Birth by caesarean section and offspring overweight and obesity in adult life: a systematic review and meta-analysis Firstname

: Overweight/obesity is one of the most important health problem worldwide. Birth by cesarean section has been shown to influence long-term health outcome including obesity. The aim of this systematic review-meta-analysis is to examine whether cesarean section increases the risk of offspring’ s overweight/obesity. The study follows the PRISMA and MOOSE guidelines. A systematic literature search was conducted on Scopus, PubMed, and WoS until December 2020. For inclusion, studies must have reported either (I) both Birth by cesarean section and adult (≥ 18 years) offspring BMI, (II) cohort or case–control study design and (III) a risk estimate. Heterogeneity testing was performed using Cochran's Q and I2 statistics. Publication bias was assessed by Egger’s test and Begg’s test. Meta-analysis was performed through a random effect model. Eleven studies with a combined population of 180.408 subjects were included in the meta-analysis. The overall analysis (n = 18) yielded a combined risk estimate for overweight/obesity of 1.19 (95% CI 1.08-1.31) and the test of heterogeneity resulted Q=53,37 (I2 = 70,37 %, P≤0•0001). The risk of offspring obesity is 1.23 (95% CI 1.09-1.39) and the test of heterogeneity resulted Q=39.50 (I2= 72,15%, P≤0•0001). cesarean section have an increased risk of developing obesity in adulthood context epigenetics mechanisms, regulation of 180,408 subjects, from eight countries, and four different continents. In addition, our review separately assessed the effects of CS on the different categories of BMI (overweight, obesity) in an adult-only population and in the subcategories of sexes and age groups. Our study is the fifth meta-analysis about the relation between CS delivery and overweight/ obesity in offspring


Introduction
Overweight and obesity represent relevant public health concerns worldwide. The prevalence of obesity has been increasing significantly in the past two decades and overweight/obesity is now recognized as a global epidemic. In 2016, 13% of the world's adult population were obese. The prevalence of obesity was about 30% in China and in USA and 25% in Europe. [1,2]. A recent NEJM report estimated that, by 2030, nearly half of adults in the USA will have obesity, with a prevalence not below 35% in any state, and nearly one in four will have severe obesity, with a prevalence higher than 25% in 25 states [3]. Overweight and obesity are associated with an increased risk of Non-Communicable Diseases (NCDs) and are responsible for about 4 million deaths every year worldwide. Reducing by one third premature deaths by NCDs is one of the targets of the 2030 Agenda for Sustainable Development of the United Nations [4,5,6]. Birth by cesarean section (CS) has been shown to influence health outcome; CS rates are increasing globally, accounting for more than 20% worldwide births. [7]. Given the increasing use of CS and the growing evidence on its effects on the health of mother and the effects on children, a deeper knowledge of its long-term effects/impact on offspring's adulthood health status has become crucial [8]. Recent literature showed that babies born by CS have different hormonal and bacterial patterns and that this fact can change neonatal physiology and health outcomes [9][10][11][12][13]. Subjects born by CS present increased risk of immune-related diseases in later life such as allergies [9], asthma [10], celiac disease [11], type 1 diabetes [12], and metabolic and cardiovascular disease risk factors such as elevate systolic blood pressure, excess fat mass and high body mass index BMI (13). Two meta-analyses, published in 2012 and 2015, to date investigate the association between CS and BMI in offspring, considering overall the period of childhood, adolescence, and to some extent adulthood [14,15]. A recent meta-analysis (2014) found a strong association between CS and increased BMI of offspring's adulthood, reporting a mean increase of 0.44 kg/m2 in BMI of adults born by CS (95% CI: 0.17, 0.72; P = 0.002; I2 = 39%) [16] whereas a recent (2015) systematic review found a slight influence between CS and the risk of overweight and obesity, which changes with age [17]. A recent update [18] of the previous review [17], included articles only to 2019, and founding similar results. Since then, several other studies with wider sample sizes and longer follow-up times have been published. Therefore, a systematic review and meta-analysis was conducted to summarize the evidence on the association between CS delivery and offspring BMI in adulthood and to assess the risk of overweight and obesity in offspring's adulthood.

Literature search strategy
Our literature search was aimed at identifying available research studies that investigated the effects of CS on offspring BMI. We identified the studies included in our meta-analysis by searching, without restrictions, multiple literature databases including Scopus, Web of Knowledge and PubMed, and selecting all the articles published until 2nd January 2021. We searched for abstracts and articles including the following terms: caesarean OR cesarean OR caesarian OR cesarian) AND (obesity OR Body Mass Index OR Overweight) AND (Adult) AND (Offspring). In addition, we supplemented this research by checking the references cited in retrieved papers and recent reviews.

Data extraction
We systematically reviewed and selected the studies meeting the following eligibility criteria: (I) assessed both mode of delivery and adult (≥ 18 years) offspring BMI, (II) used a cohort or case-control study design and (III) reported a risk estimate (hazard ratio, relative risk, prevalence ratio or OR) for the association between mode of delivery and BMI, as well as its 95% CI. When studies reported data from the same population, only the most comprehensive study was enrolled. Studies providing insufficient or overlapping data and conference abstract were excluded. Two investigators (M. C and B. DS) reviewed the eligibility of all studies according to the predetermined eligibility criteria independently. We extracted information about study characteristics (study name, authors, publication year, study design), study population characteristics, type of exposure (mode of delivery), outcomes (BMI…), and variables of adjustment. The mode of delivery was categorized in vaginal delivery (VD), as reference (including natural, forceps and vacuum extraction), and cesarean section (CS) including Planned CS (elective) and non-planned CS (non-elective). The outcomes of interest in our analysis were overweight and obesity classified according to the International Classification of BMI [19], namely 'overweight'' (BMI=25.0-29.9) and ''obese'' (BMI≥30).

Quality Evaluation
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 7 December 2021 We used the Newcastle-Ottawa Scale Assessment [20] for the evaluation of quality of the enrolled studies. Newcastle-Ottawa Scale adopted a star system scoring from 0 to 9 and a total score ≥7 indicated a high-quality study. Two investigators (B. DS. and I.G.) performed the quality evaluation of each selected study and disagreements were settled by a joint reevaluation of the original article with a third reviewer. No study was excluded based on quality criteria, in order to avoid selection bias.

Statistical Analysis
We evaluated the association between born by CS and BMI using the statistical program ProMeta version 3.0 (IDo Statistics-Internovi, Cesena, Italy). For the overall estimation, the hazard ratio was taken as an approximation to the OR, and the meta-analysis was performed as if all types of ratios were ORs. The combined risk estimate was calculated using a random-effects model in which the effect measures were ORs or hazard ratio. We assessed heterogeneity between studies by the Cochran's Q statistic (χ2), deeming p < 0.05 as significant, and I2 test, which yields results ranged from 0 to 100% (I2= 0-25%, no heterogeneity; I2= 25-50%, moderate heterogeneity; I2= 50-75%, large heterogeneity; and I2= 75-100%, extreme heterogeneity) [21]. To explore the sources of heterogeneity among studies and to test the robustness of the associations, we conducted subgroup analyses and sensitivity analyses. We further examined the influence of individual studies on the overall risk estimate, which was investigated by recalculating the pooled estimates for the remainder of the studies by omitting one study at each turn. Publication bias was evaluated using the methods of Begg and Mazumdar [22] and Egger et al. [23] (, which both test for funnel plot asymmetry, the former based on the rank correlation between the effect estimates and their sampling variances, and the latter based on a linear regression of a standard normal deviate on its precision. If the intercept of Egger's regression line deviated from zero with a p value < 0.10, the funnel plot was considered asymmetrical. In case of a small number (25 or fewer) of studies enrolled in the meta-analysis, as in the current review, this test for asymmetry possesses relatively low power to detect a real publication bias. If a potential bias was detected, sensitivity analyses were performed to assess the robustness of our findings. p values reported are from 2-sided statistical tests and differences with p < 0.05 were considered significant. This review is reported according to Metaanalysis Of Observational Studies in Epidemiology (MOOSE) [24] and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [25].

Study profiles
The study selection process is shown in Figure 1. Our initial research returned 316 citations. After excluding 142 duplicates, the analysis of titles and abstracts identified 10 studies on the potential association of CS with offspring BMI. Through the reference lists of recent relevant reviews and already selected articles, 5 additional articles were included for the analysis. From the 15 potentially eligible articles, 4 studies were excluded from the analysis after the full-text assessment, as follows ( Fig. 1): -2 studies report overlapping data -1 study is an abstract conference -1 study reports no risk estimate Eleven studies with a combined population of 180.408 subjects were included in the metaanalysis.
The forest plots are shown in Figure 2. Further analyses were performed stratifying by study design, sex, age, geographic area and level of BMI (Overweight/Obesity) ( Table 2).

Cesarean birth and overweight offspring
The overall analysis of the 5 studies pooled together (n = 6) yielded a combined risk estimate for overweight of 1.11 (95% CI 0.94-1.32) and the test of heterogeneity resulted Q=13.87 (I2= 63,96%, P=0·0160). The forest plots are reported in Figure 2.

Cesarean birth and obesity offspring
The overall analysis of the 10 studies pooled together (n = 12) yielded a combined risk estimate for obesity of 1.23 (95% CI 1.09-1.39) and the test of heterogeneity resulted Q=39.50 (I2= 72,15%, P≤0·0001). The forest plots are reported in Figure 2.

Publication Bias
Bias detection revealed a significant effect (P=0·005) using the method of Egger, while no bias was detected by the Begg and Mazumder test (P=0.063) (Figure 3). In the stratified analysis on the association between CS and obesity a significant publication bias was detected by Egger test; (P=0.032), but not by Begg and Mazumder test; P=0.217) The funnel plots of the meta-analyses are shown in figure 3.

Sensitivity analyses
Sensitivity analyses by omitting one study in turn suggested that the overall risk estimates were not substantially modified by any single study, with a range from 1.15 (95 % CI 1.05-1.27) to 1.22 (95 % CI 1.10-1.35). Heterogeneity was still observed after omitting each study in turn.

Discussion
Several studies confirmed that CS is related to different hormonal, physical and bacterial exposures and that this fact can change neonatal physiology and health outcomes [8][9][10][11][12][13]. Recently, the relationship between CS and offspring's BMI has attracted a great deal of attention. Therefore, we conducted a systematic review and meta-analysis to summarize the evidence on the association between mode of birth and BMI in offspring's adulthood and to assess the risk of obesity and overweight in offspring's adulthood associated with CS. The overall results of the current meta-analysis indicate that CS was associated with a significantly higher risk (1.19; 95% CI 1.08-1.31) of overweight/obesity in offspring's adulthood. The overall analysis (population= 180.408 subjects) showed a statistically significant increase of overweight/obesity risk in the cohort studies, in females, in age group 18-20 and in Europe. In case of obesity, significant effects were observed also in case control study and age>21. Our study shows a decrease in significance of overweight/obesity risk as age increases. The obesity risk is inversely associated to increasing age groups probably in correlation with the fact that the influence of environmental factors on obesity risk increases. This observation could be because lifestyle and environmental factors such as high food consumption, high sweetened beverages, less activity, television watching that are causative of obesity may have assumed greater importance [40]. This may be that the association between CS and overweight/obesity risk factors is more pronounced in prosperous societies, as found in our study with significant results in Europe, where breastfeeding is less common and where there is limited exposure to diverse bacteria in the immediate postnatal period. The underlying causes for obesity are complex, involving social and cultural aspects, and biological context involving epigenetics mechanisms, regulation of food intake, and even the gut microbiota [40]. Significant differences have been documented between the microbiome of infants born by caesarean section and infants born vaginally. Such differences were also proven to appear in the gut flora of normal weight people with respect to obese subjects. Children born by CS have similar patterns of intestinal colonization caused by lack of contact with maternal vaginal and intestinal flora [41,42]. The intestinal flora of infants born through caesarean section is less rich and diverse. Bacterial species protective against later obesity such as Bifidobacteria spp. and Bacteroides, are almost absent and the microbes such as that colonize their gut allows them to harvest more nutrients from diet, this is therefore presumed to persist in adulthood [41,43]. The mechanism by which CS increases the risk of overweight/obesity could be related to altered epigenetic regulation of obesity gene expression due to higher DNA methylation. CS, in fact, affects epigenetic activity in newborn infants who exhibited a higher DNAmethylation than infants vaginally born [44], on the other hand a higher DNA methylation may be associated to higher risk of overweight and obesity. The mechanism by which CS increases the risk of overweight/obesity could be related to lower rate of breastfeeding in children born by CS. The prevalence of obesity is significantly lower in breastfed people and breastfeeding is associated with a reduction in obesity risk [45]. Similarly to our study, previous meta-analyses found an increased risk of overweight/obesity in offspring's adulthood in association with CS [14,16,17,18], combined risk estimate for obesity/overweight of 1.19 (95% CI 1.08-1.31), lower compared to the results of previous meta-analyses. These discrepancies are likely due to the inclusion of more recent studies and to the increased population in study, which contributed to improve the precision and reliability of our risk estimates. Indeed, the meta-analysis by Li [17]. On the other hand, the recent meta-analysis by Quecke [18] does not consider recent studies [36,37] which enrolled about 36000 subjects, and, moreover, reported a risk of obesity of 1.22 (IC 25% 1.02-1.46), while the RR of out meta-analysis is 1.23 (1.09-1.39), with a smaller interval of confidence. Similar to Sutharsan's results, that demonstrated a decreasing effect of overweight/obesity in offspring born by CS across the life stages, our study in age group 18-20 has overweight/obesity risk of 1.27 (1.08-1.49) and in group age>20 have risk of 1.18 (0.99-1.40). Moreover, our finding about the association of CS and obesity 1.23 (95% CI 1.09-1.39), in contrast with Darmasseelane's results [16]; is stronger than the association of CS and overweight 1.11 (0.94-1.32) The study provides further evidence for CS as a risk factor for obesity in adult offspring and the association between CS and higher risk of overweight/obesity pose a public health concern. The World Health Organization (WHO) recommendations suggest a CS rate lower than 15% [46] however in developed countries CS rate is higher than recommended and an increasing number of CS have been performed in low-risk pregnancies, without a clear medical indication. Our results reinforce the importance of limiting performing CS for its influence on long term offspring health. In pregnancies with an indication for CS delivery, a greater understanding of the biological mechanism in needed to develop intervention to try to guarantee a microbial colonization like the one of a vaginal delivery and to modify the effect that CS has on obesity. As discussed above, the potential preventive role of early prebiotic or probiotic supplementation, to mimic the intestinal microflora of vaginally delivered infants, on long-lasting health consequences of an altered early life gut microbiota [47], needs to be investigated.

Strengths and Limitations
Our studies collectively represent a large population of 180,408 subjects, from eight countries, and four different continents. In addition, our review separately assessed the effects of CS on the different categories of BMI (overweight, obesity) in an adult-only population and in the subcategories of sexes and age groups. Our study is the fifth metaanalysis about the relation between CS delivery and overweight/ obesity in offspring adults, but it is the only one that have tried to study some subgroups, such as geographical area.
The studies limitations are the small number of studies included, the few risks estimated considered and the failure to investigate how much environmental factors affect obesity risk by age. Furthermore, the prevalence of CS is strongly and directly related to socioeconomic position, residual confounding may affect associations with any outcomes that also vary according to wealth. In any case the control of potential confounders is of considerable importance in studies investigating the association of C-section and obesity.
Author Contributions: All authors have contributed significantly. Author C.M., DS B. and G. I. provided the idea, designed the study, collected the data and wrote the article. Author F.R. collected the data, analyzed the data, modified the article and edited the pictures. Author G.I. performed reference collection. All authors revised the manuscript and approved the final version. All authors are included in the author list and that all are aware that the manuscript was submitted. All authors have declared no competing interests. I confirm my personal full access to all aspects of the research and writing process, and taking final responsibility for the paper.
Funding: This research received no external funding Informed Consent Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.