International Journal of Obesity

Overweight and obesity are increasingly common public health issues worldwide, leading to a wide range of diseases from metabolic syndrome to steatohepatitis and cardiovascular diseases. While the increase in the prevalence of obesity is partly attributable to changes in lifestyle (i.e. increased sedentarity and changes in eating behaviour), the metabolic and clinical impacts of these obesogenic conditions varies between sexes and genetic backgrounds. The conception of personalised treatments of obesity and its complications require a thorough understanding of the diversity of responses to conditions such as high-fat diet intake. By analysing nine genetically diverse mouse strains, we show that much like humans, mice respond to high-fat diet in a genetic- and sex-dependent manner. Physiological and molecular responses to high-fat diet are associated with expression of genes involved in immunity and mitochondrial function. Finally, we find that mitochondrial function may explain part of the diversity of physiological responses. By exploring the complex interactions between genetics and metabolic phenotypes via gene expression and molecular traits, we shed light on the importance of genetic background and sex in determining metabolic outcomes. In addition to providing the community with an extensive resource for optimizing future experiments, our work serves as an exemplary design for more generalizable translational studies.

The continuing increase in many countries in adult body mass index (BMI kg/m2) and its dispersion is contributed to by interaction between genetic susceptibilities and an increasingly obesogenic environment (OE). The determinants of OE-susceptibility are unresolved, due to uncertainty around relevant genetic and environmental architecture. We aimed to test the multi-modal distributional predictions of a Mendelian genetic architecture based on collectively common, but individually rare, large-effect variants and their ability to account for current trends in a large population-based sample. We studied publicly available adult BMI data (n = 9102) from 3 cycles of NHANES (1999, 2005, 2013). A first degree family history of diabetes served as a binary marker (FH0/FH1) of genetic obesity susceptibility. We tested for multi-modal BMI distributions non-parametrically using kernel-smoothing and conditional quantile regression (CQR), obtained parametric fits to a Mendelian model in FH1, and estimated FH x OE interactions in CQR models and ANCOVA models incorporating secular time. Non-parametric distributional analyses were consistent with multi-modality and fits to a Mendelian model in FH1 reliably identified 3 modes. Mode separation accounted for ~40% of BMI variance in FH1 providing a lower bound for the contribution of large effects. CQR identified strong FH x OE interactions and FH1 accounted for ~60% of the secular trends in BMI and its SD in ANCOVA models. Multimodality in the FH effect is inconsistent with a predominantly polygenic, small effect architecture and we conclude that large genetic effects interacting with OE provide a better quantitative explanation for current trends in BMI.

Many weight-height (Wt-Ht) models were proposed since the derivation of the "Quetelet Index", though their credibility remains elusive. This raises an interesting question about the presence of numerous Wt-Ht models and reveals that our knowledge about the Wt-Ht correlation is still far from its essence. We identified a strong linear correlation between Wt and Wt*Htc (c=-2[~]2), based on data from 359,049 participants aged 1 to 21 years, recruited from China, Japan, South Korea, Slovakia, the USA, and Bangladesh. Then, we established a statistically-robust Wt-Ht model, which is expressed as Wt=a+b*Wt*Htc (Eq. 1) or 1=a/Wt+b*Htc (Eq. 2), through linear regression. When c=1 or -1, Eq. 2 is a standard hyperbolic function, which proves that Wt is hyperbolically correlated with Ht. The coefficients a and b are sex-, age-, and geography-specific constants. As the exponent c approaches 0, the correlation between Wt and Wt*Htc, along with the standardized Wt-Ht index (sWHI) (a/Wt+b*Htc in Eq. 2), approaches 1. Further, we incorporated total fat to determine the implications of our model in obesity diagnosis. When c=1, the sWHI (a/Wt+b*Ht) is capable of screening abnormal body fat percentage (BFP) based on deviations from the Wt-Ht equilibrium (sWHI=1). BMI demonstrates advantages over other anthropometric indexes in screening abnormal BFP; nonetheless, their performances are largely similar. This is probably attributed to the strong linear correlation between Wt and Wt*Htc. Overall, the hyperbolic Wt-Ht model reveals the nature of Wt-Ht correlation and offers key insights into obesity diagnosis using anthropometric indexes.

BackgroundExposure to adverse childhood experiences (ACEs) is associated with a 30 to 50% increased risk of obesity in adolescence. The role of ACEs as a determinant of weight loss among overweight and obese children remains unclear. MethodsAmong 8568 nine-year-old children randomly sampled in 2007/2008 for the Growing up in Ireland cohort, 2210 were overweight or obese at 9 years and provided complete follow-up data at age 13 and 18 years. Structural equation and natural effects mediation models tested for a direct causal relationship between ACEs before 9 years and remission risk at 18 years, and indirect effects mediated via daily activity, diet quality, self-image and behavioural difficulties and BMI at 9 years. ResultsAmong the 1676 adolescents that were overweight or obese at age 9, 46% achieved healthy weight status by age 18; 56% (n=618) of overweight children and 27% (n=153) of obese children), 13% experienced an ACE, and 41% were female. Exposure to an ACE was associated with a higher BMI Z at ages 9 (0.47 vs 0.36, p < 0.05) and 13 years (0.39 vs 0.29, p < 0.05) and reduced the odds of achieving healthy weight status at age 18 by 27% (OR: 0.73, 95% CI:0.54- 0.99). Overweight and obese children exposed to an ACE had lower household income, higher behavioural difficulties, and lower self-concept at ages 9, 13 and 18. Children exposed to an ACE were also 2 to 3-fold more likely to have started smoking before 12 years old and 50% more likely to smoke more than 10 cigarettes a week regularly vape. Behavioural difficulties, self-concept and baseline weight status, but not smoking or dietary habits, mediated the association between ACE exposure and achieving healthy weight at age 18. ConclusionAmong overweight and obese children, exposure to ACEs indirectly reduces the likelihood of achieving healthy weight status at age 18, mediated by its effects on weight in childhood, and behavioural difficulties and self-concept in mid-adolescence. These findings highlight the complex factors that influence weight loss among overweight and obese children exposed to adverse experiences early in life.

BackgroundMost research on obesity trends and projections has focused on changes across calendar years. However, as the risk of disease increases with cumulative exposure to obesity, it is crucial to characterize the obesity landscape through a life-course perspective and across birth cohorts. ObjectiveTo enhance the accuracy of obesity epidemic projections throughout life course and across birth cohorts in the US. DesignCross-sectional and cohort study. SettingUnited States ParticipantsIndividuals participated in three National Health Examination Surveys (NHES) from 1959 to 1970 and 18 National Health and Nutrition Examination Surveys (NHANES) from 1971 to 2020. MeasurementsBody mass index (BMI) distributions by sex, race, and birth cohort. ResultsBy leveraging over 40 years of cross-sectional and longitudinal data from nationally representative surveys, we developed models to estimate historical and future BMI distributions in the US for both children and adults throughout their life course. We also calculated life-years of exposure to overweight and obesity, according to sex, race, and birth cohort. Our findings reveal significant increases in these metrics among birth cohorts since 1965 and highlight differential trends by sex and race for the 1965, 1985, and 2005 cohorts. LimitationsAssumption that model parameters will hold in the future. ConclusionOur approach significantly expands upon previous models by projecting life course with continuous BMI distributions informed by longitudinal trajectories, explicitly accounting for variations in birth cohorts. Primary Funding SourceNational Institutes of Health.

Background: Ethnic minorities and socioeconomically disadvantaged populations in the UK are at increased risk of obesity. We modelled longitudinal body mass index (BMI) trajectories through infancy, childhood, and adolescence to identify at-risk groups and modifiable risk factors. Methods: This cohort sampled 10,350 White and South Asian children born in Leicestershire, 1985-1997. We included 5,571 participants with [&ge;]3 BMI measurements between 0-18 years collected from healthcare records, questionnaires, and study visits. We used Group-Based Trajectory Modelling of BMI, separately by sex and ethnicity, and combined. We identified at-risk groups and modifiable risk factors using multinomial logistic regression, with inverse probability weighting to reduce selection bias. Results: We identified similar five BMI trajectories across sex and ethnicity: stable normal BMI (47%); persistent low BMI (30%); early overweight resolving (8%); childhood onset obesity (4%); and adolescent onset overweight (11%). Childhood onset obesity deviated from stable normal BMI at 2-4 years of age, adolescent onset overweight at 4-6 years. South Asians were at higher risk of childhood onset obesity (aOR: 1.66 [95%CI 1.08-2.53]) and adolescent onset overweight (1.29 [0.98-1.71]) than Whites. Children from deprived backgrounds (1.66 [0.92-2.82], most vs least deprived quintile) and those with less educated parents (1.67 [1.08-2.63], compulsory vs higher education) were at increased risk of childhood onset obesity. Smoking during pregnancy (1.50 [0.88-2.54]) and absence of breastfeeding (1.56 [1.07-2.29]) increased risk of childhood onset obesity. Physical activity decreased risk of childhood onset obesity (0.64 [0.44-0.93], [&ge;]4 vs 0-3 hours/week) and adolescent onset overweight (0.75 [0.59-0.94]). Conclusion: BMI trajectories diverge as early as age 2 years, revealing ethnic and social inequalities. Obesity strategies in the UK should intervene during critical windows in early life and prioritise South Asian children and those from socioeconomically deprived backgrounds.

BackgroundAdverse early life exposures might negatively affect foetal and infant development, predisposing children to obesity. We aimed to systematically identify and evaluate risk factors for childhood obesity in preconception, pregnancy, and infancy, and assess their potential as targets for future prediction and prevention strategies. MethodsThis systematic review (PROSPERO, CRD42022355152) included longitudinal studies from selected electronic databases published between inception and August 17th, 2022, identifying maternal, paternal, or infant risk factors from preconception until infancy for childhood obesity between 2 and 18 years. Screening and data extraction was performed through standardized extraction forms. We assessed risk factor quality on modifiability and predictive power using a piloted criteria template from ILSI-Europe-Marker-Validation-Initiative. FindingsWe identified 172 observational and 5 intervention studies involving n=1.879.971 children from 37, predominantly high-income, countries. 59%, 25% and 16% of studies measured childhood obesity between 2- <6 years, 6-10 years, and >10 -18 years respectively. Average reported childhood obesity prevalence was 11{middle dot}1%. Pregnancy and infancy risk factors were mostly studied. We identified 59 potential risk factors, 24 of which were consistently associated with childhood obesity risk. Higher maternal prepregnancy weight (n=28/31 positive associations from 31 studies, respectively), higher gestational weight gain (n=18/21), maternal smoking during pregnancy (n=23/29), higher birth weight (n=20/28), LGA (n=17/18), no breastfeeding (n=20/31), and higher infant weight gain (n=12/12) were the strongest risk factors, which may aid in prediction or be targets for prevention. Level of evidence was generally moderate due to unreliable exposure measurement, short follow-up/loss-to-follow up, and risk of confounding. InterpretationWe identified 7 early life risk factors, which were strongly associated with a higher risk of childhood obesity, and can contribute to future prediction and prevention strategies. These findings support implementation of prevention strategies targeting these early-life risk factors from a clinical and population perspective, where possible integrated with implementation studies. FundingThis work was conducted by an expert group of the European branch of the International Life Sciences Institute, ILSI Europe.

ObjectiveBMI alone does not capture obesity-related health heterogeneity. The Edmonton Obesity Staging System (EOSS) grades obesity severity based on comorbidities and functional impairment, whereas the Lancet Commission Diagnostic Model for Obesity (DMO) distinguishes preclinical from clinical obesity based on organ dysfunction. We assessed whether both frameworks identify overlapping phenotypes and how they classify obesity severity. MethodsA modified EOSS and DMO were applied to the UK Biobank (N {approx} 411,000). Stage distributions, cross-classification, and the impact of combining BMI with fat distribution on obesity categorization were analyzed. ResultsAbout one quarter of participants were classified with obesity under both frameworks. Most were assigned to advanced stages, with high concordance for established disease. Differences were most pronounced in early stages: DMO captured a broader spectrum of mild/subclinical organ dysfunction, whereas EOSS emphasized established disease with prognostic relevance. Discrepancies reflected differences in operationalization of e.g. metabolic, cardiovascular, and mental health. Obesity thresholds influenced classification, with [~]50% reclassified when BMI was combined with different fat distribution parameters, highlighting sensitivity of early-stage assignment. ConclusionEOSS and DMO provide complementary perspectives on obesity severity. Integrating EOSSs prognostic granularity with DMOs multidimensional approach may improve risk stratification and identify individuals most suitable for intensive interventions. STUDY IMPORTANCEO_ST_ABSWhat is already known?C_ST_ABSO_LIBMI alone poorly reflects obesity-related health risk; comorbidities, organ dysfunction, and functional impairments are crucial for precise staging. C_LIO_LITwo major frameworks exist: EOSS focuses on prognostic severity, while DMO identifies early/preclinical obesity--but their agreement and clinical implications were unclear. C_LI What does this study add?O_LIDemonstrates that EOSS emphasizes established disease and prognostic severity, whereas DMO captures a broader spectrum of early or subclinical organ dysfunction, revealing distinct phenotypes within the same BMI-defined population. C_LIO_LIHighlights that combining BMI with anthropometric measures can reclassify up to [~]50% of individuals, illustrating the sensitivity of early-stage assignment to diagnostic thresholds. C_LI How might these results change the direction of research or the focus of clinical practice?O_LIIntegrating EOSSs prognostic detail with DMOs broad, multidimensional approach enables targeted intervention, helping clinicians prioritize patients for intensive obesity management or treatment. C_LIO_LIProvides evidence for harmonizing obesity classification beyond BMI, emphasizing the need for multidimensional assessment in both research cohorts and routine clinical practice. C_LI

BackgroundFat distribution patterns, rather than total adiposity alone, critically influence the risk of obesity-related diseases. However, due to correlations between fat accumulation in different regions, the contribution of regional fat to metabolic and non-metabolic diseases remains unclear. MethodsUsing UK Biobank MRI data (N = 23,548) and a Japanese cohort (N = 642), we used archetype analysis to identify patterns of predominantly isolated fat accumulation in the liver, pancreas, visceral adipose tissue, and thigh muscle. We then characterized the type 2 diabetes (T2D) risk, biomarker profiles, and broad health burden associated with isolated fat accumulation. FindingsWe identified four distinct patterns of isolated fat accumulation in the liver, thigh muscle, pancreas, and visceral adipose tissue and replicated these patterns in the Japanese cohort. Organ-specific fat accumulation was associated with an equal or greater T2D burden than isolated visceral adiposity, despite lower BMI and visceral fat amount. Moreover, specific fat depots were linked to distinct comorbidities not observed with visceral fat alone, including knee osteoarthritis (thigh myosteatosis), COPD (pancreatic steatosis) and breast cancer (hepatic steatosis). In contrast, total and visceral fat alone were not significantly associated with many of these complications. InterpretationThese findings highlight the central role of organ-specific fat accumulation beyond general adiposity in obesity-related diseases, offering new insight into the heterogeneity of obesity. FundingEuropean Research Council, European Union, and the Japanese Society for the Promotion of Science.

ObjectiveEvaluate the impact of bariatric surgery on monthly earnings and employee status among working-age adults, and examine variations across sociodemographic characteristics. DesignRetrospective longitudinal cohort study using national, linked administrative datasets. SettingHospital inpatient services in England between 1 April 2014 and 31 December 2022. Participants40,662 individuals who had a bariatric surgery procedure and obesity diagnosis during the study period, with no bariatric surgery history in the previous 5 years, and were 25 to 64 years old at the date of surgery. We also included 49,921 individuals sampled from the general population who had not had bariatric surgery matched by age and sex to those in the cohort who had bariatric surgery. Main outcome measuresMonthly employee pay - for all months and only months where the individual was in paid employment - expressed in 2023 prices; paid employee status. ResultsAmong people living with obesity who had bariatric surgery, there was a sustained increase in monthly employee pay from six months after surgery with a mean increase of {pound}84 per month 5 years after surgery compared with the six months before surgery. Among those in paid employment, there was a sustained increase in the probability of being a paid employee from 4 months after bariatric surgery, with a mean increase of 4.3 percentage points 5 years after surgery. The increases in pay and probability of employment were greater for males. The increase in employee pay was not sustained over the 5-year follow up time for the youngest age groups. ConclusionsBariatric surgery is associated with an increased probability of being employed, resulting in increased earnings. These findings suggest that living with obesity negatively impacts labour market outcomes and that obesity management interventions are likely to generate economic benefits both to individuals and on a macroeconomic level by increasing the likelihood of employment of people living with obesity.

BackgroundChildhood with obesity is characterized by metabolic dysregulation and unique gut microbiota profiles. Nevertheless, the comprehensive understanding of gut microbiota and metabolic dysregulation of Childhood with obesity remains unclear. ObjectivesThis study aimed to investigate the causal relationship of gut microbiota and Childhood with obesity and identify the blood metabolites as potential mediators. MethodsThe exposure genome-wide association studies (GWAS) data were sourced from the GWAS Catalog, while the outcome GWAS data were obtained from the Early Growth Genetics (EGG) Consortium. The study used 473 types of gut microbiota, 233 types of blood metabolites, and Childhood with obesity from GWAS. We then performed two-sample Mendelian randomization (TSMR) and bidirectional Mendelian randomization (BDMR) analyses to explore the causal relationships between gut microbiota, blood metabolites, and Childhood with obesity. Additionally, we conducted multivariable Mendelian randomization (MVMR) and two-step Mendelian randomization (2SMR) to identify potential mediating blood metabolites in this process. ResultsMR analysis identified 13 types of gut microbiota and 12 types of blood metabolites that were causally associated with Childhood with obesity. Furthermore, there was no strong evidence that genetically predicted Childhood with obesity had an effect on these gut microbiota and blood metabolites. Further, 2SMR analysis revealed that the association between K10 sp001941205 and Childhood with obesity was mediated by the Total cholesterol to total lipids ratio in medium VLDL, accounting for 2.53% (95%CI; 2.14%-2.92%) of the association. Similarly, the relationship between SM23-33 and Childhood with obesity was mediated by the Ratio of 22:6 docosahexaenoic acid to total fatty acids, which accounted for 4.07% (95%CI; 2.70%-5.44%) of the association. ConclusionsThe present study is the first to investigate the causal relationships among 473 gut microbiota phenotypes, 233 blood metabolites, and Childhood with obesity through Mendelian randomization analysis, identifying 13 gut microbiota types with potential causal links to Childhood with obesity and suggesting that 2 blood metabolites may mediate these associations, thereby providing valuable insights for future intervention strategies aimed at addressing Childhood with obesity.

Inflammation relates to decreased effortful behaviour and altered effort-related responses in brain regions, such as the dorsomedial prefrontal cortex (dmPFC). Inflammation is prevalent in obesity, but its effects on effortful versus more convenient food choices are unknown. We investigated the role of low-grade inflammation in effortful food choice using functional MRI in a cross-sectional study (n=150 women, BMI>27 kg/m{superscript 2}) and a 12-week randomized controlled trial (n=59 women, BMI>30 kg/m{superscript 2}, CRP>3 mg/L; colchicine 0.5 mg/d vs. placebo). Inflammation was related to less high-effort choices (OR=0.27, p=0.004) and higher effort-related dmPFC signal ({beta}=0.23, p=0.025, Rpartial{superscript 2}=0.039), and mediated the association between BMI and dmPFC signal. Colchicine decreased systemic inflammation (i.e. INFLA-score) as expected ({beta}=-0.10 SD, p<0.001, Rpartial{superscript 2}=0.030), increased high-effort choices (OR=1.32, p=0.044), and marginally decreased effort-related dmPFC signal ({beta}=-0.13, p=0.053, Rpartial{superscript 2}=0.037). These findings show a causal role for inflammation in choosing convenience foods in obesity via increased effort aversion and associated dmPFC processing.

Background and AimSocioeconomically disadvantaged people are more likely to have high body mass index (BMI). However whether socioeconomic position moderates genetic susceptibility to high BMI, and whether this effect differs by sex, remains unclear. We aimed to investigate whether educational attainment (EA) moderates the association between genetic liability for high BMI and BMI trajectories across adulthood in females and males. MethodsWe analyzed data from 69,314 participants in the Trondelag Health Study (HUNT), a population-based cohort with genotyping and repeated BMI measures. A polygenic index for BMI (BMI PGI) was calculated, and participants were categorized by EA level. Using linear mixed-effects models stratified by sex, we tested the interaction between BMI PGI, EA, and age on BMI trajectories. ResultsThe relationship between BMI PGI and BMI was non-linear, showing a steeper slope in the upper deciles, and was modified by sex (p<0.001). Sex-stratified analysis showed that EA moderated the effect of BMI PGI on BMI in females (p=0.003) but not in males (p=0.089). Among highly educated females, the BMI difference between the top and bottom BMI PGI deciles was-0.99 kg/m{superscript 2} [95%CI: -1.67 to -0.30] smaller than among those with low education. In males, the corresponding difference was -0.16 kg/m{superscript 2} [-0.71 to 0.39]. Genetic influences on BMI trajectories showed consistent age-dependent patterns across all educational groups, though trajectories differed by sex. Females experienced a steady increase in BMI until age 60, after which it declined. Males had an early rapid increase, then stabilization, followed by a slight late-life decline. ConclusionHigher EA consistently moderates the effect of genetic liability for high BMI in females throughout adulthood, but this protective effect is absent in males. This sex difference suggests that gender-related socioeconomic factors may modulate the expression of BMI-related genetic variants, warranting further investigation into the underlying mechanisms.

Obesity, a leading cause of several metabolic abnormalities, is mainly due to an imbalance of energy homeostasis. IRX3 and IRX5 have been suggested as determinants of obesity in connection with the intronic variants of FTO, the strongest genetic risk factor of polygenic obesity in humans. Although the causal effects of Irx3 on obesity and its related metabolic consequences have been demonstrated in vivo, the metabolic function of Irx5 remains unclear. In this study, using mice homozygous for an Irx5-knockout (Irx5KO) allele, we show a direct link between Irx5 expression and regulation of body mass/composition and energy homeostasis. Irx5KO mice are leaner and resistant to diet-induced obesity and associated metabolic abnormalities, primarily through the loss of adiposity with an increase in basal metabolic rate with adipose thermogenesis and lower food intake. Furthermore, our long-term feeding analysis found that Irx3 mutant mouse lines also have less food intake, indicating that lower caloric intake also contributes to their lean phenotype. Together, these results demonstrate that Irx5 is critical for energy homeostasis and regulation of body mass/composition and suggest that it likely acts in other tissues beyond adipocytes.

BackgroundThough viewed as problematic for measuring individual-level adiposity, Body Mass Index (BMI) is often considered "good enough" for population inference and epidemiological research. However, we demonstrate that BMI produces statistically invalid population-level estimates of associations between key demographic risk factors (e.g., self-reported sex, race, age) and obesity when compared to more direct adiposity measurements. Further, we demonstrate how novel statistical calibration techniques can enable more valid population inference using widely available BMI data alongside a limited subset of "gold standard" measurements. MethodsUsing National Health and Nutrition Examination Survey data (2011-2023), we compare associations, broken down by demographic groups, across three different purported adiposity measures: BMI, Waist Circumference (WC), and whole-body total fat percentage from Dual-energy X-ray absorptiometry (DXA) scans. We then apply a statistical procedure for conducting inference on predicted data to calibrate BMI-based prevalence estimates toward the "gold standard" DXA-based measurements, allowing for valid population inference even for time periods where only BMI data are available. FindingsBMI-measured adiposity yields substantially different - and often contradictory - conclusions about the association between obesity status and lifestyle factors compared to the more direct, DXA-based measurements. Most concerning, the directions and magnitudes of the associations between racial groups may differ depending on whether BMI or DXA-based measurements are used. Similarly, self-reported sex-based differences in obesity prevalence show opposite patterns across measurement types. Our validation results confirm that our calibration method overcomes this challenge and successfully approximates DXA-based associations using primarily BMI-based measurements. InterpretationOur study provides empirical evidence that uncorrected BMI-based inference leads to invalid population-level estimates about the associations between obesity status and key predictors. The statistical calibration approach we present offers a practical solution for obesity researchers who must rely on BMI or similar anthropometric measures due to cost or data availability constraints, enabling more valid population inference without requiring comprehensive "gold standard" adiposity measurements.

BackgroundDiabetes, cardiovascular disease, and related cardiometabolic disturbances are increasing rapidly in the Asia-Pacific region. We investigated the contribution of excess adiposity, a key determinant of diabetes and cardiovascular risk, to unfavourable cardiometabolic profiles amongst Asian ethnic subgroups. MethodsThe Health for Life in Singapore (HELIOS) Study is a population-based cohort comprising multi-ethnic Asian men and women living in Singapore, aged 30-84 years. We analyzed data from 9,067 participants who had assessment of body composition by Dual X-Ray Absorptiometry (DEXA) and metabolic characterization. We tested the relationship of BMI and visceral Fat Mass Index (vFMI) on cardiometabolic phenotypes (glycemic indices, lipid levels, and blood pressure), disease outcomes (diabetes, hypercholesterolemia, and hypertension), and metabolic syndrome score with multivariate regression analyses. FindingsParticipants were 59.6% female, with mean (SD) age 52.8 (11.8) years. The prevalence of diabetes, hypercholesterolemia, and hypertension was 8.3%, 29% and 18.0%, respectively. Malay and Indian participants had 3-4 folds higher odds of obesity and diabetes, and showed adverse metabolic and adiposity profiles, compared to Chinese participants. Excess adiposity contributed to all adverse cardiometabolic health indices including diabetes (P<0.001). However, while vFMI explained the differences in triglycerides and blood pressure between the Asian ethnic groups, increased vFMI did not explain higher glucose levels, reduced insulin sensitivity and risk of diabetes amongst Indian participants. InterpretationVisceral adiposity is an independent risk factor for metabolic disease in Asian populations, and accounts for a large fraction of diabetes cases in each of the ethnic groups studied. However, the variation in insulin resistance and diabetes risk between Asian subgroups is not consistently explained by adiposity, indicating an important role for additional mechanisms underlying the susceptibility to cardiometabolic disease in Asian populations. FundingNanyang Technological University--the Lee Kong Chian School of Medicine, National Healthcare Group, National Medical Research Council, Singapore. Research in context Evidence before this studyWe searched Embase and MEDLINE using MeSH terms and respective alternative terms for ["body fat distribution" OR "visceral adiposity" OR "diagnostic imaging"] and ["metabolic syndrome" OR "diabetes mellitus" OR "hypertension" OR "hyperlipidemia" OR all corresponding phenotypes] from 1946 till 7th August 2023 and identified 456 relevant studies. Overall, there have been substantial attempts to characterize the impact of adiposity quantified with imaging techniques on cardiometabolic health. However, most works focused on validating novel adiposity indices (such as body shape index) or metabolic biomarkers (such as cytokines), and rarely provided insights on the contribution of excess visceral adiposity across cardiometabolic phenotypes. Some investigations focused on delineating the effect of various fat depots in the viscera on insulin resistance. Very few studies evaluated health disparity across populations; Nazare et al. characterized the impact of visceral vs. subcutaneous fat measured using Computed Tomography on various cardiometabolic outcomes across major ethnic groups in United States. In summary, it remains unclear how visceral adiposity contributes to differences in cardiometabolic health burden across large Asian ethnic groups. Added value of this studyOur multi-ethnic population cohort (n=9,067) included standardized assessments of people of Chinese, Malay, and Indian ancestries living in shared environment, bringing relevance to a wide spectrum of global Asian diaspora. We used the whole-body DEXA-based quantification of visceral fat mass which enables separate assessments of visceral adiposity and overall body fat. We show that there are major differences in adiposity and metabolic health between the Chinese, Malay, and Indian Asian people we studied, and that adiposity makes an important contribution to metabolic health in all three of these Asian ethnic subgroups. However, we also show that excess visceral adiposity only partially explains the difference in diabetes, insulin resistance and related metabolic disturbances between major Asian ethnic subgroups, indicating the presence of additional pathophysiological processes that remain to be identified. Implications of all the available evidenceExcess visceral adiposity is an important contributor to cardiovascular and metabolic health in Asian populations. Strategies to reduce excess adiposity, in particular visceral fat, in Malay and Indian subgroups offer opportunities for major improvements in cardiometabolic health in Asian people, who account for [~]60% of the global population. The difference in diabetes, insulin resistance and related metabolic disturbances between major Asian ethnic subgroups remains unexplained, providing the motivation for further research to identify additional pathophysiological processes underlying these leading global diseases.

Background/ObjectivesThis study aimed to predict body mass index (BMI) trajectories from childhood to early adulthood using explainable artificial intelligence, integrating polygenic scores (PGS), maternal, early-life, and familial factors to identify key predictors of obesity risk and inform prevention strategies. Subjects/MethodsWe analysed longitudinal data from the Raine Study Gen2 cohort, recruiting 2 868 participants. This observational study, without randomization or case-control design, collected BMI measurements at ages 8, 10, 14, 17, 20, 23, and 27 years. We applied Kolmogorov-Arnold Networks (KAN) alongside conventional machine learning models, integrating epidemiological variables (maternal and paternal anthropometrics, parental education, early-life skinfold measurements) with seven BMI-related PGS. The analysis spanned from childhood to early adulthood, with no intervention administered. ResultsThe KAN model, combining epidemiological and PGS data, achieved predictive performance with R{superscript 2} ranging from 0.81 at age 8 to 0.34 at age 27. BMI z-score at age 5 was the dominant predictor in early years, with PGS influence increasing post-adolescence. Maternal and paternal anthropometric measures, parental education, and early-life skinfold measurements were significant contributors. The interpretable KAN model revealed the dynamic interplay of genetic and environmental factors, with early-life BMI z-score and PGS emerging as key drivers of BMI trajectories across life stages. ConclusionsThese findings highlight the dynamic interplay of genetic and environmental factors across life stages, underscoring the potential of early-life BMI as a biomarker for obesity risk. Our interpretable model offers actionable insights for targeted obesity prevention strategies.

Obesity is a multifactorial disorder driven by sustained energy imbalance. The hypothalamus is an important regulator of energy homeostasis and therefore likely involved in obesity pathophysiology. Animal studies suggest that obesity-related diets induce structural changes in the hypothalamus through inflammation-like processes. Whether this translates to humans is however largely unknown. Therefore, we aimed to assess obesity-related differences in hypothalamic macro- and microstructure based on a multimodal approach using T1-weighted and diffusion-weighted magnetic resonance imaging (MRI) acquired at 3 Tesla in a large well-characterized sample of the Leipzig Research Center for Civilization Diseases (LIFE) cohort (n1 = 338, 48% females, age 21-78 years, BMI 18-43 kg/m2). We found that higher body mass index (BMI) selectively predicted higher mean proton diffusivity (MD) within the hypothalamus, indicative of compromised microstructure in the underlying tissue. Results were independent from confounders and confirmed in another independent sample (n2 = 236). In addition, while hypothalamic volume was not associated with obesity, we identified a sexual dimorphism and larger hypothalamic volumes in the left compared to the right hemisphere. Using two large samples of the general population, we showed that a higher BMI specifically relates to altered microstructure in the hypothalamus, independent from confounders such as age, sex and obesity-associated co-morbidities. This points to persisting microstructural changes in a key regulatory area of energy homeostasis occurring with excessive weight. These findings may help to better understand the pathomechanisms of obesity and other eating-related disorders.

ObjectiveChildhood obesity is a serious health concern that is not yet fully understood. Previous research has linked obesity with neurobehavioral factors such as behavior, cognition, and brain morphology. The causal directions of these relationships remain mostly untested. MethodsWe filled this gap by using the Adolescent Brain Cognitive Development study cohort comprising 11,875 children aged 9-10. First, correlations between body mass percentile and neurobehavioral measures were cross-sectionally analyzed. Effects were then aggregated by neurobehavioral domain for causal analyses. Direction of Causation twin modeling was used to test the direction of each relationship. Findings were validated by longitudinal cross-lagged panel modeling. ResultsBody mass percentile correlated with measures of impulsivity, motivation, psychopathology, eating behavior, and cognitive tests (executive functioning, language, memory, perception, working memory). Higher obesity was also associated with reduced cortical thickness in areas of the frontal and temporal lobe but with increased thickness in parietal and occipital brain areas. Similar although weaker patterns emerged for cortical surface area and volume. Twin modeling suggested causal effects of childhood obesity on eating behavior ({beta}=.26), cognition ({beta}=.05), cortical thickness ({beta}=.15), and cortical surface area ({beta}=.07). Personality/psychopathology ({beta}=.09) and eating behavior ({beta}=.16) appeared to causally influence childhood obesity. Longitudinal evidence broadly supported these findings. Results regarding cortical volume were inconsistent. ConclusionsResults supported causal effects of obesity on brain functioning and morphology, consistent with effects of obesity-related brain inflammation on cognition. The present study highlights the importance of physical health for brain development during childhood and may inform interventions aimed at preventing or reducing pediatric obesity.

BackgroundMetabolically healthy obesity may be a transient phenotype, but studies with long follow-up, especially covering late-life, are lacking. We therefore describe transitions between body mass index and metabolic health (BMIxMH) categories in a sample with up to 27 years of follow-up, from midlife to late-life. MethodsWe used cohort data from 786 Swedish twins with objective measures of BMIxMH. Metabolic health was defined as absence of metabolic syndrome (MetS). Measurements were categorized into age 50-64, 65-79, or [&ge;]80. Frequencies of transitions between 50-64 and 65-79, and between 65-79 and [&ge;]80 were calculated. One-hundred individuals had measurements in all three age categories, and their dynamic transitions across time were visualized. We also examined frequencies of attrition due to drop-out or death. ResultsThe proportion of individuals with MetS and with overweight or obesity increased over time. However, 28% of individuals with metabolically healthy normal weight and around one fifth of those with metabolically healthy overweight or obesity remained stable across all three age categories. Transitions from MetS to metabolic health were also common, occurring in 7-49% of individuals with metabolically unhealthy normal weight, overweight, and obesity. Drop-out and death during follow-up was comparable across BMIxMH categories. ConclusionsTransitions between metabolically healthy and unhealthy categories were relatively common in both directions across all BMI categories. The similar frequencies of drop-outs and deaths indicate that bias due to differences in attrition between BMIxMH categories is not a major issue in this sample.