Associations between maternal preconception and pregnancy adiposity and neuropsychiatric and behavioral outcomes in the offspring: A systematic review and meta-analysis

to identify studies on maternal adiposity and offspring neuropsychiatric outcomes. Inverse variance-weighted random-effects meta-analyses were used to pool effect estimates with 95 % confidence intervals (95 % CIs) from adjusted odds ratios (OR) and hazard ratios (HR). Estimates were computed separately for preconception and pregnancy maternal over-weight and obesity, with outcomes stratified by the type of neuropsychiatric outcome. In our meta-analyses of 42 epidemiological studies involving 3,680,937 mother-offspring pairs, we found increased risks of ADHD [OR = 1.57, 95 % CI: 1.42-1.74], autism spectrum disorder [OR = 1.42, 95 % CI: 1.22-1.65], conduct disorder [OR = 1.16, 95 % CI: 1.00-1.35], Psychotic disorder [HR = 1.61, 95 % CI: 1.41-1.83], externalizing behaviors [OR = 1.30, 95 % CI: 1.07-1.56] and peer relationship problems [OR = 1.25, 95 % CI: 1.04-1.27] in the offspring of preconception obese mothers. Similar increased risks were found in the offspring of preconception overweight mothers and those exposed to maternal adiposity during pregnancy. However, no association was found with offspring mood, anxiety, personality, eating, sleep disorders or prosocial problems. Preconception weight management may mitigate such adverse effects in the offspring.


Background
Over the past few decades, preconception and maternal obesity have emerged as a significant global epidemic among the reproductive age population, posing ongoing public health challenges (1,2).This increase in adiposity is not limited to women alone but is also observed in the general population, where paternal adiposity may also serve as a risk factor.Epidemiologic studies consistently demonstrate a significant rise in the prevalence of obesity and excessive weight gain among women of childbearing age across diverse populations (3,4).This growing concern extends to intergenerational health, as maternal adiposity has been linked with adverse perinatal outcomes, including preterm birth, low birthweight (2,5,6), small or large for gestational age and stillbirths (2,6).
These conflicting results could be due to differences in sample size or variations in covariate selection across studies.For instance, some studies have concluded that the association between maternal obesity and neuropsychiatric and behavioral outcomes in the offspring could be due to unmeasured socio-economic, familial, or maternal confounding rather than a direct causal effect of maternal adiposity (9,10,27,28).This suggests that incomplete statistical control for such putative risk factors may lead to biased estimates.However, the association between maternal adiposity and neuropsychiatric and behavioral problems in the offspring has not been systematically and comprehensively reviewed or quantified to date.One of the existing reviews (29)(30)(31) has merely examined the association between preconception maternal obesity and a few offspring neurodevelopmental outcomes (29).Subsequent reviews in 2018 included a few studies to examine the association between pre-pregnancy obesity and ASD (30), and ADHD in the offspring (31).More importantly, over the past six years, there has been a considerable increase in the number of publications on similar topics, suggesting that an update of the existing evidence through a more comprehensive systematic review and meta-analysis is warranted.Moreover, a more detailed systematic review and meta-analysis that examines the specific period of adiposity, dose-related effects, and their impact on a wide range of offspring outcomes is imperative for mechanistic studies, clinical and policy implications.We, therefore, examined the association between maternal preconception and pregnancy adiposity and adverse neuropsychiatric and behavioral outcomes in the offspring.

Study Design and Search Strategy
This systematic review and meta-analysis was conducted in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines (32).The protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO) with the registration number CRD42023414828.Two reviewers (BD and TSM) developed key search terms using the Population/participants, Interventions, Comparisons, Outcomes, and Study design (PICOS) framework (33) (Supplementary table 1).We searched citations using key terms that encompass a wide range of conditions and symptoms related to neurodevelopmental and behavioral issues including "neurodevelopmental", "oppositional defiant disorder (ODD)", "conduct disorder (CD)", "attention deficit hyperactivity disorder (ADHD)", "antisocial behavior", "neuropsychiatric disorders", "anxiety or anxiety disorder", "depression or depressive disorder", "mood disorder", "posttraumatic stress disorder", "acute stress disorder", "panic attack", "conversion disorder", "bipolar disorder", "manic attack", "eating disorder", "anorexia nervosa or bulimia nervosa", "psychosis", "schizophrenia", "delusion" and "suicide".BD and TSM developed a search strategy combining key terms and medical subject heading [MeSH terms] with relevant filters.We used Boolean operators ("AND", "OR", "NOT") to combine key words or independent searches and truncations to capture words that could have multiple endings.We searched four databases including PubMed, EMBASE, Scopus and Web of Science from inception to April 2024.The full search strategy is included as a Supplementary file 2. The electronic database search was updated in April 2024.

Study Selection and Exclusion Criteria
Citations retrieved from all electronic databases were imported into EndNote 20 and duplicates were removed.After removing duplicates, two reviewers (BD and TSM) independently screened titles and abstracts using Covidence systematic review software.Discrepancies in screening were resolved through discussion by AB.Studies were included in the meta-analysis if they: (1) examined the association between preconception or pregnancy overweight or obesity and offspring neuropsychiatric or behavioral problems; (2) were conducted on human populations and published in English language; and (3) used observational study designs (cohort studies, case-control, nested case-control, and crosssectional studies).Studies on animals, reviews, case series/reports, conference papers, conference proceedings, editorials, letters to the editors and commentaries were excluded.

Data Extraction and Quality Appraisal
TSM and BD extracted data on authors, year of publication, country, study design, study participants, study sample, study design/method, key findings and conclusion using a pre-prepared data extraction sheet.The extracted data was further checked consistency by AB.The methodological quality of the included studies was assessed by two independent reviewers (BD and TSM) using the Newcastle-Ottawa Scale (NOS) (34).This tool consists of three domains: selection of study participants, comparability of study groups, and reporting and ascertainment of outcomes.Each primary study was graded out of nine points as per the NOS coding manual (35) and summarized in three categories as good (if total score >7), fair (if total score 5 -6) or poor (if total score <5).Studies were deemed to be at high risk of bias if the NOS score was < 6. Disagreements in the assessment of methodological quality ratings were resolved through discussion.

Study outcome
The outcomes of interest of this systematic review and meta-analysis are any neuropsychiatric and behavioral outcomes in the offspring (excluding addictive behaviors), ADHD, ASD, mood disorders, anxiety disorders, conduct disorder, psychotic disorders, personality disorders, eating disorders, sleep problems, peer relationship and prosocial problems, and internalizing and externalizing behaviors at any age.

Statistical Analysis
We performed inverse-variance-weighted random-effects meta-analysis to address heterogeneity among studies included in the review using a STATA version 18.All primary studies that reported effect size using odds ratios/ relative risks (OR/RR), hazard ratio (HR) or provided data to calculate these were included in the final meta-analysis.Otherwise, we tabulated the results and presented a narrative review of the studies.For the studies that reported multiple effect estimates, the estimate with the most extensive adjustment were used for meta-analyses.For the studies that reported multiple outcomes, each outcome was reported separately in this review.Inverse variance weighted random effects meta-analysis models were used to combine studies to estimate the association between maternal preconception and pregnancy adiposity and offspring neuropsychiatric and behavioral adverse outcomes and to account for heterogeneity across the studies included in the review.To explore which offspring outcomes were specifically associated with maternal preconception and pregnancy overweight or obesity, we stratified meta-analyses by classifying outcomes to ADHD, ASD, mood disorders, anxiety disorders, conduct disorder, psychotic disorders, personality disorder, eating disorders, sleep related, peer relationship and prosocial problems, and internalizing and externalizing symptoms.Further, the estimates for maternal preconception and pregnancy adiposity were computed separately to check whether the effect estimates were sensitive to the specified time of exposure, i.e., preconception and pregnancy.Sub-group analyses were also performed according to the maternal adiposity, determined based on their body mass index (BMI) and classified according to the WHO criteria as overweight (25.0-29.9kg/m 2 ) and obesity (30.0-34.9kg/m 2 ) (36).Tau-and I 2 -test were used to detect the magnitude of statistical heterogeneity between studies.I 2 -test scores 75 % was considered to high heterogeneity between studies.We checked potential publication bias by inspection of the funnel plot and Egger's test for regression asymmetry.As a sensitivity analysis, we calculated the statistical power of each study included in the meta-analysis to determine if they had sufficient power to detect the effect of exposure on outcomes (Supplementary File 3).
Three studies were multi-national: one study in the UK and Netherlands (27), one in Sweden, Denmark, and Finland (13) and one in the USA and Canada (62).Of the 42 included studies, 39 were cohort studies and three were case control studies.The sample sizes for the included studies are ranging from the smallest with 62 participants to the largest cohort with 673,632 participants (Table 1).All the studies included in this review scored > 7 in the Newcastle Ottawa scale, suggesting good methodological quality (Supplementary file 5).

Meta-analysis: effect of preconception and pregnancy overweight/obesity on offspring neuropsychiatric and behavioral outcomes
Of the 42 studies included in this review, 16 reported on ADHD, seven on ASD, 11 on mood and anxiety-related disorders, 10 on internalizing behaviors, seven on externalizing behaviors, and six on peer relationship problems.The remaining studies reported on prosocial behavior, eating disorders, psychotic disorders, personality disorders, and other eating disorders.Most studies included in this review reported fully adjusted effect estimates, whereas one study presented an unadjusted effect estimate (36) (Table 1).
Offspring exposed to maternal adiposity both during preconception period and pregnancy were at increased risk of ASD.We found that offspring exposed to maternal preconception overweight [pooled OR=1.09, 95% CI:1.02-1.16]and obesity [pooled OR= 1.42, 95% CI:1.22-1.65]were at 9% and 42% increased risk of ASD, respectively, when compared with non-exposed offspring (Figure 4).We also observed a two-fold higher risk of ASD in offspring exposed to maternal obesity during pregnancy [pooled OR = 2.23, 95% CI: 1.23-4.05](Figure 5).
Offspring born to mothers who were obese prior to conception had a 16% higher likelihood of experiencing CD [pooled OR = 1.16, 95% CI: 1.00-1.35].However, maternal preconception overweight status was not associated with CD in offspring (Figure 6).Similarly, our meta-analysis suggested insufficient statistical evidence to support associations between offspring mood disorders and internalizing behaviors with maternal preconception adiposity (Figure 7).The meta-analysis of studies on externalizing behaviors found maternal preconception obesity associated with an increased risk [pooled OR = 1.30, 95% CI: 1.07-1.56].Conversely, maternal preconception overweight was not associated with an increased risk of offspring externalizing behaviors [pooled OR = 1.15, 95% CI: 0.92-1.44],suggesting a dose-response association (Figure 8).Similarly, unlike maternal preconception overweight, maternal preconception obesity was associated with an increased risk of offspring peer relationship problems [pooled OR = 1.47, 95% CI: 1.09-1.97](Figure 9).However, we were unable to examine the association between maternal preconception and pregnancy adiposity and certain offspring outcomes, such as anxiety disorder, sleep disorder, schizophrenia, and eating disorders, including anorexia nervosa, as only one study each examined those specific outcomes.Based on our narrative review of those studies, maternal preconception obesity was associated with an increased risk of anxiety disorder, while maternal pregnancy adiposity was associated with an increased risk of offspring peer relationship problems.All other offspring outcomes were not associated with maternal preconception and pregnancy adiposity (Supplementary file 6).
Similarly, maternal pregnancy adiposity was not associated with offspring psychotic disorder and prosocial behaviors (Supplementary file 7-8).
To further explore whether the estimated effects of maternal adiposity on offspring neuropsychiatric and behavioral outcomes were influenced by level of statistical adjustments in each study, we conducted additional sensitivity analyses.For instance, the effect estimates for ADHD in the offspring exposed to maternal preconception overweight [pooled OR=1.51, 95% CI:1.45-1.58,I 2 = 97.90%;Pvalue = 0.001] appeared to be elevated in the studies that did not adjust for maternal mental health problems compared to those that did adjust for this risk factor [pooled OR=1.18, 95% CI:1.12-1.24,I 2 = 9.1%; P-value = 0.001].Similarly, elevated effect estimates for ADHD were seen for offspring exposed to maternal preconception obesity in studies that did not control for maternal mental health problems [pooled OR=1.61, 95% CI:1.54-1.69,I 2 = 25.8;P-value = 0.001] when compared to those adjusted for this risk factor [pooled OR=1.28, 95% CI:1.21-1.36,I 2 = 76.4%;P-value = 0.001].We did not conduct additional sensitivity analysis based on the level of statistical adjustment for other risk factors due to a lack of sufficient data in the included studies.

Publication bias
Based on visual inspection of the funnel plot and egger's test for small-study effects, with a p-value of 0.514, indicates no significant evidence of publication bias in the meta-analysis of 42 studies (Supplementary file 9).

Confounding variables
Confounders may influence or modify the association between maternal preconception and pregnancy adiposity and adverse neurodevelopmental and psychiatric outcomes in the offspring.The studies included in this systematic review and meta-analysis partially or fully adjusted for various potential confounders.Maternal education and age were commonly adjusted for in most studies, while other covariates were inconsistently included in the final statistical models of the included studies.For instance, 33 studies included maternal age as a covariate, 25 studies adjusted for maternal mental health problems, and 18 studies controlled for maternal tobacco smoking (Table 2).

Discussion
This systematic review and meta-analysis aimed to examine prospective associations between maternal preconception and pregnancy adiposity and a spectrum of offspring neuropsychiatric disorders and behavioral problems.We retrieved 42 observational epidemiological studies, covering a range of disorders, including ADHD, ASD, CD, psychotic, mood, anxiety, personality and eating disorders, prosocial behavior, sleep and peer relationship problems, and internalizing and externalizing behaviors.Our meta-analyses found evidence supporting prospective positive associations between maternal preconception and pregnancy adiposity and the majority of offspring outcomes examined.
Although the number of studies included in our meta-analyses and the magnitude of prospective associations varied, our findings are in line with existing reviews (29)(30)(31).
The rising global prevalence of obesity among women of reproductive age highlights the relevance of examining the mechanistic relationship between preconception and maternal adiposity and the development of neuropsychiatric and behavioral outcomes in the offspring (63).Yet, the exact mechanisms are not fully understood despite several plausible suggestions proposed to elucidate such associations (64).Some of the plausible potential mechanisms linking neuropsychiatric and behavioral outcomes in offspring of preconception or pregnancy overweight or obese mothers include chronic low-grade inflammation, oxidative stress, dysregulated fatty acid metabolism, and hormonal imbalances, which can affect the intrauterine environment and disrupt fetal brain development (63,65).
Indirect mechanistic and causal associations are also conceivable, as maternal preconception and pregnancy adiposity are known to increase the risk of various acute and chronic maternal mental health conditions.These mental health conditions may potentially mediate the observed association between maternal adiposity and offspring behavior and mental health, making it challenging to disentangle the specific effect of maternal overweight/obesity on the offspring (65).In support of this explanation, we also noted attenuation in the strength of associations in the studies that adjusted for maternal mental health conditions, suggesting a predisposition to maternal or paternal mental health problems may partly explain some part of the association reported in our study.
Maternal obesity is frequently accompanied by comorbid conditions such as gestational diabetes and hypertensive disorders of pregnancy, which have been independently linked to an increased risk of neuropsychiatric and behavioral disorders in the offspring.Alternatively, pleiotropy may be driving some of this comorbidity.Most notably, fatty acids such as arachidonic acid have been shown to play a vital role in fetal brain development and may also be a potential risk factor in gestational diabetes (66).Therefore, it is imperative to consider and investigate the potential mediating roles of these comorbid conditions in order to gain a comprehensive understanding of the complex interplay between maternal obesity and offspring neuropsychiatric and behavioral outcomes (11,17).
Maternal obesity, diabetes and fatty acids share similar metabolic pathways, highlighting their intertwined roles in influencing offspring's behavioral and mental health outcomes (17,67).For instance, arachidonic acid, an omega-6 fatty acid that forms a key component of phospholipids, is known to play a critical role in early brain development (67).Further, arachidonic acid, along with its metabolites, modulates various brain processes, including neurotransmission and neuroinflammation (68).Corresponding neuroinflammation has been associated with a number of neuropsychiatric and behavioral problem in the offspring (69).Some studies have also suggested that maternal obesity may result in adverse neuropsychiatric and behavioral outcomes in the offspring through indirect pathways.
A meta-analysis synthesizing the results of 22 observational epidemiological studies has suggested that maternal overweight and obesity increase the odds of child obesity by 264% and 89%, respectively (70).This, in turn, was associated with some form of discrimination, social isolation and bullying, which have an impact on their neuropsychiatric and behavioral well-being (71,72).
In addition to the overall association between maternal adiposity and offspring neuropsychiatric and behavioral outcomes, some evidence highlights severe maternal obesity (BMI ≥35 kg/m²) is linked with specific neuropsychiatric outcomes in offspring (11,17,19,21).Notably, ADHD co-morbid with ASD (7) and co-occurring psycho-neurotic, mood, stress, and somatization disorders (11) were observed in offspring born to mothers with severe obesity.his underscores the need for more mechanistic studies that consider the severity of maternal obesity as a contributing factor to the increased risk of adverse outcomes in offspring, addressing the unique risks associated with this population.Indeed, it is also essential to highlight the increased risk of behavioral and mental health outcomes in offspring associated with both the combined and independent effects of obesity and/or excessive gestational weight gain (18,73,74).
Epidemiological evidence suggests that maternal adiposity increases the risk of preterm birth, with significant associations to spontaneous preterm labor, membrane rupture, and medically indicated preterm birth (75)(76)(77).For instance, the odds of having a spontaneous preterm birth and medically indicated preterm birth increased approximately two-fold in the offspring of obese mothers (76,77).These are associated with increased risks of neurobehavioral and psychiatric outcomes later in life (78)(79)(80).More specifically, individuals born preterm are 2-3 times more likely to be diagnosed with ADHD, ASD, anxiety, and other behavioral problems later in life (81,82).A few studies in our review that considered prematurity or being born preterm in their final statistical models observed an attenuation in the strength of the association, suggesting that being born preterm may play a role in the associations reported by our study.This comprehensive systematic review and meta-analysis has several strengths.A robust search strategy was employed to identify relevant studies, and a rigorous methodological quality assessment was conducted by three authors, ensuring the reliability of the findings.Meta-analyses were stratified by offspring outcomes, and maternal BMI status was stratified by BMI status and period of exposure, allowing us to investigate whether the timing of maternal obesity during pregnancy or preconception has differential effects on offspring outcomes.However, it is important to acknowledge the limitation of the study that we were unable to provide pooled estimates of neuropsychiatric and behavioral outcomes of offspring across detailed categories of obesity due to inconsistent cutoff points for BMI categories reported in the original studies.Moreover, the majority of studies included in this review had no information about gestational weight gaina well-known risk factor for a number of adverse outcomes in offspring (83).Further, inclusion of studies reporting neuropsychiatric and behavioral outcomes as reported by both mothers and teachers may have introduced heterogeneity in measurement approaches, potentially affecting the overall effect estimates in the meta-analysis.While this potential impact is less likely due to the limited number of studies presenting this issue, it is important to consider these biases when interpreting the results and drawing conclusions from this review.It remains difficult to isolate whether maternal adiposity before and during pregnancy contributes as a cause of adverse neuropsychiatric and behavioral outcomes in the offspring.Furthermore, there were large variations in the sample sizes of the studies included in our review.Furthermore, there were large variations in the sample size of the studies included in our review.The meta-analysis accounts for the standard error (SE), which is a function of sample size, and gives more weight to larger studies, thus systematically accounting for large sampling errors due to small sample sizes.We also confirm that we did not observe notable differences in the findings of studies based on their sample size.However, we note that studies with small sample sizes are more susceptible to the winner's curse, thereby impacting the pooled effect estimates of our meta-analyses.Moreover, alternative methods, such as negative control analysis, may provide insights into whether maternal adiposity before conception and during pregnancy influences neuropsychiatric and behavioral outcomes in the offspring.This can be done by comparing the association between maternal and paternal adiposity with offspring neuropsychiatric and behavioral outcomes.Assessing the impact of paternal adiposity on adverse outcomes in offspring can help determine if the observed associations are a result of genetic predisposition to neurodevelopmental disorders or if they are likely causal.Although this review does not include a dedicated investigation into the effects of paternal adiposity on offspring outcomes, existing evidence generally suggests no significant association between parental adiposity and offspring neuropsychiatric adverse outcomes such as ASD (30).It is also plausible that the associations observed in this systematic review and meta-analysis could be due to confounders, such as maternal and familial socio-economic positions, prenatal substance use, and parental mental health problems, which are linked with maternal preconception and pregnancy adiposity and the outcomes studied in the offspring.Some of these confounders were not consistently adjusted in the studies included in this review, and this may have affected the associations reported in our study.

Conclusion
This systematic review and meta-analysis provides evidence of prospective positive associations between maternal preconception and pregnancy adiposity and neuropsychiatric and behavioral outcomes in the offspring.However, to elucidate causal pathways and validate these findings, further mechanistic studies, such as genetically sensitive studies and sibling-controlled analyses, are warranted.These studies are also necessary to inform targeted intervention programs focusing on the management of preconception and pregnancy adiposity.Importantly, future studies should comprehensively address potential confounding factors such as socio-demographic, familial, environmental, and genetic/biological confounders that could influence outcomes.The findings of this comprehensive review contribute to the existing body of knowledge and have important clinical, policy, and research implications.

Figure 2 .
Figure 2. The effects of maternal preconception overweight and obesity on offspring ADHD.

Figure 3 .
Figure 3.The effects of maternal pregnancy overweight and obesity on offspring ADHD.

Figure 4 .
Figure 4.The effects of maternal preconception overweight and obesity on offspring ASD.

Figure 5 .
Figure 5.The effects of maternal pregnancy overweight and obesity on offspring ASD.

Figure 6 .
Figure 6.The effects of maternal preconception overweight and obesity on offspring conduct disorder.

Figure 7 .
Figure 7.The effects of maternal preconception overweight and obesity on offspring mood disorders (a) and internalizing behaviors (b).

Figure 8 .
Figure 8.The effects of maternal preconception overweight and obesity on offspring externalizing behaviors.

Figure 9 .
Figure 9.The effects of preconception maternal overweight on offspring peer relationship problems.