Association of point‐of‐care ultrasound‐derived subcutaneous fat thickness with <scp>DXA</scp> ‐measured body fat percentage in older adults

Accurate body fat measurement is essential for evaluating health risks in older adults, and multiple techniques, including hydrostatic weighing, bioimpedance analysis, skinfold thickness measurements, computed tomography (CT), magnetic resonance imaging (MRI), and dual-energy x-ray absorptiometry (DXA), are employed to assess BF%.1 DXA is widely recognised as a reliable, precise, and practical gold standard, particularly in clinical and research contexts, due to its ability to accurately differentiate between fat mass (FM), fat-free mass (FFM), and bone mineral content.2, 3 However, DXA is limited by high equipment costs, operational complexity, radiation exposure, and restricted availability in specific clinical settings.1, 2 Consequently, alternative methods such as ultrasound have garnered interest due to their accessibility and non-invasive nature.4, 5 B-mode ultrasound has emerged as a promising alternative for measuring subcutaneous fat thickness (SFT), offering advantages like portability, ease of use, cost-effectiveness, and absence of radiation exposure.4, 5 However, ultrasound measurement of muscle and fat thickness is limited by its operator dependence, lack of standardised protocols, potential measurement error due to probe pressure and angle, and the requirement for specific training and skill.6 This study aims to investigate the relationship between BF% assessed by DXA and SFT measured by B-mode point-of-care ultrasound (POCUS) in older adults. We hypothesised that SFT (quadriceps, biceps) would correlate significantly with DXA-derived BF%, and SFT can be used to predict total BF% in older adults. This is a cross-sectional study. The University of British Columbia Institutional Review Board (IRB number: H20-01355) approved the study protocol, and all participants signed a consent form. Participants over the age of 65 were recruited through geriatric physicians in geriatrics and fall prevention. SFT was measured using bedside POCUS with participants in a supine position, knees comfortably extended at ~10–20 degrees. A bedside POCUS device (Vscan with Dual probe, GE Healthcare, IL) was used in B-mode for all SFT measurements with minimum pressure. SFT was measured as the distance from the skin layer to where the fat meets the muscle. Quadriceps SFT was measured at the midpoint (50%) between the greater trochanter and the lateral knee joint space. Biceps SFT was measured with the arm fully extended, at the midpoint between the acromion and the cubital fossa. All SFT measures were taken by one technician (BF) trained under the Head of Point-of-Care Ultrasound service (SA) at Vancouver General Hospital. Body fat measurements were obtained using a whole-body scan performed with a Norland XR-36 scanner (Norland Medical Systems, White Plains, NY). The scanner was calibrated daily using a standard calibration block supplied by the manufacturer. Standard DXA procedures involved positioning the participant centrally on the scanning table, ensuring the entire body was within the scan field. Participants were instructed to remain motionless throughout the scan to ensure image clarity and accuracy. According to established guidelines, fat mass data were reported in grams and percentages for total body and regional segments (arms, legs, and trunk).2, 3 The study assessed body composition using fat percentages rather than absolute fat mass. Fat percentage provides a relative measure of adiposity by expressing fat content as a proportion of total body weight, which allows for comparisons across individuals of different body sizes. Pearson correlation analysis was performed between POCUS-measured SFT at two locations and BF% in the arm, leg, trunk, and whole body (total BF%). A univariate regression analysis was conducted to examine the associations between DXA-measured total BF% and SFT measurements at the quadriceps and biceps. Statistical significance was defined as p < 0.05. All statistical analyses were done using R Version 2024.09.0 + 375 (Posit Software, PBC). The mean age of the participants was 74 ± 6 years. Quadriceps SFT measured 0.77 ± 0.36 cm in women compared with 0.34 ± 0.14 cm in men. Total body fat percentage was significantly higher in women (34% ± 7) than in men (24% ± 6). Table 1 summarises the correlations between quadriceps and biceps SFT and total BF%. Quadriceps SFT demonstrated strong correlations with arm, leg, and total BF% (r > 0.70), whereas biceps SFT showed moderate correlations (r = 0.55 for arm and total BF%). Notably, correlation coefficients between SFTs and BF% were higher in females compared with males. Table 2 presents the univariate regression analysis between quadriceps and biceps SFT and DXA-measured total BF%. Quadriceps SFT showed the strongest association with total BF% (R2 = 0.48). Stratified analyses revealed that females demonstrated markedly higher explanatory power than males, with R2 being 1.5–2 times greater (quadriceps: 0.35 vs. 0.24; biceps: 0.47 vs. 0.22). We developed predictive models for total BF% using quadriceps and biceps SFT measurements. Our study demonstrates that SFT, measured by ultrasound, significantly correlates with BF% measured by DXA in older adults. SFT demonstrated a stronger correlation with DXA-measured BF% overall, with females showing higher correlations than males. These results show the potential of ultrasound as a practical and non-invasive tool for assessing body composition, especially in geriatric populations where ease and accessibility are crucial.4, 5 Subcutaneous adipose tissue represents a significant portion of total body fat, and its thickness often mirrors changes in total adiposity.7, 8 Our findings of correlations between ultrasound-derived SFT and DXA-assessed BF% yielded a lower R2 value than the earlier validation studies in different populations.5, 7 In middle-aged and older adults aged between 50 and 76, the estimated BF% measured by ultrasound and DXA-measured BF% was strongly correlated (adjusted R2 = 0.845, p < 0.001).9 Another study among adults older than 60 years of age found that BF% was predicted using waist circumference, hip circumference, triceps skinfold, and gender (R2 = 0.79, SEE = 3.94%). Predicted BF% in the validation group showed a high correlation with the same cohort cross-validation group (R2 = 0.82, SEE = 4.05%) and with an independent cohort (R2 = 0.76, SEE = 4.42%).10 This study was conducted at a single centre, which limits its generalisability. The cross-sectional design does not allow for causal inference regarding the relationship between SFT and BF%. Although using a single technician minimised inter-rater variability, intra-rater reliability was not assessed. Measurement procedures and techniques are not yet standardised and require experienced technicians who need time to learn and practice.4 The prediction model has not undergone internal and external validation. Its performance, accuracy, and generalisability have not been formally evaluated within the dataset used for development (internal validation) or tested on independent datasets (external validation). Future research should focus on establishing standardised measurement protocols and threshold values for ultrasound-derived SFT in older adults. B-mode ultrasound measurement of quadriceps and biceps SFT provides a reasonable, non-invasive estimate of BF% in older adults. Kenneth M. Madden, Boris Feldman, Shane Arishenkoff, and Graydon S. Meneilly conceptualised and designed the study methods. Boris Feldman collected the data and measured muscle thickness by ultrasound. Boris Feldman measured the pixel intensity by importing the captured images. Uyanga Ganbat analysed the data and drafted the manuscript. Kenneth M. Madden, Altan-Ochir Byambaa, and Jonathan Little contributed to the manuscript writing and editing. All authors read and approved the manuscript. The authors have nothing to report. This work was supported by Vancouver Coastal Health Research Institute, Innovation and Translational Research Award F20-00139. The authors declare no conflicts of interest. The Human Subjects Committee of the University of British Columbia approved the study protocol (IRB number: H20-01355; Clinical trials number: NCT04370912). All participants of the study consented to participate in writing. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.70134. Not available due to confidentiality.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.