Frailty is associated with inflammation, impaired glucose metabolism and reduced bone mineral density after adjustment for fat mass index: a uk biobank study

Abstract Background/Aims Frailty, a common syndrome associated with aging, represents a huge public health burden. An improved understanding of the mechanisms underlying frailty are essential, especially in younger populations in whom frailty could be prevented. Fundamental aging processes such as chronic inflammation are associated with frailty, but the independence of these relationships from age, sex, lifestyle and adiposity is unclear. Using UK Biobank, we investigated associations between frailty, blood biomarkers and bone health, independent of these characteristics. Methods 502,640 participants aged 40-69 years were recruited to UK Biobank 2006-10. Venous blood samples were obtained and fat mass was measured by bioelectrical impedance. From 2014 onwards, a subset attended an imaging follow-up, including whole-body DXA (GE Lunar iDXA), grip strength (Jamar dynamometer), and a questionnaire. Frailty was defined using a modification of Fried’s classification (at least 3 of weight loss, mental exhaustion, low physical activity, slow gait speed and low grip strength). The presence of 1-2 criteria designated pre-frailty. Linear regression was used to discern associations between frailty status, biochemical markers (CRP, 25(OH)-vitamin D, IGF-1, Calcium, Phosphate, ALP, Creatinine, Urate, glucose and HbA1c) and bone outcomes, adjusting for age, sex, fat mass, smoking, alcohol and educational level. Non-frail was the reference category and blood biomarkers were standardised (β representing the mean difference in SD). Results 22,332 participants (11,484 women and 10,848 men) with frailty assessment and either DXA bone measures or blood biochemistry were included in the analysis; of these 547 (2.4%) were frail and 9359 (41.9%) pre-frail. Frail participants were more likely to be female [59.6% vs. 50.9%], with higher BMI [mean (SD) 30.7 (6.4) vs. 25.9 (4.0) kg/m2] and of older age [mean (SD) 63.2 (7.9) vs. 62.6 (7.3) years]. After full adjustment, frail participants had lower 25(OH)-vitamin D levels [-0.36 SD (-0.45,-0.27)], higher CRP [+0.29 SD (95% CI 0.21, 0.37)], and higher HbA1c [+0.32 SD (0.23, 0.40)], all p < 0.001. Frail participants had lower femoral neck [-0.18 SD (-0.26, -0.10)] and lumbar spine bone mineral density (BMD) [-0.13 SD (-0.21, -0.06)], both p < 0.001. BMD associations were only apparent after fat mass adjustment, prior to fat mass adjustment the direction of association was positive. Similar associations were observed for pre-frail vs. non-frail participants. Conclusion In the UK Biobank cohort of middle-older aged people, frailty is associated with high levels of systemic inflammation, low 25(OH)-vitamin D, poorer glucose handling and lower BMD, independent of age, sex and lifestyle. The inversion of BMD associations after adjustment for fat mass is intriguing and suggests substantial phenotypic complexity in the associations. Our findings support further investigation of frailty phenotypes at younger ages and elucidation of possible contributions from conditions such as metabolic syndrome and obesity-related chronic inflammation. Disclosure E.M. Curtis: None. S. D'Angelo: None. S. Woolford: None. R. Durdin: None. Z. Raisi-Estabragh: None. K.A. Ward: None. C. Cooper: None. N.C. Harvey: None.