Аннотация
Introduction: Because muscle and bone are biomechanically linked, it is important to examine the tissues as a functional unit. The muscle-to-bone ratio (MBR) and soft tissue-to-bone ratio (SBR) are metrics that examine the weight placed on the bone from muscle and/or fat masses. Currently, two methods exist for calculating MBR and SBR: 1) dual X-ray absorptiometry (DXA), and 2) International Society for Advanced Kinanthropometry (ISAK) profile. This study compared DXA- and ISAK-derived methods of determining total-body MBR and SBR. Methods: A total of 31 (n=16 male, n=15 female) participants, averaging 31.4 (±11.4) years old with a body mass index (BMI) of 24.6 (±4.0) kg/m 2 had their lean mass (LM), fat mass (FM) and bone mineral content (BMC) measured using DXA (Hologic Horizon A; Hologic Inc., Marlborough, MA, USA). After the full-body DXA scan, an ISAK-certified, Level 2 anthropometrist conducted an ISAK full-profile assessment consisting of 43 measurements including skinfold thickness, girths, lengths and heights, and breadths and depths. DXA-derived MBR was obtained by dividing LM by BMC. SBR was determined by dividing soft tissue mass (LM+FM) by BMC. ISAK-derived total body composition measurement provides estimates of total-body fat, muscle, bone, residual, and skin percentages. These percentages were multiplied by the participants’ weight to obtain the values as masses. ISAK-derived muscle mass, residual mass, and skin mass were combined to obtain a comparable measure of LM. ISAK-derived MBR was obtained by dividing LM by bone mass (BM) and ISAK-derived SBR was obtained by dividing (LM+FM) by BM. Results: DXA and ISAK measures of LM (2.5±11.7 vs. 56.7±13.0 kg, p< 0.001), FM (17.7±7.4 vs. 22.4±6.6 kg, p=0.002) and BM (2.5±0.5 vs. 8.2±2.0 kg, p< 0.001 ) were significantly different. DXA and ISAK derived MBR (22.8±3.1 vs. 3.6±0.8) and SBR (30.3±5.2 vs. 6.5±1.1) were also significantly (p< 0.001) different. There was a significant positive correlation between DXA and ISAK derived LM (r=0.81, p< 0.001) and BMC/BM (r=0.66, p< 0.001), but not FM (r=0.32, p=0.84). Both DXA and ISAK derived MBR (r = -0.11, p = 0.54) and SBR (r=0.02, p=0.91), were not significantly correlated. Bland-Altman analysis revealed a moderate level of agreement (LoA= 29.48, -1.19) between DXA and ISAK LM due to the distribution of values outside 1.96 standard deviation. Between DXA and ISAK FM, the median difference was positive (+5.1 kg), indicating that ISAK overestimates FM compared to DXA. Most data points fell within the non-parametric LoA of 2.5th and 2.5th percentiles (27.9 to -13.2), indicating good agreement. There was a good level of agreement (LoA = -2.38, -9.01) between DXA BMC and ISAK BM observed by the distribution of all values within the 1.96 standard deviation. However, ISAK BM overestimated the DXA BMC indicated by a positive bias of +5.7 kg. Conclusion: DXA- and ISAK-derived methods of quantifying BM, LM and FM are systematically different. These differences result in unrelated determinations of both MBR and SBR. Therefore, MBR and SBR determined by DXA and ISAK should not be used interchangeably. This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.