The effect of birth weight on body composition: Evidence from a birth cohort and a Mendelian randomization study


Authors: Junxi Liu aff001;  Shiu Lun Au Yeung aff001;  Baoting He aff001;  Man Ki Kwok aff001;  Gabriel Matthew Leung aff001;  C. Mary Schooling aff001
Authors place of work: School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China aff001;  City University of New York Graduate School of Public Health and Health Policy, New York, New York, United States of America aff002
Published in the journal: PLoS ONE 14(9)
Category: Research Article
doi: 10.1371/journal.pone.0222141

Summary

Background

Lower birth weight is associated with diabetes although the underlying mechanisms are unclear. Muscle mass could be a modifiable link and hence a target of intervention. We assessed the associations of birth weight with muscle and fat mass observationally in a population with little socio-economic patterning of birth weight and using Mendelian randomization (MR) for validation.

Methods

In the population-representative “Children of 1997” birth cohort (n = 8,327), we used multivariable linear regression to assess the adjusted associations of birth weight (kg) with muscle mass (kg) and body fat (%) at ~17.5 years. Genetically predicted birth weight (effect size) was applied to summary genetic associations with fat-free mass and fat mass (kg) from the UK Biobank (n = ~331,000) to obtain unconfounded estimates using inverse-variance weighting.

Results

Observationally, birth weight was positively associated with muscle mass (3.29 kg per kg birth weight, 95% confidence interval (CI) 2.83 to 3.75) and body fat (1.09% per kg birth weight, 95% CI 0.54 to 1.65). Stronger associations with muscle mass were observed in boys than in girls (p for interaction 0.004). Using MR, birth weight was positively associated with fat-free mass (0.77 kg per birth weight z-score, 95% CI 0.22 to 1.33) and fat mass (0.58, 95% CI 0.01 to 1.15). No difference by sex was evident.

Conclusion

Higher birth weight increasing muscle mass may be relevant to lower birth weight increasing the risk of diabetes and suggests post-natal muscle mass as a potential target of intervention.

Keywords:

Biology and life sciences – Physiology – Physiological parameters – Body weight – Birth weight – Biochemistry – Lipids – Fats – Computational biology – Genome analysis – Genome-wide association studies – Genetics – Genomics – Genome analysis – Genome-wide association studies – Human genetics – Genome-wide association studies – Medicine and health sciences – Physiology – Physiological parameters – Body weight – Birth weight – Endocrinology – Endocrine disorders – Diabetes mellitus – Metabolic disorders – Diabetes mellitus – Women's health – Maternal health – Birth – Obstetrics and gynecology – Birth – Research and analysis methods – Research design – Cohort studies – People and places – Population groupings – Age groups – Children – Families – Children

Introduction

Observationally, lower birth weight is associated with higher risk of many chronic diseases including cardiovascular disease, diabetes and poor liver function,[14] but is also associated with lower risk of hormone-related cancers including breast and prostate cancer.[5, 6] Although these observations are open to confounding by factors such as socio-economic position (SEP), different associations by diseases suggest some of these associations may be causal. Mendelian randomization (MR) studies, taking advantages of the random allocation of genetic endowment at conception to obtain un-confounded estimates,[7] suggest an inverse association of birth weight with diabetes,[3, 4] but practical implications for prevention are unclear given birth weight is a complex phenotype. Elucidating the pathways linking birth weight with diabetes may provide additional insights into the identification of intervention targets, since birth weight is difficult to change[8] and does not have an “optimal” definition.[9]

Observationally, birth weight is positively associated with muscle mass in both teenagers and adults.[10, 11] Randomized controlled trials shows resistance training increases muscle mass and improves Hemoglobin A1c.[12] As such, muscle mass could be a modifiable downstream effect of birth weight, partially driven by sex hormones,[13, 14] potentially with sex-specific effects, consistent with the associations of lower birth weight with lower risk of breast and prostate cancers.[5, 6] However, previous observational studies assessing the role of birth weight in muscle mass sometimes adjusted for factors on the causal pathway, such as body mass index (BMI), height and physical activity, but may not fully adjusted for SEP.[15, 16]

To clarify the role of birth weight in body composition, we conducted two analyses with different assumptions and study designs (Fig 1). First, in an observational setting, we prospectively assessed the overall and sex-specific associations of birth weight with body composition (muscle mass, grip strength, and fat percentage) in a unique population, Hong Kong’s “Children of 1997” birth cohort. In Hong Kong, the usual associations of higher SEP with higher birth weight and greater gestational age are almost absent,[17] and obesity has little socio-economic patterning in young people.[18] Therefore, Hong Kong is an ideal setting to assess the associations of birth weight and gestational age with body composition. We also assessed whether these associations differed by sex given the sex-difference in body composition since such differences are likely interpretable even when associations are confounded.[19] Second, using an MR design, we validated our findings, by assessing the associations of birth weight predicted by maternal genetics independent of fetal genetics, as a proxy of maternal intrauterine environment,[20] on body composition (fat-free mass, grip strength, and fat mass) in the largest publicly available genome wide association study (GWAS).[21] Taking advantage of the random allocation of genetic endowment at conception, MR studies provide un-confounded estimates and give the result of a lifelong difference in the risk factor between groups.[7]

<h2>Directed acyclic graph of the observational analysis and the Mendelian randomization analysis.</h2>
Fig. 1.

Directed acyclic graph of the observational analysis and the Mendelian randomization analysis.


Material and methods

Ethics statement

Ethical approval for the study, including comprehensive health related analyses, was obtained from Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster (HKU/HA HKW IRB). Informed written consent was obtained from the parents/guardians, or from the participant if 18 years or older, before participation in the Biobank Clinical Follow-up.

The MR study only uses published or publicly-available data. No original data were collected for the MR study. Ethical approval for each of the studies included in the investigation can be found in the original publications (including informed consent from each participant).

Observational study—The “Children of 1997” birth cohort

“Children of 1997” is a population-representative Chinese birth cohort (n = 8327), based on 88% of births in Hong Kong in April and May 1997.[22] The original study was designed to assess the associations of second-hand smoke exposure and breastfeeding with health services utilization in the first 18 months of life. Recruitment took place at all Maternal and Child Health Centers (MCHCs) in Hong Kong. Parents are strongly encouraged to take their children to the MCHCs for free preventive care and vaccinations to age 5 years. Parental and infant characteristics were obtained at recruitment. Contact was re-established in 2007. A Biobank clinical follow-up was conducted from 2013–2016 at ~17.5 years, when body composition was assessed from bio-impedance analysis using a Tanita segmental body composition monitor (Tanita BC-545, Tanita Co., Tokyo, Japan). Grip strength was measured using a Takei T.K.K.5401 GRIP D handgrip dynamometer (Takei Scientific Instruments Co. Ltd, Tokyo, Japan).

Exposure—Birth weight, gestational age-specific birth weight z-score, and gestational age

Birth weight recorded in grams was considered in kilograms and as internally generated gestational age-specific birth weight z-scores. Gestational age recorded in days was considered in weeks. Gestational age was calculated from the actual and expected dates of delivery reported by the mothers or primary caregivers at the initial MCHCs visit. The reported expected date of delivery is based on the date of the last menstrual period and any dating scans.

Outcome—Body composition

Muscle was assessed from whole-body muscle mass (kg), and dominant hand grip strength (kg). Fat mass was assessed from body fat percentage.

Mendelian randomization study

Exposure—Genetic predictors of maternal only effects on birth weight

Single nucleotide polymorphisms (SNPs) predicting maternal effects on birth weight independent of fetal genetics (z-score transformed) at genome-wide significance (p-value<5×10−8) adjusted for gestational age where available (only available in <15% of the sample) and study-specific covariates were obtained from a GWAS consisting of two components, the Early Growth Genetics (EGG) Consortium (n = 12,319, 10 studies in the EGG consortium of European descent imputed up to the HapMap 2 reference panel, and n = 7,542, 2 studies in of European descent imputed up to the HRC panel) and the UK Biobank (n = 190,406, white European). A structural equation model was used to decompose the contributions of maternal genetic and fetal effects on birth weight (264,498 individuals own birth weight and 179,360 individuals offspring birth weight).[20]

We obtained independent SNPs (R2>0.01) with the lowest p-value using the “Clumping” function of the MR-Base (TwoSampleMR) R package, with the 1000 Genomes catalog.[23] Potentially pleiotropic effects of these SNPs were obtained from up-to-date genotype to phenotype cross-references, i.e., GWAS Catalog (https://www.ebi.ac.uk/gwas/), Ensembl (http://www.ensembl.org/index.html) and Phenoscanner (http://www.phenoscanner.medschl.cam.ac.uk/). We also checked for potential pleiotropic effects and confounding of these SNPs from the Bonferroni corrected significance (12 traits × 30 SNPs, p-value<1×10−4) of their associations with alcohol consumption (past and current), smoking (past and current), physical activity (light, moderate, and vigorous), socioeconomic position (income and education), age of voice braking, age of menarche, and height in the UK Biobank summary statistics.[21]

Outcome—Genetic associations with body composition

Genetic associations with fat-free mass (kg), grip strength (kg) (left and right hand), and fat mass (kg) were obtained from the UK Biobank (~331,000 people of genetically verified white British ancestry). The genetic associations were assessed from multivariable linear regression adjusted for the first 20 principal components, sex, age, age-squared, the sex and age interaction and the sex and age-squared interaction.[21]

Statistical analyses

Observational analyses

We compared “Children of 1997” who were included and excluded on baseline characteristics using chi-squared tests, and Cohen effect sizes[24] to obtain the magnitude of the differences between groups. Cohen effect sizes are usually categorized as 0.20 for small, 0.50 for medium and 0.80 for large for continuous variables, and as 0.10 for small, 0.30 for medium and 0.50 for large for categorical variables.

The associations of muscle mass, grip strength and fat percentage with potential confounders were assessed using independent t-tests or analysis of variance for continuous variables and chi-square tests for categorical variables. We used multivariable linear regression to obtain the observational associations of birth weight, birth weight z-score and gestational age with body composition adjusting for second-hand and maternal smoking, parental education, parental occupation, household income, type of housing, and sex. We additionally adjusted for gestational age in the association of birth weight with body composition. Sex differences were assessed from the significance of interaction terms adjusted for the other potential confounding interactions with sex.

Taking missingness into account, multiple imputation and inverse probability weighting were applied.[25] Firstly, we created 20 sets of imputed data accounting for missing confounders and exposures for all participants. Secondly, logistic regression was used to predict loss-to-follow-up based on gestational age (log-transformed because of the long tail of the distribution), second-hand and maternal smoking, sex, type of housing, type of hospital at delivery, maternal migrant status, maternal age, and, breastfeeding with the lowest Akaike information criterion value. We also used the Hosmer-Lemeshow test to check model fit. Additionally, weights were checked to ensure acceptable stability. Unstable weights indicate model misspecification.[25] Lastly, we combined each inverse probability weighting effect estimator and its corresponding sandwich variance estimator according to Rubin’s Rules.[26]

Mendelian randomization

The strength of the genetic instruments was assessed from the F-statistic, obtained using an approximation (square of SNP on exposure divided by variance of SNP on exposure).[27, 28] A higher F-statistic indicates a lower risk of weak instrument bias.[27] The effects of birth weight on the outcomes were obtained from a meta-analysis of SNP-specific Wald estimates (SNP-outcome association divided by SNP-exposure association) using inverse variance weighting with multiplicative random effects assuming balanced pleiotropy. Heterogeneity of the Wald estimates was assessed from the I2 statistic, where a high I2 may indicate the presence of invalid SNPs.[29] Differences by sex were additionally assessed.[30] Power calculations were performed using the approximation that the sample size for Mendelian randomization equates to that of the same regression analysis with the sample size divided by the r2 for genetic variant on exposure.[31]

Sensitivity analyses relevant to the observational designs

A complete case analysis was conducted as a validation without taking missingness into account.

Sensitivity analyses relevant to Mendelian randomization

As sensitivity analyses, we excluded SNPs which may be invalid. These included 1)SNPs associated with potentially pleiotropic effects on muscle or fat given in Ensembl or the GWAS Catalog; 2) SNPs associated with potential confounders and/or pleiotropic effects in the UK Biobank at Bonferroni corrected significance (p-value<1×10−4) and in PhenoScanner (p-value <1×10−5).

Estimates were obtained from sensitivity analyses with different assumptions.

Specifically, we used a weighted median which may generate correct estimates if >50% of weight is contributed by valid SNPs.[32] MR-Egger was used which generates correct estimates if all the SNPs are invalid instruments as long as the instrument strength independent of direct effect assumption is satisfied.[29] A non-null intercept from MR-Egger indicates potential directional pleiotropy and an invalid inverse variance weighting estimate.[32] The Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) was additionally used, which detects and corrects for pleiotropic outliers assuming >50% of the instruments are valid, balanced pleiotropy and the instrument strength independent of direct effect assumption are satisfied.[33, 34]

All statistical analyses were conducted using R version 3.4.2 (R Foundation for Statistical Computing, Vienna, Austria). The R packages MendelianRandomization [35] and MRPRESSO [34] were used to generate the estimates.

Results

Children of 1997

Among the originally recruited 8327 participants, 6850 are contactable and living in Hong Kong. 3460 (51%) participated in the Biobank clinical follow-up, of which 3455 had muscle mass, grip strength or fat percentage (Fig 2). The mean and standard deviation (SD) of muscle mass, grip strength and fat percentage were 42.6kg (SD 8.8kg), 25.8kg (SD 8.3kg) and 21.7% (SD 8.8%). Boys had higher muscle mass and grip strength but lower fat percentage than girls. Body composition had little association with SEP (Table 1). Differences between participants included and excluded from the study were found for gestational age, sex, second-hand and maternal smoking exposure, and SEP using chi-squared tests, but the magnitude of these differences was small (Cohen effect size <0.15) (S1 Table).

<h2>Flowchart of the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.</h2>
Fig. 2.

Flowchart of the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.


Tab. 1.

Baseline characteristics muscle mass, grip strength, and fat percentage among participants in Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.

<h2>Baseline characteristics muscle mass, grip strength, and fat percentage among participants in Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.</h2>

Observationally, birth weight and birth weight z-score were positively associated with muscle mass, grip strength, and, fat percentage. The associations were strengthened after adjusting for gestational age. Gestational age was not associated with muscle mass, grip strength or fat percentage. Associations with muscle muss differed by sex for birth weight z-score and birth weight adjusted for gestational age, with stronger associations in boys (Table 2). Similar estimates were obtained in the complete case analyses (S2 Table).

Tab. 2.

Adjusted associations of birth weight, birth weight z-score and gestational age with body composition with inverse probability weighting (IPW) and multiple imputation (MI) in the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.

<h2>Adjusted associations of birth weight, birth weight z-score and gestational age with body composition with inverse probability weighting (IPW) and multiple imputation (MI) in the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.</h2>

Mendelian randomization

Genetic instruments for maternal only effects on birth weight

Altogether, 30 SNPs independently predicted effects of maternal genetics net of infant genetics on birth weight (p-value<5×10−8) in people of European ancestry.[20] The average of SNP-specific F statistics was 79, and all were >30 (S3 Table); the variance explained (r2) was 0.013. As such, the MR study had 80% power with 5% alpha to detect a difference of 0.04 of an effect size in fat-free mass and fat mass per z-score of birth weight.

Of the 30 SNPs predicting birth weight, 5 palindromic SNPs were aligned (S3 Table); 5 SNPs had potentially pleiotropic effects, i.e., (height and metabolic response) in Ensembl or the GWAS Catalog. Of the remaining 25 SNPs, 15 remained after excluding SNPs related to height, menarche, income, and basal metabolic rate in the UK Biobank (p-value<1×10−4) and in PhenoScanner (p-value <1×10−5) (S4 and S5 Tables).

Mendelian randomization estimates

Based on all 30 SNPs, genetically predicted birth weight (maternal effects net of infant effects) was positively associated with fat-free mass, fat mass, and grip strength. No sex differences were evident. After excluding 5 potentially pleiotropic SNPs, the positive associations remained, however, the associations were not robust after additionally excluding 10 potentially pleiotropic and confounded SNPs (S5 Table). Detecting and correcting for pleiotropic outliers, MR-PRESSO indicated robust positive estimates, in particular with fat mass (Fig 3). MR-Egger had wider confidence intervals but had no indication of potential pleiotropy (S5 Table).

<h2>Mendelian randomization estimates of the effect of genetically predicted birth weight (maternal effects net of infant effects) (per z-score) on body composition with and without potentially pleiotropic SNPs and potentially confounded SNPs using MR-PRESSO.</h2>
Fig. 3.

Mendelian randomization estimates of the effect of genetically predicted birth weight (maternal effects net of infant effects) (per z-score) on body composition with and without potentially pleiotropic SNPs and potentially confounded SNPs using MR-PRESSO.


SNP = 30: all SNPs; SNP = 25, excluding maternal genotype related SNPs, and potential pleiotropic SNPs from GWAS catalog and Ensembl: rs560887 (G6PC2), rs2971669 (GCK), rs148982377 (ZNF789), rs2168101 (LMO1), rs10830963 (MTNR1B); excluding potential pleiotropic and/or confounded SNPs in UK Biobank in Bonferroni corrected significance (p-value<1×10−4) and in PhenoScanner (p-value<1×10−5): rs934232 (ZFP36L2), rs34471628 (DUSP1), rs9379084 (RREB1), rs6911024 (MICA), rs6995390 (ZFHX4), rs10814916 (GLIS3), rs111867185 (AGBL2), rs6487930 (IPO8), rs180438 (SLC38A4), rs597808 (ATXN2). MR-PRESSO: Mendelian randomization pleiotropy residual sum and outlier.

Discussion

Using two different designs, with different assumptions and data sources, we found consistent evidence that birth weight was positively associated with muscle mass (fat-free mass), grip strength and fat percentage (fat mass). These findings are consistent with previous observational studies,[10, 36, 37] but add by validating these observations in a setting with little socioeconomic patterning of birth weight and the use of MR.

These two study designs have contrasting limitations. First, residual confounding could not be ruled out in the observational design. SEP is hard to measure precisely and eliminate. In Hong Kong, the usual positive association of SEP with birth weight and gestational age is almost absent,[17] and SEP has little association with adiposity in young people.[18] However, other familial factors might affect birth weight and body composition.[38, 39] It is also difficult to disentangle correlated factors reliably in an observational study. Second, follow-up was incomplete (51%). Selection bias is unlikely, given no major difference between the participants with and without body composition indices. Moreover, differences by sex were observed, which are less open to confounding.[19] Third, MR studies have stringent assumptions, i.e., the genetic instruments should strongly predict the exposure, should not be confounded and should only be linked with the outcomes via the exposure. To examine the robustness of our findings, we excluded SNPs which may have pleiotropic effects or be associated with potential confounders, and the results were similar. MR-PRESSO also gave consistently positive sex-specific estimates after taking potential pleiotropy into account (Fig 3). Although some of the I2 were large, after excluding potentially pleiotropic and/or confounded SNPs, they became smaller. MR-Egger regression did not show directional pleiotropy even though the intercept test might be underpowered. Fourth, the overlap of the GWAS of birth weight with UK Biobank is ~90%, which might bias estimates towards the exposure-outcome association, nevertheless, the F statistic was 79 suggesting weak instrument bias is less likely.[27] Fifth, the MR study mainly pertains to people of European ancestry. However, restricting the MR study to the European ancestry could mitigate the potential confounding bias caused by hidden population structure, if the genetic associations vary by ethnic groups.[40] Ethnic differences between the MR study and the observational study is another concern, although we usually expect causal factors to act consistently across populations, unless we have evidence that the causal mechanism differs or is less relevant in some specific populations. Given the distribution of body composition varies by ethnicity, it is possible that the drivers of body composition also vary by ethnicity. However, more parsimoniously, it is likely that the drivers of body composition are similar across populations but their relevance varies. However, causes are usually consistent although not relevant in all contexts. Replicating the MR study in a Chinese population would be very helpful. Sixth, using summary statistics from different samples in the MR study means differences by age and sex could not be comprehensively assessed since no sex-specific genetic predictors of birth weight are available and hence we were only able to assess differences by sex observationally. Seventh, canalization might compensate for genetic variation in birth weight. However, whether such canalization exists is unknown. Eighth, MR provides an estimate of the effect of lifetime exposure rather than indicating the exact size of the corresponding intervention, as such it indicates an etiological pathway. Birth weight is affected by maternal and fetal genetics.[20, 41, 42] We used maternal genetics predictors net of infant genetics so the associations found with offspring body composition indicate the role of the intrauterine environment. Whether the intrauterine environment is a modifiable target of intervention, or whether subsequent consequences of the intrauterine environment would be more suitable for intervention requires investigation. Lastly, different genetic effects by generation is a concern. Given summary data was used, the genetic effects of maternal genetics net of infant genetics with offspring body composition were approximated by the genetic effects of maternal genetics net of infant genetics with maternal body composition. However, effects of genetic are likely consistent across generations.[43] We cannot rule out the possibility of the gene-environmental and/or gene-gene interactions leading to heritable epigenetic changes, which requires further exploration with individual maternal and infant genetic data.[43]

Positive associations of birth weight with body composition seem intuitive and might arise for several reasons. Development before birth is critical for skeletal muscle and adiposity. Specifically, myogenesis forms most fiber, and muscle fiber numbers do not increase after birth.[36, 44] Similarly, fat cell number is complete at birth and postnatal fat mass is mainly via increasing adipocyte size.[45, 46] Mechanisms driving differential development of muscle and fat cells before birth are unclear, but likely related to nutrition, acting via hormones. We have previously proposed that lower levels of androgens might cause higher diabetes risks via lower muscle mass.[13, 14, 47] Lower birth weight might indicate lower levels of androgens thus generating positive associations of birth weight with muscle mass and the stronger associations in men seen in both the observational and MR designs, although a difference by sex was not evident in the MR design. From an etiological perspective, a causal association of birth weight with muscle mass provides a potential mechanistic, a modifiable pathway from lower birth weight to higher diabetes risks.[3, 4, 47] Given birth weight is difficult to change, such findings suggest that muscle building might reduce diabetes risk due to lower birth weight. Such a mechanism, might also help explain a higher risk of diabetes in Asia with low prevalence of obesity, lower birth weight, and lower muscle mass than in western settings.[4852] Asians have more than double the risk of developing diabetes than Europeans at the same BMI.[48] However, it is possible that the observed associations do not extend to the extremes of the birth weight distribution, where birth weight may be a symptom of specific pathology. Given this is likely to be rare, we do not have sufficient sample size to assess this possibility. These findings are consistent with the idea of evolutionary public health, i.e., that the trade-off of growth and reproduction against longevity may inform understanding of chronic diseases and the identification of interventions.

Conclusion

Higher birth weight might increase fat-free mass and fat mass. Our study provides some indications that low fat-free mass may explain why lower birth weight increases diabetes risk and suggests muscle building as an attractive target of intervention.

Supporting information

S1 Table [docx]
Baseline characteristics of the participants who were included (n = 3455) and excluded (n = 4872) in the analyses of the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.

S2 Table [docx]
Adjusted associations of birth weight, birth weight z-score and gestational age with body composition in complete case analysis in the Hong Kong’s “Children of 1997” birth cohort, Hong Kong, China, 1997 to 2016.

S3 Table [docx]
Single nucleotide polymorphisms (SNPs) independently predicted effects of maternal genetics net of infant genetics on birth weight in Europeans from the Early Growth Genetics (EGG) Consortium (p-value<5×10).

S4 Table [docx]
Single nucleotide polymorphisms (SNPs) with potential pleiotropic effects, and/or potential confounders from Ensembl, GWAS Catalog, PhenoScanner, and UK Biobank.

S5 Table [docx]
Estimates of the effect of genetically predicted birth weight (maternal effects net of infant effects) (per z-score) on body composition with and without potentially pleiotropic single nucleotide polymorphisms (SNPs) and potentially confounded SNPs using Mendelian randomization with different methodological approaches.


Zdroje

1. Risnes KR, Vatten LJ, Baker JL, Jameson K, Sovio U, Kajantie E, et al. Birthweight and mortality in adulthood: a systematic review and meta-analysis. International journal of epidemiology. 2011;40(3):647–61. Epub 2011/02/18. doi: 10.1093/ije/dyq267 21324938.

2. Liu JX, Au Yeung SL, Kwok MK, Leung JYY, Lin SL, Hui LL, et al. Birth weight, gestational age and late adolescent liver function using twin status as instrumental variable in a Hong Kong Chinese birth cohort: "Children of 1997". Preventive medicine. 2018;111:190–7. Epub 2018/03/17. doi: 10.1016/j.ypmed.2018.03.006 29545162.

3. Wang T, Huang T, Li Y, Zheng Y, Manson JE, Hu FB, et al. Low birthweight and risk of type 2 diabetes: a Mendelian randomisation study. Diabetologia. 2016;59(9):1920–7. Epub 2016/06/24. doi: 10.1007/s00125-016-4019-z 27333884; PubMed Central PMCID: PMC4970938.

4. Zanetti D, Tikkanen E, Gustafsson S, Priest JR, Burgess S, Ingelsson E. Birthweight, Type 2 Diabetes Mellitus, and Cardiovascular Disease: Addressing the Barker Hypothesis With Mendelian Randomization. Circulation Genomic and precision medicine. 2018;11(6):e002054. Epub 2018/06/08. doi: 10.1161/CIRCGEN.117.002054 29875125.

5. Ahlgren M, Melbye M, Wohlfahrt J, Sørensen TIA. Growth Patterns and the Risk of Breast Cancer in Women. New England Journal of Medicine. 2004;351(16):1619–26. doi: 10.1056/NEJMoa040576 15483280.

6. Zhou CK, Sutcliffe S, Welsh J, Mackinnon K, Kuh D, Hardy R, et al. Is birthweight associated with total and aggressive/lethal prostate cancer risks? A systematic review and meta-analysis. British journal of cancer. 2016;114(7):839–48. Epub 2016/03/02. doi: 10.1038/bjc.2016.38 26930450; PubMed Central PMCID: PMC4955914.

7. Lawlor DA, Harbord RM, Sterne JA, Timpson N, Davey Smith G. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Statistics in medicine. 2008;27(8):1133–63. doi: 10.1002/sim.3034 17886233.

8. Mathews F, Yudkin P, Neil A. Influence of maternal nutrition on outcome of pregnancy: prospective cohort study. BMJ (Clinical research ed). 1999;319(7206):339–43. Epub 1999/08/06. doi: 10.1136/bmj.319.7206.339 10435950; PubMed Central PMCID: PMC28185.

9. Fisher D, Baird J, Payne L, Lucas P, Kleijnen J, Roberts H, et al. Are infant size and growth related to burden of disease in adulthood? A systematic review of literature. International journal of epidemiology. 2006;35(5):1196–210. Epub 2006/07/18. doi: 10.1093/ije/dyl130 16845132.

10. Chomtho S, Wells JC, Williams JE, Lucas A, Fewtrell MS. Associations between birth weight and later body composition: evidence from the 4-component model. The American journal of clinical nutrition. 2008;88(4):1040–8. Epub 2008/10/10. doi: 10.1093/ajcn/88.4.1040 18842792.

11. Yliharsila H, Kajantie E, Osmond C, Forsen T, Barker DJ, Eriksson JG. Birth size, adult body composition and muscle strength in later life. International journal of obesity (2005). 2007;31(9):1392–9. Epub 2007/03/16. doi: 10.1038/sj.ijo.0803612 17356523.

12. Strasser B, Siebert U, Schobersberger W. Resistance training in the treatment of the metabolic syndrome: a systematic review and meta-analysis of the effect of resistance training on metabolic clustering in patients with abnormal glucose metabolism. Sports medicine (Auckland, NZ). 2010;40(5):397–415. Epub 2010/05/04. doi: 10.2165/11531380-000000000-00000 20433212.

13. Schooling CM, Jiang C, Zhang W, Lam TH, Cheng KK, Leung GM. Adolescent build and diabetes: the Guangzhou Biobank Cohort Study. Annals of epidemiology. 2011;21(1):61–6. Epub 2010/12/07. doi: 10.1016/j.annepidem.2010.08.010 21130371.

14. Hou WW, Tse MA, Lam TH, Leung GM, Schooling CM. Adolescent testosterone, muscle mass and glucose metabolism: evidence from the 'Children of 1997' birth cohort in Hong Kong. Diabetic medicine: a journal of the British Diabetic Association. 2015;32(4):505–12. Epub 2014/10/14. doi: 10.1111/dme.12602 25307068.

15. Madden D. The relationship between low birth weight and socioeconomic status in Ireland. Journal of biosocial science. 2014;46(2):248–65. Epub 2013/05/02. doi: 10.1017/S0021932013000187 23631865.

16. Gigante DP, Horta BL, Matijasevich A, Mola CL, Barros AJ, Santos IS, et al. Gestational age and newborn size according to parental social mobility: an intergenerational cohort study. Journal of epidemiology and community health. 2015;69(10):944–9. Epub 2015/06/26. doi: 10.1136/jech-2014-205377 26109560; PubMed Central PMCID: PMC4602273.

17. Leung JY, Leung GM, Schooling CM. Socioeconomic disparities in preterm birth and birth weight in a non-Western developed setting: evidence from Hong Kong's 'Children of 1997' birth cohort. Journal of epidemiology and community health. 2016;70(11):1074–81. Epub 2016/05/12. doi: 10.1136/jech-2015-206668 27165846.

18. Schooling CM, Yau C, Cowling BJ, Lam TH, Leung GM. Socio-economic disparities of childhood Body Mass Index in a newly developed population: evidence from Hong Kong's 'Children of 1997' birth cohort. Archives of disease in childhood. 2010;95(6):437–43. Epub 2010/04/27. doi: 10.1136/adc.2009.168542 20418337.

19. VanderWeele TJ. Explanation in causal inference: methods for mediation: Oxford University Press; 2015.

20. Warrington NM, Beaumont RN, Horikoshi M, Day FR, Helgeland Ø, Laurin C, et al. Maternal and fetal genetic effects on birth weight and their relevance to cardio-metabolic risk factors. Nature genetics. 2019;51(5):804–14. doi: 10.1038/s41588-019-0403-1 31043758

21. Howrigan D. DETAILS AND CONSIDERATIONS OF THE UK BIOBANK GWAS: THE NEALE LAB; September 20, 2017. Available from: http://www.nealelab.is/blog/2017/9/11/details-and-considerations-of-the-uk-biobank-gwas.

22. Schooling CM, Hui LL, Ho LM, Lam TH, Leung GM. Cohort profile: 'children of 1997': a Hong Kong Chinese birth cohort. Int J Epidemiol. 2012;41(3):611–20. doi: 10.1093/ije/dyq243 21224275.

23. Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. eLife. 2018;7:e34408. doi: 10.7554/eLife.34408 29846171

24. Cohen J. Statistical power analysis for the behavioral sciences: Academic Press; 1977.

25. Seaman SR, White IR. Review of inverse probability weighting for dealing with missing data. Statistical methods in medical research. 2013;22(3):278–95. Epub 2011/01/12. doi: 10.1177/0962280210395740 21220355.

26. Seaman SR, White IR, Copas AJ, Li L. Combining Multiple Imputation and Inverse-Probability Weighting. Biometrics. 2012;68(1):129–37. doi: 10.1111/j.1541-0420.2011.01666.x PMC3412287. 22050039

27. Burgess S, Davies NM, Thompson SG. Bias due to participant overlap in two‐sample Mendelian randomization. Genetic Epidemiology. 2016;40(7):597–608. doi: 10.1002/gepi.21998 PMC5082560. 27625185

28. Bowden J, Del Greco MF, Minelli C, Davey Smith G, Sheehan NA, Thompson JR. Assessing the suitability of summary data for two-sample Mendelian randomization analyses using MR-Egger regression: the role of the I2 statistic. International journal of epidemiology. 2016;45(6):1961–74. Epub 2016/09/13. doi: 10.1093/ije/dyw220 27616674; PubMed Central PMCID: PMC5446088.

29. Burgess S, Bowden J., Fall T., Ingelsson E., Thompson S. G. Sensitivity analyses for robust causal inference from Mendelian randomization analyses with multiple genetic variants. Epidemiology (Cambridge, Mass). 2016.

30. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ (Clinical research ed). 2003;326(7382):219. Epub 2003/01/25. doi: 10.1136/bmj.326.7382.219 12543843; PubMed Central PMCID: PMC1125071.

31. Freeman G, Cowling BJ, Schooling CM. Power and sample size calculations for Mendelian randomization studies using one genetic instrument. Int J Epidemiol. 2013;42(4):1157–63. Epub 2013/08/13. doi: 10.1093/ije/dyt110 23934314.

32. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. 2016;40(4):304–14. Epub 2016/04/12. doi: 10.1002/gepi.21965 27061298; PubMed Central PMCID: PMC4849733.

33. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. International journal of epidemiology. 2015;44(2):512–25. Epub 2015/06/08. doi: 10.1093/ije/dyv080 26050253; PubMed Central PMCID: PMC4469799.

34. Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nature genetics. 2018;50(5):693–8. doi: 10.1038/s41588-018-0099-7 29686387

35. Yavorska OO, Burgess S. MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. International journal of epidemiology. 2017;46(6):1734–9. Epub 2017/04/12. doi: 10.1093/ije/dyx034 28398548; PubMed Central PMCID: PMC5510723.

36. Dodds R, Denison HJ, Ntani G, Cooper R, Cooper C, Sayer AA, et al. Birth weight and muscle strength: A systematic review and meta-analysis. The journal of nutrition, health & aging. 2012;16(7):609–15. doi: 10.1007/s12603-012-0053-9 22836701

37. Simpson J, Smith AD, Fraser A, Sattar N, Lindsay RS, Ring SM, et al. Programming of Adiposity in Childhood and Adolescence: Associations With Birth Weight and Cord Blood Adipokines. The Journal of clinical endocrinology and metabolism. 2017;102(2):499–506. Epub 2016/11/15. doi: 10.1210/jc.2016-2342 27841944; PubMed Central PMCID: PMC5413167.

38. Patro B, Liber A, Zalewski B, Poston L, Szajewska H, Koletzko B. Maternal and paternal body mass index and offspring obesity: a systematic review. Annals of nutrition & metabolism. 2013;63(1–2):32–41. Epub 2013/07/28. doi: 10.1159/000350313 23887153.

39. Tyrrell J, Richmond RC, Palmer TM, et al. Genetic evidence for causal relationships between maternal obesity-related traits and birth weight. Jama. 2016;315(11):1129–40. doi: 10.1001/jama.2016.1975 26978208

40. Davies NM, Holmes MV, Davey Smith G. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ (Clinical research ed). 2018;362:k601. doi: 10.1136/bmj.k601 30002074

41. Evans DM, Moen G-H, Hwang L-D, Lawlor DA, Warrington NM. Elucidating the role of maternal environmental exposures on offspring health and disease using two-sample Mendelian randomization. International journal of epidemiology. 2019;48(3):861–75. doi: 10.1093/ije/dyz019 30815700

42. Lawlor D, Richmond R, Warrington N, McMahon G, Smith G, Bowden J, et al. Using Mendelian randomization to determine causal effects of maternal pregnancy (intrauterine) exposures on offspring outcomes: Sources of bias and methods for assessing them [version 1; peer review: 4 approved]. Wellcome Open Research. 2017;2(11). doi: 10.12688/wellcomeopenres.10567.1 28405635

43. Nelson VR, Nadeau JH. Transgenerational genetic effects. Epigenomics. 2010;2(6):797–806. doi: 10.2217/epi.10.57 22122083.

44. Du M, Yan X, Tong JF, Zhao J, Zhu MJ. Maternal obesity, inflammation, and fetal skeletal muscle development. Biology of reproduction. 2010;82(1):4–12. Epub 2009/06/12. doi: 10.1095/biolreprod.109.077099 19516021; PubMed Central PMCID: PMC2802110.

45. Chiavaroli V, Derraik JG, Hofman PL, Cutfield WS. Born Large for Gestational Age: Bigger Is Not Always Better. The Journal of pediatrics. 2016;170:307–11. Epub 2015/12/29. doi: 10.1016/j.jpeds.2015.11.043 26707580.

46. Spalding KL, Arner E, Westermark PO, Bernard S, Buchholz BA, Bergmann O, et al. Dynamics of fat cell turnover in humans. Nature. 2008;453(7196):783–7. Epub 2008/05/06. doi: 10.1038/nature06902 18454136.

47. Yeung CHC, Au Yeung SL, Fong SSM, Schooling CM. Lean mass, grip strength and risk of type 2 diabetes: a bi-directional Mendelian randomisation study. Diabetologia. 2019. Epub 2019/02/25. doi: 10.1007/s00125-019-4826-0 30798333.

48. Pan WH, Flegal KM, Chang HY, Yeh WT, Yeh CJ, Lee WC. Body mass index and obesity-related metabolic disorders in Taiwanese and US whites and blacks: implications for definitions of overweight and obesity for Asians. The American journal of clinical nutrition. 2004;79(1):31–9. Epub 2003/12/20. doi: 10.1093/ajcn/79.1.31 14684394.

49. Spanakis EK, Golden SH. Race/Ethnic Difference in Diabetes and Diabetic Complications. Current diabetes reports. 2013;13(6):10.1007/s11892-013-0421-9. doi: 10.1007/s11892-013-0421-9 PMC3830901. 24037313

50. Madan A, Holland S, Humbert JE, Benitz WE. Racial Differences in Birth Weight of Term Infants in a Northern California Population. Journal Of Perinatology. 2002;22:230. doi: 10.1038/sj.jp.7210703 11948387

51. Silva AM, Shen W, Heo M, Gallagher D, Wang Z, Sardinha LB, et al. Ethnicity-Related Skeletal Muscle Differences Across the Lifespan. American journal of human biology: the official journal of the Human Biology Council. 2010;22(1):76–82. doi: 10.1002/ajhb.20956 PMC2795070. 19533617

52. Lear SA, Kohli S, Bondy GP, Tchernof A, Sniderman AD. Ethnic Variation in Fat and Lean Body Mass and the Association with Insulin Resistance. The Journal of Clinical Endocrinology & Metabolism. 2009;94(12):4696–702. doi: 10.1210/jc.2009-1030 19820012


Článek vyšel v časopise

PLOS One


2019 Číslo 9

Nejčtenější v tomto čísle

Tomuto tématu se dále věnují…


Kurzy

Zvyšte si kvalifikaci online z pohodlí domova

Léčba bolesti v ordinaci praktického lékaře
nový kurz
Autoři: MUDr. PhDr. Zdeňka Nováková, Ph.D.

Revmatoidní artritida: včas a k cíli
Autoři: MUDr. Heřman Mann

Jistoty a nástrahy antikoagulační léčby aneb kardiolog - neurolog - farmakolog - nefrolog - právník diskutují
Autoři: doc. MUDr. Štěpán Havránek, Ph.D., prof. MUDr. Roman Herzig, Ph.D., doc. MUDr. Karel Urbánek, Ph.D., prim. MUDr. Jan Vachek, MUDr. et Mgr. Jolana Těšínová, Ph.D.

Léčba akutní pooperační bolesti
Autoři: doc. MUDr. Jiří Málek, CSc.

Nové antipsychotikum kariprazin v léčbě schizofrenie
Autoři: prof. MUDr. Cyril Höschl, DrSc., FRCPsych.

Všechny kurzy
Kurzy