Dissecting unique and common variance across body and brain health indicators using age prediction.

Ageing is a heterogeneous multisystem process involving different rates of decline in physiological integrity across biological systems. The current study dissects the unique and common variance across body and brain health indicators and parses inter-individual heterogeneity in the multisystem ageing process. Using machine-learning regression models on the UK Biobank data set (N = 32,593, age range 44.6-82.3, mean age 64.1 years), we first estimated tissue-specific brain age for white and gray matter based on diffusion and T1-weighted magnetic resonance imaging (MRI) data, respectively. Next, bodily health traits, including cardiometabolic, anthropometric, and body composition measures of adipose and muscle tissue from bioimpedance and body MRI, were combined to predict 'body age'. The results showed that the body age model demonstrated comparable age prediction accuracy to models trained solely on brain MRI data. The correlation between body age and brain age predictions was 0.62 for the T1 and 0.64 for the diffusion-based model, indicating a degree of unique variance in brain and bodily ageing processes. Bayesian multilevel modelling carried out to quantify the associations between health traits and predicted age discrepancies showed that higher systolic blood pressure and higher muscle-fat infiltration were related to older-appearing body age compared to brain age. Conversely, higher hand-grip strength and muscle volume were related to a younger-appearing body age. Our findings corroborate the common notion of a close connection between somatic and brain health. However, they also suggest that health traits may differentially influence age predictions beyond what is captured by the brain imaging data, potentially contributing to heterogeneous ageing rates across biological systems and individuals.

Intramuscular fat infiltration evaluated by magnetic resonance imaging predicts the extensibility of the supraspinatus muscle.

INTRODUCTION: Rotator cuff (RC) tears result in muscle atrophy and fat infiltration within the RC muscles. An estimation of muscle quality and deformation, or extensibility, is useful in selecting the most appropriate surgical procedure. We determined if noninvasive quantitative assessment of intramuscular fat using MRI could be used to predict extensibility of the supraspinatus muscle. METHODS: Seventeen cadaveric shoulders were imaged to assess intramuscular fat infiltration. Extensibility and histological evaluations were then performed. RESULTS: Quantitative fat infiltration positively correlated with histological findings and presented a positive correlation with muscle extensibility (r = 0.69; P = 0.002). Extensibility was not significantly different between shoulders graded with a higher fat content versus those with low fat when implementing qualitative methods. DISCUSSION: A noninvasive prediction of whole-muscle extensibility may directly guide pre-operative planning to determine if the torn edge could efficiently cover the original footprint while aiding in postoperative evaluation of RC repair. Muscle Nerve 57: 129-135, 2018.

Effect of liraglutide on thigh muscle fat and muscle composition in adults with overweight or obesity: Results from a randomized clinical trial.

BACKGROUND: Excess muscle fat is observed in obesity and associated with greater burden of cardiovascular risk factors and higher risk of mortality. Liraglutide reduces total body weight and visceral fat but its effect on muscle fat and adverse muscle composition is unknown. METHODS: This is a pre-specified secondary analysis of a randomized, double-blind, placebo-controlled trial that examined the effects of liraglutide plus a lifestyle intervention on visceral adipose tissue and ectopic fat among adults without diabetes with body mass index ≥30 kg/m2 or ≥27 kg/m2 and metabolic syndrome. Participants were randomly assigned to a once-daily subcutaneous injection of liraglutide (target dose 3.0 mg) or matching placebo for 40 weeks. Body fat distribution and muscle composition was assessed by magnetic resonance imaging at baseline and 40-week follow-up. Muscle composition was described by the combination of thigh muscle fat and muscle volume. Treatment difference (95% confidence intervals [CI]) was calculated by least-square means adjusted for baseline thigh muscle fat. The association between changes in thigh muscle fat and changes in body weight were assessed using Spearman correlation coefficients. The effect of liraglutide versus placebo on adverse muscle composition, denoted by high thigh muscle fat and low thigh muscle volume, was explored. RESULTS: Among the 128 participants with follow-up imaging (92.2% women, 36.7% Black), median muscle fat at baseline was 7.8%. The mean percent change in thigh muscle fat over median follow-up of 36 weeks was -2.87% among participants randomized to liraglutide (n = 73) and 0.05% in the placebo group (absolute change: -0.23% vs. 0.01%). The estimated treatment difference adjusted for baseline thigh muscle fat was -0.24% (95% CI, -0.41 to -0.06, P-value 0.009). Longitudinal change in thigh muscle fat was significantly associated with change in body weight in the placebo group but not the liraglutide group. The proportion of participants with adverse muscle composition decreased from 11.0% to 8.2% over follow-up with liraglutide, but there was no change with placebo. CONCLUSIONS: In a cohort of predominantly women with overweight or obesity in the absence of diabetes, once-daily subcutaneous liraglutide was associated with a reduction in thigh muscle fat and adverse muscle composition compared with placebo. The contribution of muscle fat improvement to the cardiometabolic benefits of liraglutide requires further study.

Coronary artery calcium, hepatic steatosis, and atherosclerotic cardiovascular disease risk in patients with type 2 diabetes mellitus: Results from the Dallas heart study.

INTRODUCTION: Cardiovascular disease (CVD) risk amongst those with type 2 diabetes (T2D) is heterogenous. The role of imaging-based cardiometabolic biomarkers (e.g., coronary artery calcium [CAC] score, and hepatic triglyceride content [HTC]) in CVD risk stratification in T2D is unclear. To better understand this, we sought to evaluate the individual and joint associations between CAC and hepatic steatosis (HS) with clinical atherosclerotic CVD (ASCVD) in Dallas Heart Study (DHS) participants with and without T2D. METHODS: We examined participants in the DHS, a multi-ethnic cohort study, without self-reported ASCVD. CAC scoring was performed via computed tomography with the mean of two consecutive scores used. HTC was measured using magnetic resonance spectroscopy, and HS was defined as HTC >5.5% The primary outcome was incident ASCVD, defined as coronary heart disease (CHD; myocardial infarction, percutaneous coronary intervention, or coronary artery bypass graft surgery), ischemic stroke, transient ischemic attack, or CVD death. Cox regression analyses, and interaction testing was performed to evaluate the individual and joint associations between CAC and HS with ASCVD. The association between HS and coronary heart disease was validated in the UK Biobank (UKB). RESULTS: A total of 1252 DHS participants were included with mean age 44.8 ± 9.3 years, mean body mass index 28.7 ± 5.9 kg/m2, 55% female, and 59% black with an overall prevalence of T2D of 9.7%. CAC scores were significantly higher (p < 0.01) and HS was significantly more prevalent in those with T2D (p < 0.01). Over a median of 12.3 years, 8.3% of participants experienced ASCVD events. The ASCVD event rate was significantly higher in participants with T2D (20.5% vs 7.0%, p < 0.01). Continuous CAC was associated with ASCVD events in the overall cohort regardless of T2D status with a significant interaction present between CAC and T2D status on ASCVD, Pinteraction = 0.02. HTC was not associated with ASCVD risk in participants without T2D but was inversely associated with risk in participants with T2D (HR 0.91, 95% CI 0.83-0.99 per 1% increase in HTC, p = 0.02), Pinteraction = 0.02. Amongst 37,266 UKB participants, 4.5% had T2D. CHD events occurred in 2.2% of participants, with 10.2% of events occurring amongst those with T2D. An inverse relationship between HTC and CHD was also found amongst those with T2D in UKB with a significant interaction between T2D status and HTC on CHD (HR per 1% increase in HTC 0.95, 95% CI 0.91-0.99, p = 0.01, Pinteraction = 0.02). CONCLUSIONS: In the DHS, we found that CAC was associated with ASCVD risk independent of T2D status. We did not observe an association between HTC and ASCVD in participants without T2D, but there was an inverse association between HTC and ASCVD in those with T2D that was replicated in the UKB cohort. Further investigation is warranted to understand the possible protective association of HS in participants with T2D.

The link between liver fat and cardiometabolic diseases is highlighted by genome-wide association study of MRI-derived measures of body composition.

Obesity and associated morbidities, metabolic associated fatty liver disease (MAFLD) included, constitute some of the largest public health threats worldwide. Body composition and related risk factors are known to be heritable and identification of their genetic determinants may aid in the development of better prevention and treatment strategies. Recently, large-scale whole-body MRI data has become available, providing more specific measures of body composition than anthropometrics such as body mass index. Here, we aimed to elucidate the genetic architecture of body composition, by conducting genome-wide association studies (GWAS) of these MRI-derived measures. We ran both univariate and multivariate GWAS on fourteen MRI-derived measurements of adipose and muscle tissue distribution, derived from scans from 33,588 White European UK Biobank participants (mean age of 64.5 years, 51.4% female). Through multivariate analysis, we discovered 100 loci with distributed effects across the body composition measures and 241 significant genes primarily involved in immune system functioning. Liver fat stood out, with a highly discoverable and oligogenic architecture and the strongest genetic associations. Comparison with 21 common cardiometabolic traits revealed both shared and specific genetic influences, with higher mean heritability for the MRI measures (h2 = .25 vs. .13, p = 1.8x10-7). We found substantial genetic correlations between the body composition measures and a range of cardiometabolic diseases, with the strongest correlation between liver fat and type 2 diabetes (rg = .49, p = 2.7x10-22). These findings show that MRI-derived body composition measures complement conventional body anthropometrics and other biomarkers of cardiometabolic health, highlighting the central role of liver fat, and improving our knowledge of the genetic architecture of body composition and related diseases.

Population-based body-brain mapping links brain morphology with anthropometrics and body composition.

Understanding complex body-brain processes and the interplay between adipose tissue and brain health is important for understanding comorbidity between psychiatric and cardiometabolic disorders. We investigated associations between brain structure and anthropometric and body composition measures using brain magnetic resonance imaging (MRI; n = 24,728) and body MRI (n = 4973) of generally healthy participants in the UK Biobank. We derived regional and global measures of brain morphometry using FreeSurfer and tested their association with (i) anthropometric measures, and (ii) adipose and muscle tissue measured from body MRI. We identified several significant associations with small effect sizes. Anthropometric measures showed negative, nonlinear, associations with cerebellar/cortical gray matter, and brain stem structures, and positive associations with ventricular volumes. Subcortical structures exhibited mixed effect directionality, with strongest positive association for accumbens. Adipose tissue measures, including liver fat and muscle fat infiltration, were negatively associated with cortical/cerebellum structures, while total thigh muscle volume was positively associated with brain stem and accumbens. Regional investigations of cortical area, thickness, and volume indicated widespread and largely negative associations with anthropometric and adipose tissue measures, with an opposite pattern for thigh muscle volume. Self-reported diabetes, hypertension, or hypercholesterolemia were associated with brain structure. The findings provide new insight into physiological body-brain associations suggestive of shared mechanisms between cardiometabolic risk factors and brain health. Whereas the causality needs to be determined, the observed patterns of body-brain relationships provide a foundation for understanding the underlying mechanisms linking psychiatric disorders with obesity and cardiovascular disease, with potential for the development of new prevention strategies.