Laparoscopic sleeve gastrectomy (LSG), the most common bariatric surgical procedure, leads to durable weight loss and improves obesity-related comorbidities. However, it induces abnormalities in bone metabolism. One unexplored potential contributor is the gut microbiome, which influences bone metabolism and is altered after surgery. We characterized the relationship between the gut microbiome and skeletal health in severe obesity and after LSG. In a prospective cohort study, 23 adults with severe obesity underwent skeletal health assessment and stool collection preoperatively and 6mo after LSG. Gut microbial diversity and composition were characterized using 16S rRNA gene sequencing, and fecal concentrations of short-chain fatty acids (SCFA) were measured with LC-MS/MS. Spearman's correlations and PERMANOVA analyses were applied to assess relationships between the gut microbiome and bone health measures including serum bone turnover markers (C-terminal telopeptide of type 1 collagen [CTx] and procollagen type 1N-terminal propeptide [P1NP]), areal BMD, intestinal calcium absorption, and calciotropic hormones. Six months after LSG, CTx and P1NP increased (by median 188% and 61%, P< .01) and femoral neck BMD decreased (mean -3.3%, P< .01). Concurrently, there was a decrease in relative abundance of the phylum Firmicutes. Although there were no change in overall microbial diversity or fecal SCFA concentrations after LSG, those with greater within-subject change in gut community microbial composition (β-diversity) postoperatively had greater increases in P1NP level (ρ= 0.48, P= .02) and greater bone loss at the femoral neck (ρ= -0.43, P= .04). In addition, within-participant shifts in microbial richness/evenness (α-diversity) were associated with changes in IGF-1 levels (ρ= 0.56, P< .01). The lower the postoperative fecal butyrate concentration, the lower the IGF-1 level (ρ= 0.43, P= .04). Meanwhile, the larger the decrease in butyrate concentration, the higher the postoperative CTx (ρ= -0.43, P= .04). These findings suggest that LSG-induced gut microbiome alteration may influence skeletal outcomes postoperatively, and microbial influences on butyrate formation and IGF-1 are possible mechanisms.

Cadmium (Cd) is a heavy metal and natural element found in soil and crops with increasing concentrations linked to phosphate fertilizers and sewage sludge applied to crop lands. A large fraction of older US men and woman have documented Cd exposure. Cd exposure has proven health concerns such as risk of lung cancer from inhalation and impaired renal function; however, growing evidence suggests it also influences bone and muscle health. Given that low levels of Cd could affect bone and muscle, we have designed prospective studies using the two largest and most detailed US studies of bone health in older men and women: the Osteoporotic Fractures in Men Study and the Study of Osteoporotic Fractures. We are investigating the association of urinary cadmium (U-Cd), as a surrogate for long-term Cd exposure, with bone and muscle health. Building off suggestive evidence from mechanistic and cross-sectional studies, this will be the first well-powered prospective study of incident fracture outcomes, bone loss, and muscle loss in relation to U-Cd, an established biomarker of long-term Cd exposure. The following is a proposed protocol for the intended study; if successful, the proposed studies could be influential in directing future US policy to decrease Cd exposure in the US population similar to recent policies adopted by the European Union to limit Cd in fertilizers.

The D3-creatine (D3-Cr) dilution method is of emerging interest for estimating total-body skeletal muscle mass. This review explores the association of muscle mass estimated via D3-Cr with various clinical outcomes and provides a summary of the literature comparing D3-Cr with other body composition techniques. A literature search was conducted on PubMed/MEDLINE and Web of Science for studies using D3-Cr to measure muscle in adult populations (ie, ≥18 years old) from inception until September 2023. Out of the 23 included studies, 15 investigated the correlation between D3-Cr and clinical outcomes. More consistent associations were reported for mortality (100%, n = 2), mobility disability (100%; n = 5), falls and fractures (100%; n = 3), physical performance (63.3%; n = 11), muscle strength (44.4%; n = 9), and muscle composition (33.3%; n = 3). However, conflicting findings were also reported for such correlations. Among the 23 studies, 14 compared D3-Cr-estimated muscle with other body composition techniques, including magnetic resonance imaging (MRI) as a reference method. Strong and positive correlations were found between D3-Cr and MRI. Nonetheless, variations in muscle measurements were noted, with differences in D3-Cr values ranging from 0.62kg lower to 13.47kg higher compared to MRI. D3-Cr-estimated muscle mass may be a valuable predictor of clinical outcomes showing consistent associations with falls and fractures, mobility disability, and mortality. However, less consistent associations were found with muscle strength and composition, and physical performance. Although a strong correlation exists between D3-Cr-estimated muscle mass and MRI measurements, under- or overestimation may occur.

Abstract Performance fatigability manifests as insufficient energy to complete daily physical tasks and worsens with aging, exacerbating vulnerability to disability. Skeletal muscle energetics also declines with aging. Thus, we hypothesized muscle energetics may be an important contributor to performance fatigability. In the Study of Muscle, Mobility and Aging (SOMMA), participants completed a usual-paced 400m walk while wearing a wrist-worn ActiGraph, from which raw data were used to derive the Pittsburgh Performance Fatigability Index (PPFI, higher=more severe fatigability) that quantifies percent decline in the entire individual cadence-versus-time trajectory. Maximal oxidative phosphorylation (maxOXPHOS) in skeletal muscle mitochondria was quantified in vitro using high-resolution respirometry in permeabilized fiber bundles from vastus lateralis muscle biopsies. Maximal adenosine triphosphate production (ATPmax) was assessed in vivo by 31P magnetic resonance spectroscopy. We conducted separate tobit regressions to examine associations of maxOXPHOS and ATPmax with PPFI, adjusting for technician/site, age, sex, race, height, weight, and mins/day moderate-to-vigorous physical activity measured by ActiGraph in free-living, in N=795 participants with complete PPFI scores and &amp;gt;1 energetics measure (70-94 yrs, 58% women). Median PPFI scores were 1.4% [IQR: 0-2.9%]. After adjustment, each SD (18.4 pmol/(s*mg)) lower maxOXPHOS was associated with 0.55% (95% CI: 0.26, 0.85) higher PPFI scores, while each SD (0.2 mM/sec) lower ATPmax was associated with 0.54% (95% CI: 0.27, 0.81) higher PPFI scores. Our results indicate that lower skeletal muscle energetics were associated with more severe performance fatigability. This suggests that therapeutics targeting muscle energetics may thereby potentially mitigate fatigability and lessen susceptibility to disability among older adults.

Abstract Cardiopulmonary exercise testing (CPET) is the gold standard to measure cardiorespiratory fitness (CRF), assessed as peak oxygen consumption (VO2peak), yet is not always feasible in older adults due to cost, safety, and participant burden. Usual-paced 400-m Long Distance Corridor Walk (LDCW), a measure of mobility, may be an alternative to traditional CPET. We examined whether usual-paced 400-m LDCW was associated with VO2peak among older adults using cross-sectional data in SOMMA. Participants (N=820, age = 76.2±4.9, 58% women, 86% White) completed a usual-paced 400-m LDCW (seconds) and a treadmill CPET using a modified symptom-limited protocol at baseline. VO2peak (mL/kg/min) was defined as the highest 20-second average O2 consumption achieved during CPET. Mean VO2peak was 20.2±4.8 mL/kg/min and mean usual-paced 400-m LDCW time was 388.8±69.7 seconds. Usual-paced 400-m LDCW time was moderately inversely correlated with VO2peak (r=-0.52, p&amp;lt;0.0001). Potential covariates considered for the step-wise linear regression included age, sex, body mass index (BMI), Borg Rating of Perceived Exertion (RPE, range 6-20) at end of usual-paced 400-m LDCW, min/week moderate-vigorous physical activity (MVPA, CHAMPS questionnaire), 4-m gait speed, and Short Physical Performance Battery (range 0-12). The final parsimonious model included usual-paced 400-m LDCW time, age, sex, BMI, RPE, and MVPA. Every 30-second higher usual-paced 400-m LDCW time was associated with a 1.98 mL/kg/min lower VO2peak (p &amp;lt; 0.001), explaining 30.5% of the 55.7% total variance in VO2peak. Usual-paced 400-m LDCW was a strong estimator of CRF (VO2peak) after adjustment, providing an inexpensive and safer alternative among older adults across ranges of physical function and fitness.

Abstract We investigated the association of diabetes with cardiorespiratory fitness and skeletal muscle mitochondrial energetics in older adults (N=876, mean age ± SD: 76.3 ± 5.0 yrs.; 59% females) from the Study of Muscle, Mobility and Aging (SOMMA). Participants were grouped by self-reported diabetes status (N=131, with diabetes). Cardiorespiratory fitness (modified Balke protocol), mitochondrial energetics (31P magnetic resonance spectroscopy [31P-MRS] and high resolution respirometry from skeletal muscle biopsy), physical activity (wearable accelerometer) and body composition (magnetic resonance imaging) were measured. Generalized linear regression models were used to calculate means adjusted by age, race, gender, BMI, and co-morbidities. Thigh fat-free muscle volume (p=0.12) and abdominal subcutaneous adipose tissue depot (p=0.15) were similar between those with and without diabetes. However, visceral (18.84%, p&amp;gt;0.01) and intramuscular adipose tissue (6.57%, p=0.02) were higher in older adults with diabetes. VO2peak (-4.79%, p&amp;gt;0.01), and mitochondria energetics measured by ATPmax (-5.10%, p=0.04), maximum complex I+II carbohydrate (maxOXPHOSCHO) (-6.67%, p=0.02) and fatty acid (maxOXPHOSFAO) (-8.70%, p=0.01) stimulated respiration were lower in those with diabetes compared to those without. Following further adjustments for both physical activity and visceral adiposity, revealed that only VO2peak (-4.65%, p&amp;gt;0.01) and maxOXPHOSFAO (-7.38%, p=0.02) remained significantly lower in older adults with diabetes. Overall, our data suggests that cardiorespiratory fitness and skeletal muscle mitochondria energetics are reduced in older adults with diabetes, independent of visceral adiposity and physical activity.

Abstract Cardiorespiratory fitness (VO2peak) declines with age, and this may be due in part to weakened circadian functions. Rhythmic patterns of rest-activity behavior over 24h represents a novel feature of physical activity, indicative of circadian behavior. Whether VO2peak relates to rest-activity rhythms (RAR) is unknown. We determined cross-sectional associations between VO2peak and RAR in SOMMA (N=714, Age: ≥70 yrs). Activity data from wrist-worn accelerometry (ActiGraph GT9X) were collected in one-minute epochs, over 8, 24h periods. An extension to the traditional cosine model was used to map RAR to activity data, calculating these parameters: strength of the rhythm (amplitude); robustness of rhythm (pseudo-F statistic); and timing of peak activity (acrophase). RAR parameters were expressed as quartiles. Linear models examining associations of RAR and VO2peak were adjusted for potential confounders (age, race, height, comorbidities and lifestyle). Interactions of RAR predictors with sex were examined. Associations between VO2peak and rhythmic amplitude appears to differ by sex (p-interaction=0.07). Among men, VO2peak was positively associated with amplitude (adjusted mean Q1: 21.3 vs. Q4: 23.9, p-trend=0.005)-even after adjusting for moderate-to-vigorous physical activity levels (MVPA)-although associations were more modest and not significant in women. Those with higher VO2peak had an earlier acrophase (time of peak activity) (adjusted mean: Q1 &amp;lt;1:30PM: 21.0 vs. Q4&amp;gt;3:09PM 19.3, p-trend &amp;lt;0.0001) and no sex interaction was observed. There were no associations between the robustness of RAR and VO2peak. Analyses reveal that better cardiorespiratory fitness relates to features of circadian behavior independent of MVPA, supporting emerging research that circadian-based interventions promote healthy aging.

Abstract An age-related decline in muscle mitochondrial energetics is known to contribute to the loss of muscle function in older adults. Women experience a higher prevalence of mobility impairment compared to men, but it is unknown whether gender-specific differences in muscle energetics underlie this disparity. In the Study of Muscle, Mobility and Aging (SOMMA), muscle energetics were characterized using in vivo 31-Phosphorus Magnetic Resonance Spectroscopy and High-Resolution Respirometry of the vastus lateralis. A Short Physical Performance Battery score ≤ 8 was used to define lower-extremity mobility impairment. In this analysis of 773 participants age 70-94, women had greater odds (OR=1.83, p=0.03) of mobility impairment compared to men, which was largely explained by significantly lower muscle energetics (Maximal OXPHOS in Women= 55.06 +/- 15.95; Men= 65.80 +/- 19.74; p&amp;lt;0.0001) using mediation modeling (adjusted OR=1.25, p=0.45). The mobility impairment disparity could be further explained by BMI, race, physical activity, and number of comorbidities (adjusted OR=0.98, p=0.96). However, none of these covariates, including age, fully explained gender differences in muscle mitochondrial energetics (e.g. Adjusted Maximal OXPHOS in Women= 56.5, 95%CI=54.8, 58.1; Men= 64.4, 95%CI=62.5, 66.2; p&amp;lt;0.0001). Notably, high BMI and low gait speed were associated with lower muscle energetics, though BMI slightly more so in women (ATPmax BMI/gender interaction, p= 0.02) than men, and age was only significantly negatively associated with muscle energetics in men (e.g. Maximal ETS capacity age/gender interaction, p=0.04). In conclusion, women had lower muscle mitochondrial energetics compared to men, which largely explained their greater odds of lower-extremity mobility impairment.

Abstract While gut dysbiosis has been linked to frailty in aging, its association with early mobility impairments is unclear. Here, our primary goal was to determine the cross-sectional associations between walking speed and gut microbiome in 740 older men (84±4y) from MrOS with available stool samples and 400m walking speed measured in 2014–16. We also analyzed the retrospective longitudinal associations between changes in 6-meter walking speed (from 2005-06 to 2014-16) and gut microbiome composition among participants with available data (702/740). The gut microbiome composition was determined by 16S sequencing (DADA2 and SILVA). We examined diversity, taxa abundance (by ANCOM-BC), and performed network analysis (by NetCoMi) to uncover microbial communities interactions by walking speed levels. Higher walking speed (m/s) was associated with greater microbiome Shannon α-diversity (R=0.11; P=0.004). Decline in walking speed was associated with lower Shannon α-diversity (R=0.07; P=0.054). Faster walking speed and less decline in walking speed were associated with higher abundance of genus-level bacteria that produce short-chain fatty acids, and possess anti-inflammatory properties, including Paraprevotella, Fusicatenibacter, and Alistipes, adjusting for age, race, site, education, health, marital status, weight, height, physical activity, batch, medications, energy, and fiber intake (P&amp;lt; 0.05). The gut microbiome networks of participants in the first vs. last quartile of walking speed (≤0.9 vs. ≥1.2 m/s) exhibited distinct characteristics, including different cluster numbers, hubs, and centrality measures (P&amp;lt; 0.05). Faster walking speed and its less decline were associated with higher gut microbiome diversity, suggesting potential role of microbiome in preserving mobility in aging.

Abstract Greater perceived fatigability has been associated with neurological diseases, but we do not know whether there are associations with cognition among at-risk older adults. At baseline, SOMMA participants completed the Pittsburgh Fatigability Scale (PFS) Physical and Mental subscales (each range 0–50; higher scores = greater fatigability; clinically meaningful increment=4-point physical, 3-point mental) and four cognitive function assessments [Digit Symbol Substitution Test (DSST), Montreal Cognitive Assessment (MoCA), Trail Making Test Part B (TMT-B), and California Verbal Learning Test-Second Edition, Short Form (CVLT-II SF)]. Linear regression quantified associations between PFS subscales and cognitive assessment scores adjusting for site, age, sex, race, education, marital status, and history of stroke, cancer, heart failure, and lung disease. In the 873 participants (59.2% women; age 76.3+/-5.0 years; 85% White), 54% had greater physical fatigability (PFS Physical≥15) and 23% had greater mental fatigability (PFS Mental≥13). Prevalence of cognitive impairment was 2% moderate (MoCA 10-17) and 0% severe ( &amp;lt;10). After adjustments, for each 4-point higher PFS Physical score participants had 0.8 fewer correct DSST items [Beta coefficient and 95% confidence interval: -0.8 (-1.2, -0.4); n=866] and 2.0 seconds slower TMT-B time [2.0 (0.1, 3.8); n=835]. Associations were similar for each 3-point higher PFS Mental score [DSST: -0.7 (-1.1, -0.4) and TMT-B time: 2.5 (1.0, 4.1)]. Neither PFS subscale was associated with MoCA or CVLT. Our results suggest that higher perceived physical and mental fatigability scores may both be indicative of cognitive impairment, particularly in processing speed.