Pangilinan F, Brody LC. Genetic Risk Factors for Folate-Responsive Neural Tube Defects. Annu Rev Nutr. 2017;37:269-91. 3. Ingstrup KG, Wu CS, Olsen J, Nohr EA, Bech BH, Li J. Maternal Antenatal Bereavement and Neural Tube Defect in Live-Born Offspring: A Cohort Study. PLoS One. 2016;11(9):e0163355. 4. Agopian AJ, Tinker SC, Lupo PJ, Canfield MA, Mitchell LE, National Birth Defects Prevention S. Proportion of neural tube defects attributable to known risk factors. Birth Defects Res A Clin Mol Teratol. 2013;97(1):42-6. 5. Aguilar-Garduno C, Lacasana M, Blanco-Munoz J, Borja-Aburto VH, Garcia AM. Parental occupational exposure to organic solvents and anencephaly in Mexico. Occup Environ Med. 2010;67(1):32-7. 6. Ye R, Ren A, Zhang L, Li Z, Liu J, Pei L, et al. Tea drinking as a risk factor for neural tube defects in northern China. Epidemiology. 2011;22(4):491-6. 7. Correa A, Stolley A, Liu Y. Prenatal tea consumption and risks of anencephaly and spina bifida. Ann Epidemiol. 2000;10(7):476-7. 8. Fedrick J. Anencephalus and maternal tea drinking: evidence for a possible association. Proc R Soc Med. 1974;67(5):356-60. 9. Yazdy MM, Tinker SC, Mitchell AA, Demmer LA, Werler MM. Maternal tea consumption during early pregnancy and the risk of spina bifida. Birth Defects Res A Clin Mol Teratol. 2012;94(10):756-61. 10. Petrova JG, Vaktskjold A. The incidence of neural tube defects in Norway and the Arkhangelskaja Oblast in Russia and the association with maternal age. Acta Obstet Gynecol Scand. 2009;88(6):667-72. 11. Vieira AR, Castillo Taucher S. Maternal age and neural tube defects: evidence for a greater effect in spina bifida than in anencephaly. Rev Med Chil. 2005;133(1):62-70. 12. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. 13. Deeks J. When can odds ratios mislead? Odds ratios should be used only in casecontrol studies and logistic regression analyses. BMJ. 1998;317(7166):1155-6; author reply 6-7. 14. Wells GA, Shea B, O'Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. 15. Sterne JA, Hernan MA, Reeves BC, Savovic J, Berkman ND, Viswanathan M, Henry D, Altman DG, Ansari MT, Boutron I, Carpenter JR, Chan AW, Churchill R, Deeks JJ, Hrobjartsson A, Kirkham J, Juni P, Loke YK, Pigott, TD, Ramsay, CR, Regidor D, Rothstein HR, Sandhu L, Santaguida PL, Schunemann HJ, Shea B, Shrier 16

Jo

ur

na

lP

re

-p

ro of

I, Tugwell P, Turner L, Valentine JC, Waddington H, Waters E, Wells GA, Whiting PF, Higgins, JP. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355: i4919. 16. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50(4):1088-101. 17. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629-34. 18. Zhang Y, Li Z, Zhang L, Liu J, Jin L, Ren A. Association between severe nausea and vomiting in early pregnancy and the risk of neural tube defects in Northern China. Birth Defects Res. 2018;110(5):406-12. 19. Lu QB, Wang ZP, Gao LJ, Gong R, Sun XH, Wang M, et al. Nausea and vomiting in early pregnancy and the risk of neural tube defects: a case-control study. Sci Rep. 2015;5:7674. 20. Talebian A, Soltani B, Sehat M, Zahedi A, Noorian A, Talebian M. Incidence and Risk Factors of Neural Tube Defects in Kashan, Central Iran. Iran J Child Neurol. 2015;9(3):50-6. 21. Carmichael SL, Ma C, Tinker S, Rasmussen SA, Shaw GM, National Birth Defects Prevention, Study. Maternal stressors and social support as risks for delivering babies with structural birth defects. Paediatr Perinat Epidemiol. 2014;28(4):338-44. 22. Deb R, Arora J, Saraswathy KN, Kalla AK. Association of sociodemographic and nutritional factors with risk of neural tube defects in the North Indian population: a case-control study. Public Health Nutr. 2014;17(2):376-82. 23. Li Z, Zhang L, Li H, Ye R, Liu, J, Ren A. Maternal severe stressful life events and risk of neural tube defects among rural Chinese. Birth Defects Res A Clin Mol Teratol. 2013; 97(2): 109-14. 24. Anderka M, Mitchell AA, Louik C, Werler MM, Hernández-Diaz S, Rasmussen SA; National Birth Defects Prevention Study. Medications used to treat nausea and vomiting of pregnancy and the risk of selected birth defects. Birth Defects Res A Clin Mol Teratol. 2012; 94(1): 22-30. 25. Garne E, Loane M, Dolk H, Barisic I, Addor MC, Arriola L, et al. Spectrum of congenital anomalies in pregnancies with pregestational diabetes. Birth Defects Res A Clin Mol Teratol.2012; 94(3): 134-40. 26. De Marco P, Merello E, Calevo MG, Mascelli S, Pastorino D, Crocetti L, et al. Maternal periconceptional factors affect the risk of spina bifida-affected pregnancies: an Italian case-control study. Childs Nerv Syst.2011; 27(7): 1073-81. 27. Lu QB, Wang ZP, Gao LJ, Gong R, Sun XH, Zhao ZT. Previous abortion and the risk of neural tube defects: a case-control study. J Reprod Med. 2011;56(9-10):431-6. 28. Yin Z, Xu W, Xu C, Zhang S, Zheng Y, Wang W, et al. A population-based casecontrol study of risk factors for neural tube defects in Shenyang, China. Childs Nerv Syst.2011; 27(1): 149-154. 29. Grewal J, Carmichael SL, Song J, Shaw GM. Neural tube defects: an analysis of neighbourhood- and individual-level socio-economic characteristics. Paediatr Perinat Epidemiol. 2009;23(2):116-24. 17

Jo

ur

na

lP

re

-p

ro of

30. Carmichael SL, Shaw GM, Yang W, Abrams B, Lammer EJ. Maternal stressful life events and risks of birth defects. Epidemiology. 2007;18(3):356-61. 31. Suarez L, Brender JD, Langlois PH, Zhan FB, Moody K. Maternal exposures to hazardous waste sites and industrial facilities and risk of neural tube defects in offspring. Ann Epidemiol. 2007;17(10):772-7. 32. Czeizel AE, Puho E, Acs N, Banhidy F. Inverse association between severe nausea and vomiting in pregnancy and some congenital abnormalities. Am J Med Genet A. 2006;140(5):453-62. 33. Macintosh MC, Fleming KM, Bailey JA, Golightly S, Miller A. 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(7560): 177. 34. Blanco-Munoz J, Lacasana M, Borja-Aburto VH. Maternal miscarriage history and risk of anencephaly. Paediatr Perinat Epidemiol. 2006;20(3):210-8. 35. Li Z, Ren A, Zhang L, Guo Z, Li Z. A population-based case-control study of risk factors for neural tube defects in four high-prevalence areas of Shanxi province, China. Paediatr Perinat Epidemiol. 2006;20(1):43-53. 36. Anderson JL, Waller DK, Canfield MA, Shaw GM, Watkins ML, Werler MM. Maternal obesity, gestational diabetes, and central nervous system birth defects. Epidemiology. 2005;16(1):87-92. 37. Mandiracioglu A, Ulman I, Luleci E, Ulman C. The incidence and risk factors of neural tube defects in Izmir, Turkey: a nested case-control study. Turk J Pediatr. 2004;46(3):214-20. 38. Ray JG, Vermeulen MJ, Meier C, Wyatt PR. Risk of congenital anomalies detected during antenatal serum screening in women with pregestational diabetes. QJM. 2004;97(10):651-3. 39. Morris SE, Thomson AO, Jarup L,d e Hoogh C, Briggs DJ, Elliott P. No excess risk of adverse birth outcomes in populations living near special waste landfill sites in Scotland. Scott Med J. 2003;48(4): 105-7. 40. Suarez L, Cardarelli K, Hendricks K. Maternal stress, social support, and risk of neural tube defects among Mexican Americans. Epidemiology. 2003;14(5):612-6. 41. Farley TF, Hambidge SJ, Daley MF. Association of low maternal education with neural tube defects in Colorado, 1989-1998. Public Health. 2002;116(2):89-94. 42. Elliott P, Briggs D, Morris S, de Hoogh C, Hurt C, Jensen TK, et al. Risk of adverse birth outcomes in populations living near landfill sites. BMJ. 2001;323(7309):363-8. 43. Todoroff K, Shaw GM. Prior spontaneous abortion, prior elective termination, interpregnancy interval, and risk of neural tube defects. Am J Epidemiol. 2000;151(5):505-11. 44. Carmichael SL, Shaw GM. Maternal life event stress and congenital anomalies. Epidemiology. 2000;11(1):30-5. 45. Wasserman CR, Shaw GM, Selvin S, Gould JB, Syme SL. Socioeconomic status, neighborhood social conditions, and neural tube defects. Am J Public Health. 1998;88(11):1674-80. 18

Jo

ur

na

lP

re

-p

ro of

46. Dolk H, Vrijheid M, Armstrong B, Abramsky L, Bianchi F, Garne E, et al. Risk of congenital anomalies near hazardous-waste landfill sites in Europe: the EUROHAZCON study. Lancet. 1998;352(9126):423-7. 47. Croen LA, Shaw GM, Sanbonmatsu L, Selvin S, Buffler PA. Maternal residential proximity to hazardous waste sites and risk for selected congenital malformations. Epidemiology. 1997; 8(4): 347-54. 48. Canfield MA, Annegers JF, Brender JD, Cooper SP, Greenberg F. Hispanic origin and neural tube defects in Houston/Harris County, Texas. II. Risk factors. Am J Epidemiol. 1996;143(1):12-24. 49. Janssen PA, Rothman I, Schwartz SM. Congenital malformations in newborns of women with established and gestational diabetes in Washington State, 1984-91. Paediatr Perinat Epidemiol. 1996;10(1):52-63. 50. Kurinczuk JJ, Clarke M. A case-control study to investigate the role of recent spontaneous abortion in the aetiology of neural tube defects. Paediatr Perinat Epidemiol. 1993;7(2):167-76. 51. Yang Q, Wen SW, Leader A, Chen XK, Lipson J, Walker M. Paternal age and birth defects: how strong is the association? Hum Reprod. 2007;22(3):696-701. 52. Kazaura M, Lie RT, Skjaerven R. Paternal age and the risk of birth defects in Norway. Ann Epidemiol. 2004;14(8):566-70. 53. McIntosh GC, Olshan AF, Baird PA. Paternal age and the risk of birth defects in offspring. Epidemiology. 1995;6(3):282-8. 54. Norman SM, Odibo AO, Longman RE, Roehl KA, Macones GA, Cahill AG. Neural tube defects and associated low birth weight. Am J Perinatol. 2012;29(6):4736. 55. Gu X, Lin L, Zheng X, Zhang T, Song X, Wang J, et al. High prevalence of NTDs in Shanxi Province: a combined epidemiological approach. Birth Defects Res A Clin Mol Teratol. 2007;79(10):702-7. 56. Nili F, Jahangiri M. Risk factors for neural tube defects: a study at universityaffiliated hospitals in Tehran. Arch Iran Med. 2006;9(1):20-5. 57. Khoury MJ, Erickson JD, Cordero JF, McCarthy BJ. Congenital malformations and intrauterine growth retardation: a population study. Pediatrics. 1988;82(1):83-90 58. Anomalies EESoC.EUROCAT Prevalence Tables. 2012-2016. http://www.eurocat-network.eu/accessprevalencedata/prevalencetables. 59. Hobel CJ, Dunkel-Schetter C, Roesch SC, Castro LC, Arora CP. Maternal plasma corticotropin-releasing hormone associated with stress at 20 weeks' gestation in pregnancies ending in preterm delivery. Am J Obstet Gynecol. 1999;180(1 Pt 3):S257-63. 60. Wadhwa PD, Dunkel-Schetter C, Chicz-DeMet A, Porto M, Sandman CA. Prenatal psychosocial factors and the neuroendocrine axis in human pregnancy. Psychosom Med. 1996; 58(5): 432-46. 61. Wadhwa PD, Sandman CA, Garite TJ. The neurobiology of stress in human pregnancy: implications for prematurity and development of the fetal central nervous system. Prog Brain Res. 2001;133:131-42. 62. Wang M, Wang ZP, Gong R, Zhao ZT. Maternal smoking during pregnancy and 19

Jo

ur

na

lP

re

-p

ro of

neural tube defects in offspring: a meta-analysis. Childs Nerv Syst. 2014;30(1):83-9. 63. Li Z, Zhang L, Jin L, Ye R.Raynes-Greenow C.Ren A.et al. Poor sleep during the periconceptional period increases risk for neural tube defects in offspring. Birth Defects Res A Clin Mol Teratol. 2015; 103(0): 780-86. 64. van Gelder MM, Reefhuis J, Caton AR, Werler MM, Druschel CM, Roeleveld N, et al. Maternal periconceptional illicit drug use and the risk of congenital malformations. Epidemiology. 2009;20(1):60-6. 65. Carmichael SL, Yang W, Shaw GM. Periconceptional nutrient intakes and risks of neural tube defects in California. Birth Defects Res A Clin Mol Teratol. 2010;88(8):670-8. 66. Peterson C, Grosse SD, Li R, Sharma AJ, Razzaghi H, Herman WH. et al.Preventable health and cost burden of adverse birth outcomes associated with pregestational diabetes in the United States. Am J Obstet Gynecol.2015;212(1): 74 e71-79. 67. Starikov RS, Inman K, Has P, Iqbal SN, Coviello E, He M. Correlation of placental pathology and perinatal outcomes with Hemoglobin A1c in early pregnancy in gravidas with pregestational diabetes mellitus. Placenta . 2017;52: 94-9. 68. Brown JE, Murtaugh MA, Jacobs DR, Jr., Margellos HC. Variation in newborn size according to pregnancy weight change by trimester. Am J Clin Nutr. 2002;76(1):205-9. 69. Penrose LS. Parental age and mutation. Lancet. 1955;269(6885):312-3. 70. Meng X, Sun Y, Duan W, Jia C. Meta-analysis of the association of maternal smoking and passive smoking during pregnancy with neural tube defects. Int J Gynaecol Obstet. 2018;140(1):18-25. 71. Agha MM, Glazier RH, Moineddin R, Moore AM, Guttmann A. Food fortification and decline in the prevalence of neural tube defects: does public intervention reduce the socioeconomic gap in prevalence? Int J Environ Res Public Health. 2013;10(4):1312-23.

20

Figure captions

na

lP

re

-p

ro of

Figure 1. Flow gram of study selection process

Jo

ur

Figure 2. Begg’s funnel plot for publication bias test

21

ro of -p re lP

Jo

ur

na

Figure 3. Subgroup analysis of paternal age and spina bifida risk

22

ro of -p re lP na ur

Jo

Figure 4. Sensitivity analysis of maternal stressful life events and NTDs risk

23

24

ro of

-p

re

lP

na

ur

Jo

Table 1Characteristics of the included maternal studies in meta-analysis Author Year

Number

OR/R

egion

(case/control)

R

LL

UL

China

Ingstrup et al. 2016

832/1271

3.25

2.56

4.12

Denmark 1115/1733075 0.92

0.58

1.46

China

459/459

1.97

1.1

3.52

Iran

91/209

4.9

1.9

12.8

552/2974

1.5

India

284/568

1.1

China

631/862

1.1

2

0.79

1.53

4.2

1.4

12.6

Grewal et al. 2009

0.86

1.16

Europe

China

518/6424 3526/NR

China

0.87

133/273 459/459 631/857

0.71

1.07

1.57 0.68 1.12

1.14 0.45

2.16 1.02

Jo

Suarez et al. 2007 Czeizel et al. 2006

0.74

1.69

7

study

5

study

6

study

6

3.4

1.4

8.3

6

case-control study

7

case-control study

5

study

7

study

8

case-control

China

227/227

1.68

1.11

2.53

USA

283/552

1.7

1.2

2.6

USA

study

case-control

study

8

case-control study

5

case-control 295/695

1.58

1.24

2

study

6

case-control USA

655/4368

1

0.6

1.7

study

7

case-control Hungary

1202/38151

0.77

0.54

1.08

study

6

retrospective

Macintosh et al. 2006

7

case-control

Carmichael et al. 2007

study

-p

1

re

USA

ur

Yin et al. 2011

1038/5859

na

Ye et al. 2011

study

case-control

Italy

Lu et al. 2011

8

case-control

De Marco et al. 2011

study

case-control

lP

Garne et al. 2012

8

case-control

USA

Yazdy et al. 2012

cohort study

case-control

Anderka et al. 2012

8

ro of

Li et al. 2013

study

case-control

USA

Deb et al. 2014

quality

case-control

Carmichael et al. 2014

Study

retrospective

Talebian et al. 2015

Study design case-control

Zhang et al. 2018

Lu et al. 2015

Country/R

Britain

NR

4.2

25

2

7.8

cohort study

7

case-control

Blanco-Munoz et al. 2006

Mexico

58/58

4.58

1.22

17.23

China

158/226

2.33

0.96

5.65

7

case-control

Li et al. 2006

study

5

case-control

Anderson et al. USA

2005

353/497

0.56

0.1

3.1

study

7

case-control

Mandiracioglu et Turkey

al. 2004

44/88

3.33

1.29

8.61

study

6

retrospective

Ray et al. 2004

Canada

285/NR

6.8

NR

0.71

0.36

1.42

MexicoAmerica

cohort study

7

study

7

case-control

184/225

2.99

1.8

4.7

study

5

case-control

Farley et al. 2002

USA

224/930

1.8

1.1

3.1

study

7

case-control

Britain

NR

1.05

1.01

1.1

study

-p

Elliott et al. 2001

7

case-control

USA

276/3029

USA

399/426

Carmichael and Wasserman et al. USA Europe

538/539 NR

na

1998

261/464

ur

USA

Janssen et al. 1996

3.85

USA

507/517 91/451

0.61

1.1

1.5

1.1

2.1

1.2

0.8

1.8

7

1.86

study

5

case-control study

5

study

6

case-control 1.24

2.79

study

8

case-control 1

0.7

1.3

study

7

case-control 0.98

0.53

1.81

study

6

retrospective

USA USA

91/451

2.03

1.16

3.57

Clarke 1993

England

177/354

1.03

0.74

1.43

Canfield et al. 1996

study

case-control

Jo

Canfield et al. 1996

0.82

lP

USA

Shaw 2000

1.5

case-control

Todoroff and Shaw 2000

2.4

re

Correa et al. 2000

Croen et al. 1997

0.9

ro of

Scotland

Suarez et al. 2003

Dolk et al. 1998

2.5

case-control

Morris et al. 2003

Kurinczuk

study

NR

6.2

0.9

44

cohort study

8

case-control

and

study

6

case-control study

5

case-control Fedrick 1974

Britain

459/1763

1.57

1.07

2.32

study

5

Abbreviations:OR/RR, odds ratio/relative ratio; LL, low limit of 95%CI; UL, upper limit of 95%CI; NR, not report.

26

Table 2Characteristics of the included paternal studies in meta-analysis anencephaly Countr Author Year

y/

spinal bifida

number

Pater

number

Region (case/con OR/ trol) RR

LL

UL

nal

(case/con OR/ trol)

RR

LL UL

age

Stu Study

dy

design qual ity retrospe

2007

USA 538/NR 0.85

0.51

1.35

1195/NR 1.36

1.02 1.8

<20

ctive cohort

ro of

Yang et al.

7

study

retrospe

Kazaura et al. 2004

Norway 512/NR 1.4

0.6

3.4

791/NR 1.4

0.8 2.8

<20

ctive

cohort

7

al. 1995

Britain 127/251 9

1.5

55

595/1185 2

re

McIntosh et

-p

study

0.8 4.7

case<20

control study

Abbreviations:OR/RR, odds ratio/relative ratio; LL, low limit of 95%CI; UL, upper limit of 95%CI; NR,

Jo

ur

na

lP

not report.

27

6

Table 3 Characteristics of the included neonatal studies in meta-analysis Author Year

Country/Reg

Number

ion

(case/control)

OR/RR

LL

UL

Study design

Study quality

case-control Deb et al. 2014

India

284/568

1.1

0.79

1.53

study

5

retrospective Norman et al. 2012

USA

128/66786

2.5

1.6

3.9

cohort study

8

case-control China

126/6236

1.5

1.04

2.17

study

ro of

Gu et al. 2007

7

case-control

Gu et al. 2007

and

0.57

Jahangiri.

2006 Nili

0.95

Farley et al. 2002 Khoury et al. 1988

study

7

case-control

Iran

192/193

1.12

Iran

192/193

2.98

1.1

Jahangiri.

2006

1.59 1.39

study

7

case-control

USA USA

224/930

-p

and

128/5773

1.53

1.4

1

re

Nili

China

284/NR

7.18

2.02

4.61 1.9

study case-control study

lP

na ur Jo 28

7

retrospective 25.5

cohort study

Abbreviations:OR/RR, odds ratio/relative ratio; LL, low limit of 95%CI; UL, upper limit of 95%CI; NR, not report.

7

7

Table 4 Meta-analysis of risk factors for Neural Tube Defects First authors Year No.of case/total

tea drinking

Yazdy 2012

193/518

Ye 2011

20/631

Correa 2000

NR/276

Fedrick 1974

427/464

history of maternal abortion Talebian 2015

17/27

De Marco 2011

38/ 144

Lu 2011

86/153

Yin 2011

78/131

Blanco-Munoz 2006

17/21

Todoroff 2000

122/271

Canfield 1996

27/118

Kurinczuk 1993

6/35

De Marco 2011

23/97

Lu 2011

35/54

Kurinczuk 1993

6/35

stressful life events

Ingstrup 2016 Carmichael 2014 Li 2013

0.082

<0.001

1.31(0.95-1.80)

0.096

<0.001

0.93(0.55-1.56)

0.841

0.023

1.61(1.24-2.08)

<0.001

0.009

1.42(1.19-1.70)

<0.001

0.054

1.48(0.73-3.00)

0.274

<0.001

2.24(1.21-4.12)

0.01

0.043

31/38

281/556 71/111

Carmichael 2000

85/199

Deb 2014

104/372

Grewal 2008

112/269

Li 2006

146/341

na

Mandiracioǧlu 2004

40/99

Farley 2002

28/93

ur Jo

1.69(0.94-3.06)

NR

Suarez 2003

low maternal education level

19/34407

lP

Carmichael 2007

PHet

re

abortion

P-value*

CI)

-p

history of spontaneous

OR/RR(95%

ro of

Risk factor

Wasserman 1998

204/353

Canfield 1996

37/191

nausea and vomiting during Zhang 2018

258/832

the first trimester of pregnancy

pregestational diabetes

Lu 2015

53/81

Anderka 2012

NR

Czeizel 2006

97/3874

Garne 2012

41/NR

Macintosh 2006

10/2400

Anderson 2005

10/13 29

hazardous waste sites

Ray 2004

4/2069

Janssen 1996

2/1511

Suarez 2007

20/137

Morris 2003

14/149

Elliott 2001

3508/4648

Dolk 1998

130/NR

Croen 1997

123/238

Yang 2007

65/NR

Kazaura 2004

8/NR

McIntosh 1995

15/32

Norman 2012

25/5642

Khoury 1988

626/NR

Deb 2014

104/372

Gu 2007

68/2829

Nili 2006

147/238

Farley 2002

129/589

Gu 2007

66/3075

Nili 2006

132/228

1.10(0.89-1.37)

0.381

0.061

1.41(1.10-1.81)

0.007

0.718

5.53(1.95-15.70)

0.001

<0.001

low birth weight

female gender

birth season

re

*: Significant positive results (P<0.05).

1.54(1.10-2.14)

0.012

0.025

1.11(0.99-1.25)

0.071

0.54

-p

25-29)

ro of

low paternal age (<20 vs.

Jo

ur

na

lP

Abbreviations: OR/RR, odds ratio/relative ratio; CI, confidence interval; NR, not report.

30