Advancing the Assessment of Regional Body Composition in Children and Adolescents Using Regional Raw Bioelectrical Impedance Data

Advancing the Assessment of Regional Body Composition in Children and Adolescents Using Regional Raw Bioelectrical Impedance Data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Advancing the Assessment of Regional Body Composition in Children and Adolescents Using Regional Raw Bioelectrical Impedance Data Gil Rosa, Ruben Francisco, Inês Correia, Ana Bernardino, João Magalhães, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9224431/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background/Objectives Regional bioelectrical impedance analysis (BIA) offers a practical yet unexplored approach to characterizing limb-specific physiological properties in youth. Thus, we aimed to examine the cross-sectional and prospective associations between regional raw BIA parameters, including stature- and limb length-normalized indices, and regional body composition in children and adolescents. Subjects/Methods A total of 681 participants underwent regional BIA at 50kHz to obtain phase angle (PhA), resistance index (RI), RI normalized to limb length (RI-specific), reactance index (XcI), XcI normalized to limb length (XcI-specific), parallel XcI (XcpI), XcpI normalized to limb length (XcpI-specific), and capacitance (Cap). Lean soft tissue (LST), fat mass (FM), and bone mineral content (BMC) were measured with dual-energy X-ray absorptiometry. Multiple regression models with false-positive detection were considered. Results Regional BIA variables showed weak-to-strong cross-sectional associations with LST and BMC ( β = |0.143| to |0.921|). RI-specific consistently exhibited the strongest relationships ( β = -0.879 to -0.544) outperforming RI and all Xc-derived indices. PhA and Cap were positively associated with LST and BMC but showed lower explanatory power ( β = 0.116 to 0.838). For FM, the associations were mostly weak ( β < |0.583|), with PhA showing the highest associations. In prospective analyses, effect sizes were attenuated overall; however, changes in RI-related indices remained consistently associated with LST and BMC ( β = -0.593 to -0.136), while no associations were observed for FM. Conclusions Regional raw BIA variables enhance the characterization of limb composition in youth, with R-related indices offering the greatest sensitivity to regional LST and BMC. Health sciences/Health care/Nutrition Health sciences/Health care/Paediatrics Bioimpedance Dual-energy X-ray absorptiometry Pediatric Phase angle Regional analysis Figures Figure 1 Figure 2 Introduction Bioelectrical impedance analysis (BIA) has become increasingly relevant across clinical and research fields due to its ability to characterize and discriminate indicators of health, nutritional status, body composition, physical function, and even mortality (1). As a practical alternative to costly and less accessible methods, BIA has traditionally been implemented through prediction equations to estimate health-related outcomes (2, 3). However, these equations are susceptible to latent sources of error arising from population and BIA device specificity, hydration variability, and multi-step modeling assumptions (4, 5). Over the last two decades, there has been a growing interest in the direct interpretation of raw BIA variables (1), which may provide more physiologically established information (5, 6). Among raw BIA variables, particular attention has been given to whole-body phase angle (PhA), which represents the physiological ratio between capacitive reactance (Xc, the delay of alternating current penetrating cell membrane) and resistance (R, the direct opposition to current flow) (7). PhA is expressed as the arctangent of Xc/R x 180/π. In pediatric populations, whole-body PhA measured mostly at 50kHz has been shown to be useful for informing body composition mostly measured at the molecular level [e.g., lean soft tissue (LST), fat-free mass (FFM), and fat mass (FM)] (8, 9, 10, 11) and physical fitness (8, 12, 13, 14). Owing to its consistent associations with morphofunctional characteristics across diverse health contexts in youth, PhA is gaining recognition as a holistic indicator of health (15, 16). Beyond whole-body information, modern regional BIA devices now allow the acquisition of raw BIA parameters from individual body segments. This approach is based on the principle of equipotentials (17), which allows automatic selection of multiple current pathways without the need for electrode repositioning (18). Particularly among children and adolescents, there is a biological rationale for exploring the usefulness of these regional raw BIA parameters, as rapid growth and maturation lead to limb-specific morphological and functional changes that whole-body measures may not capture. These adaptations include longitudinal bone elongation, regional LST or FM accretion, and progressive improvements in neuromuscular performance. Accordingly, the extent to which regional PhA and other raw BIA parameters explain such limb-specific domains of body composition and physical function remains insufficiently investigated in youth (19, 20). Moreover, the physiological interpretation of regional BIA measurements not only on their direct acquisition but also on how raw BIA parameters are normalized to reflect segment-specific conductive pathways. To date, most evidence has relied on standardizing raw BIA variables to stature, under the rationale that stature normalization accounts for the conductive pathway length (i.e., following Pouillet's equation (21)), thus improving interpretability relative to absolute values (10). Accordingly, raw BIA stature-normalized data [e.g., impedance (Z) and R] have been used to develop predictive equations for regional LST and FM (22, 23). However, this approach assumes that total body height adequately represents the electrically relevant conductive pathway, an assumption that may not hold when the objective is true segmental assessment and may attenuate physiological specificity and obscure region-specific associations. From a biophysical standpoint, the conductive pathway of interest in regional evaluation is more plausibly defined by the anatomical length of the specific limb assessed rather than overall stature (24). Therefore, adjusting raw BIA variables to segment length may provide a more anatomically coherent representation of regional characteristics. Whether segmental length adjustment of raw BIA variables yields greater physiological relevance than conventional stature normalization in a regional approach remains to be tested. To further expand the potential applications of regional BIA in youth, the present aim of this investigation is to determine whether distinct regional raw BIA parameters provide cross-sectional and prospective information on limb-specific body composition in a large cohort of children and adolescents. In addition, it also aims to examine whether normalization of regional R to limb segment length (RI-specific) offers greater informational value than traditional stature-normalized adjustment (RI). Materials & Methods Sample and study design A mixed cross-sectional and prospective design was used to assess 681 children and adolescents (49% females) enrolled in the PRO-BODY project (10). The participants were assessed between September 2022 and July 2023 (Moment 1, baseline) and between September 2023 and July 2024 (Moment 2, follow-up), and were excluded if they had sensory, motor, or intellectual impairments, were institutionalized, or presented a mismatch between chronological age and biological maturation status. Additional information on the investigation protocol, including detailed inclusion and exclusion criteria and ethical approval, has been reported elsewhere (10, 11). Written informed consent was obtained from legal guardians, and verbal assent was provided by all participants. Bioelectrical impedance analysis A phase-sensitive BIA device (BIA 101 BIVA PRO, Akern, Florence, Italy) with eight-point Ag/AgCI low Z [i.e., <30 ohms (Ω)] electrodes (BIATRODES ® , Akern Srl, Florence, Italy) was used to measure PhA (º), R (Ω), and Xc (Ω). The BIA device was calibrated each morning using the manufacturer’s standard control circuit (R = 380 Ω; Xc = 45 Ω). Following verification of pre-test conditions (i.e., minimum 4-hour fast, no structured, vigorous-intensity physical activity in the previous 12 hours), and after instructing participants to empty their bladder and to remove all metal accessories, the technician cleaned the electrode sites with 70% isopropyl alcohol and applied four emitting and four sensing electrodes according to standardized procedures (25). Assessments were performed after a 5-minute resting period using an alternating current of 245 μA RMS at 50 kHz, delivered through the manufacturer-defined regional reading channels. The participants were lying supine on a non-conductive surface, with a leg opening of 45° and arms positioned at 30° from the trunk, in a room free from electromagnetic interference and with an ambient temperature between 22ºC and 25ºC. In the present investigation, the coefficients of variation (CV) for the dominant and non-dominant arms and legs PhA measures ranged from 1.8% to 2.3% in females (n=6) and from 1.6% to 2.5% in males (n=6). In addition to the abovementioned raw BIA parameters derived from series models, parallel equivalents of Xc (i.e., Xcp; Ω) and Cap (pF) were obtained using basic converting formulas described elsewhere (26). Parallel R and Z were not included due to redundancy with series R (27). Series R and both series- and parallel-Xc values were subsequently normalized to stature squared (m 2 ), yielding the R index (RI), the Xc index (XcI), and the Xcp index (XcpI). These variables were also normalized to each segment length (m 2 ), producing RI-specific, XcI-specific, and XcpI-specific, respectively. Anthropometry Body mass (kg) and stature (cm) were measured to the nearest 0.01 kg and 0.1 cm, respectively, using an electronic scale with an integrated stadiometer (SECA 796, Hamburg, Germany), with participants wearing minimal clothing. Body mass index (BMI) was calculated as body mass divided by stature squared (kg/m 2 ). Sitting height was measured to the nearest 0.1 cm with participants seated upright on a 40-cm bench, whereas total upper limb length was measured to the nearest 0.1 cm from the acromion to the tip of the middle finger with the arm fully extended. Total lower limb length was calculated by subtracting trunk length (i.e., sitting height - bench height) from stature. The CVs for stature, sitting height, and arm length measurements remained under 0.3% in both sexes (females, n=6; males, n=6). Body Composition Bone mineral content (BMC, kg), FM (kg), and FFM (kg) were measured using a dual-energy X-ray absorptiometry (DXA) fan-beam densitometer (Hologic Explorer-W, fan-beam densitometer, software QDR for Windows version 12.4, Waltham, MA, USA). Additional information on DXA procedures has been previously described (10). Regional information was derived using Hologic APEX software (Version 3.3.0.1). Regional LST (kg) was obtained by subtracting BMC from the FFM of each limb. The CVs for the dominant and non-dominant arms and legs FM, FFM, and BMC measures ranged from 0.1% to 0.8% in both sexes (females, n=6; males, n=6). Hydration Status A salivary sample was collected from each participant using a non-stimulated passive drool method. Salivary osmolality was subsequently measured by freezing-point depression using a semi-micro osmometer (K-7400, KNAUER, Berlin, Germany), according to procedures described elsewhere (10). In this investigation, the CV for salivary osmolality assessment was 1.1% and 1.3% for 8 and 7 participants, respectively. Statistics Descriptive characteristics were presented as means and standard deviations for each analytical sample. For prospective analyses, absolute differences between baseline and follow-up as well as proportional changes relative to baseline were expressed as percentages [e.g., (PhA at follow-up - PhA at baseline / PhA at baseline) x 100]. Within-participant changes over time were evaluated using paired t-tests (or Wilcoxon signed-rank tests when normality assumptions were not met). Multiple linear regression was used to examine the cross-sectional and prospective associations between regional raw BIA parameters and regional body components. In cross-sectional analysis, both unadjusted (M1) and adjusted (M2) models were applied separately for each sex, adjusting for ethnicity, presence of chronic diseases (described in STable 1 ), BMI, salivary osmolality, fasting time, body temperature, and menstruation status (in females). In prospective analysis, the changes (Δ) in regional raw BIA, and body composition parameters were expressed as the proportion of increase or decrease relative to baseline. In addition, models were run unadjusted (M1) and adjusted (M2) for ethnicity, presence of chronic diseases, BMI, age at baseline, each baseline raw BIA variable, and each body composition measure. To ensure reliable estimations of the predictive value of regional raw BIA parameters, while avoiding model overfitting and instability in the standardized coefficients, several statistical procedures were considered: Sample size calculation required a minimum sample size of n≥ 112 for females and n ≥111 for males in the cross-sectional analysis, and a minimum of n≥ 111 for both females and males in the prospective analysis (based on a 5% type I error and 80% power) (28). Multicollinearity of independent variables was verified using the variance inflation factor (i.e., above 5). This assessment led to the removal of age from the cross-sectional models due to collinearity with raw BIA parameters (e.g., PhA and R), particularly in males. Residuals' linearity, homoscedasticity, and normality were verified through diagram plot analysis. The Benjamini-Hochberg False Discovery Rate (FDR) test was applied across all raw BIA variables within each regional outcome to avoid inflation of type I error resulting from multiple testing (i.e., only associations that remained significant after FDR adjustment were considered valid). To compare the explanatory contribution of RI and RI-specific across adjusted models (semi-partial R 2 ; R 2 Part), Z-Steiger’s Z test and nonparametric bootstrap resampling were used, respectively. All statistical analyses were performed using SPSS software (version 30.0, IBM Corp., Armonk, NY, USA), considering a significance level of α=0.05. Results Descriptive characteristics of anthropometric, raw BIA variables, and body composition of cross-sectional (n = 613, 13.9 ± 2.1y; 52% females) and prospective analytical samples (n = 562, 13.4 ± 2.1y at baseline, 14.4 ± 2.1y at follow-up; 52% females) are provided in STables 2 and 3 , respectively. Regional body composition Lean soft tissue (LST). The adjusted cross-sectional associations showed a highly consistent pattern across sex and limb segments (Fig. 1, panel A ), with RI and RI-specific displaying the strongest inverse associations with LST ( β = -0.879 to -0.598; p < 0.001), whereas Xc-related parameters in series (XcI and XcI-specific; β = -0.499 to -0.209; p < 0.001) and in parallel (XcpI and XcpI-specific; β = -0.828 to -0.504; p < 0.001; Table 1) showed smaller effect sizes. Positive associations were also observed for regional PhA and Cap ( β = 0.283 to 0.838; p < 0.001). In the prospective analyses, ΔRI and ΔRI-specific retained the highest associations with ΔLST across sexes and limb segments (females: β = -0.554 to -0.263; males: β = -0.593 to -0.463; p < 0.001) (Fig. 2, panel A; Table 2). Among Xc-related parameters, ΔXcpI and ΔXcpI-specific showed higher prospective associations with regional ΔLST (females: β = -0.381 to -0.217; males: β = -0.522 to -0.411; p < 0.001) than XcI and XcI-specific. Among ΔPhA and ΔCap, positive associations were observed (females: β < 0.323; males: β < 0.551; p < 0.001), with ΔCap emerging as a relevant long-term informative marker of ΔLST in males ( β = 0.444 to 0.551; p < 0.001). Fat mass (FM). The adjusted cross-sectional associations with regional FM were weaker and more variable than those observed for LST (Fig. 1, panel B ). Although RI and RI-specific showed only very small positive associations in females ( β < 0.163; p < 0.001; in the arms), higher associations were observed in males ( β = 0.346 to 0.490; p < 0.001; in the arms and legs). A similar trend was observed for parallel-transformed Xc variables, with XcpI and XcpI-specific being associated with FM of the arms and legs in females ( β < 0.197) and males ( β = 0.370 to 0.494; both p < 0.001). Likewise, for PhA and Cap, females exhibited a lower magnitude of associations than males (females: β = -0.223 to -0.149; males: β = -0.583 to -0.368; both p 0.05; Fig. 2, panel B ). Bone mineral content (BMC). Both RI and RI-specific showed the strongest inverse associations with BMC across all limbs in females ( β = -0.659 to -0.544) and in males ( β = -0.772 to -0.726; both p < 0.001; Fig. 1, panel C ). Among Xc-related variables, the highest associations were observed between parallel-transformed variables (females: β = -0.544 to -0.398; males: β = -0.708 to -0.642; both p < 0.001) rather than in raw series variables (females: β = -0.261 to -0.162; males: β = -0.415 to -0.363; both p < 0.001). Regarding PhA and Cap, positive associations with BMC were observed, with higher values in males ( β = 0.423 to 0.721) than in females ( β < 0.332; both p ≤ 0.05). In the prospective analysis, ΔR-related variables continue to show the highest associations with ΔBMC of all limbs in both sexes (females: β = -0.303 to -0.136; males: β = -0.484 to -0.331; both p 0.05), all ΔXc-related parameters were negatively associated with ΔBMC (females: β = -0.241 to -0.096; males: β = -0.419 to -0.201; both p < 0.001). In terms of ΔPhA and ΔCap, most of the previously observed positive associations become trivial in females, although they remained valid, particularly for ΔCap in males ( β = 0.110 to 0.375; p < 0.001). The explained variance of each raw BIA variable, by outcome domain, sex, limb involvement, and dominance, is displayed in Figs. 1–2 and STables 4–7. Regarding the comparison of explained variance between RI and RI-specific, no major differences were observed in the variables' ability to explain regional body components in the cross-sectional analysis (Table 3). Nevertheless, in prospective analyses, RI-specific explained variance was approximately twice that of RI for lower-limb ΔLST and ΔBMC in females. In contrast, RI did not outperform RI-specific in any model or sex-specific analysis (all p > 0.05). Although no consistent statistical advantage of ΔRI-specific over ΔRI was observed, sensitivity analyses suggested that this pattern became more evident with advancing biological maturation, with ΔRI-specific showing larger effect sizes than ΔRI in older youth (i.e., after peak height velocity), particularly for non-dominant leg ΔLST in female adolescents and for both leg ΔLST and ΔBMC in males ( STable 8) . Discussion The present investigation provides the first comprehensive evaluation of regional body composition through multiple regional raw BIA variables in youth, encompassing both series and parallel models and indices normalized to stature and segment length. By examining all limb segments (arms and legs) and integrating both cross-sectional and prospective analyses, we demonstrate that most regional raw BIA variables were associated with regional body composition (LST and BMC) in youth, with generally stronger associations observed in males. In addition, the comparison between conventional RI and RI-specific did not reveal a consistent advantage of regional adjustment in explaining regional body composition. Nevertheless, some improvements were observed in specific contexts, particularly in prospective analyses, among females, and among more biologically mature males. While most evidence in the field has focused on whole-body BIA variables, either for prediction equation development or direct physiological interpretation, a concomitant line of work has advanced single- and multifrequency BIA devices capable of providing isolated segmental measurements. Instead of relying on a single current loop (e.g., wrist-ankle), these devices create multiple independent circuits, each isolating a specific body region by controlling where current is injected and where voltage is sensed (i.e., equipotential principle; using four to eight electrode configurations) ( 17 ). A reference investigation ( 29 ) exploring the potential of regional BIA applications (i.e., to date, the tenth-most-cited article in the BIA field ( 1 )) has shown its predictive value for estimating regional LST in middle-aged adults. Since then, only a few investigations, involving predominantly adults and older populations, have developed regional BIA-derived equations with high concordance with DXA-measured LST (r > 0.88) and FM (r > 0.40 ) ( 22 , 23 , 30 ). In youth, an opposite trend is observed, with predictive equations based on regional BIA devices remaining largely absent ( 19 , 31 ). Only a limited number of investigations have explored models incorporating combined scores of regional BIA variables (e.g., the sum of raw BIA variables), showing some potential for estimating whole-body FFM of children ( 32 ). According to the authors, the 8-electrode configuration provided more accurate FFM predictions than 4-electrode systems, suggesting a greater ability to capture intrinsic properties of regional body composition distribution ( 32 ). However, despite these preliminary applications of regional BIA devices in children, no equations currently exist to directly estimate regional body components from raw regional BIA variables ( 19 ). Currently, most research in pediatric populations primarily focuses on the direct interpretation of raw BIA parameters in relation to body composition, thus minimizing model-dependent assumptions and reducing potential sources of estimation error observed in predictive equations (5). In one of the earliest investigations, Fuller et al. ( 24 ) documented a valid relationship between regional raw BIA variables (i.e., Z and Z normalized to segment length) and regional body components (i.e., FFM and FM) in youth. According to the authors, both regional Z indices were more strongly associated with regional FFM than with FM, thus supporting the rationale that the availability of highly metabolizable components with higher water content (FFM) ( 33 ) increases regional conductivity, thereby reducing R ( 34 ). Although a higher proportion of cell membranes would theoretically decrease Xc due to greater capacitive properties ( 34 ), the influence of Xc on Z is known to be comparatively small. As a result, R remains the dominant biophysical determinant of Z, explaining the stronger relationship between Z and FFM observed in their study ( 24 ). Our findings further supported this idea by showing higher cross-sectional and prospective associations between regional raw BIA variables, especially R indices, and regional LST rather than with regional FM. Interesting findings were also documented for the regional BMC. Despite this molecular body component traditionally being considered a poor electrical conductor ( 35 ) and, therefore, expected to show positive associations with R and negative associations with Xc, an opposite trend emerged. This finding is consistent with the close interrelationship between muscle and bone that develops during growth, in which both components adapt in a coordinated, mechanically driven manner ( 36 ). In youth, this relationship often follows allometric scaling ( 37 ), such that increases in skeletal muscle mass are proportionally matched by increases in bone size, strength, and mineral content. As a result, greater BMC in healthy youth coexists with higher proportions of metabolically active, water-rich components, producing association patterns that more closely resemble those of LST than of non-conducting components (e.g., FM) ( 38 ). These observations underscore that, particularly in healthy youth, the electrical behavior of BMC cannot be interpreted in isolation but must be considered in the context of its coupling with other biological components (e.g., LST, body cell mass, skeletal muscle tissue). Accordingly, associations between raw BIA parameters and BMC are likely indirect and should therefore be interpreted with caution. Another finding was that normalizing R to segment length (RI-specific) did not produce meaningful advantages in regional body composition over the conventional approach of normalizing these variables to stature (RI), which remains the most commonly used method in regional BIA ( 22 , 23 , 30 , 32 ). This finding may be interpreted considering that, in youth, R is primarily modulated by segment cross-sectional area (CSA) rather than segment length. Although body segments are typically shorter in length (favoring R), they are characterized by smaller CSA, which contributes to the relatively higher R values that are observed in early ages ( 39 ). This also implies that the assumption of constant tissue resistivity may not fully hold across growth, as changes in body composition (e.g., FFM and LST hydration fraction) may act as moderating factors. Therefore, while additional adjustment for segment length yielded limited physiological gain relative to normalization to stature, which may be advantageous in large-scale studies where time efficiency is relevant, further research is needed to better understand the role of regional CSA in modulating regional R-related indices. Even so, our findings show that these may not apply to all limbs similarly. For instance, we have observed a slight advantage of RI-specific relative to RI when explaining LST and BMC of the lower limbs (2–4% higher R 2 Part), both in females and in more biologically mature males. One possible explanation is that the relevance of length adjustment may vary across body segments according to their morphology. While the CSA of the arms remains relatively small (forearm circumference: 20cm at 10y; 23-26cm at 17y; arm circumference: 21cm at 10y; 25-27cm at 17y), meaning that R is predominantly driven by a limited conductive area, the circumferences of the lower limb (calf and thigh) are expected to increase up to ~ 35cm and to ~ 55cm, respectively, at 17y, making segment length a more physiologically relevant component of the conductive pathway ( 40 ). This idea is theoretically supported by the closer adherence to Pouillet’s law, in which R increases proportionally with the length of the region of interest ( 21 ). Nevertheless, this concept warrants further investigation in older populations and with alternative BIA methodologies (e.g., bioelectrical impedance vector analysis), where segment-specific adjustments to regional impedance variables may provide greater physiological interpretability. Beyond regional R indices, the association patterns observed for regional PhA, Cap, and Xc-related variables followed previous whole-body observations in youth (i.e., higher associations in males) ( 8 , 10 , 11 , 12 ), in which PhA and Cap tended to show positive associations with LST, and negative associations with FM, whereas Xc indices displayed opposite trends. Moreover, we observed that PhA, Cap, and both stature- and segment-normalized Xcp was more consistently associated with regional body composition than series-derived Xc, for which most associations were absent. While these findings reinforce the relevance of PhA as a marker whose regional application can reflect both regional (current investigation) and whole-body composition (i.e., highly metabolically active components) ( 41 ), they also provide the first evidence of the value of examining regional parallel-transformed and derived raw BIA variables (e.g., Xcp and Cap), which exhibited greater alignment with the body's physiological organization ( 10 , 11 , 26 , 27 , 42 ), and appear to offer greater physiological insight than traditional series measures. Nevertheless, future research should extend the application of these raw BIA parameters to regional body composition at higher physiological levels (e.g., cellular-level, body cell mass; tissue-level, skeletal muscle mass). Finally, the near-identical association patterns identified across dominant and non-dominant limbs in both the upper- and lower-body reinforce the methodological stability of regional raw BIA and its ability to capture morphological and functional characteristics independent of habitual limb use. Furthermore, the consistency of the associations observed in the present investigation, even after statistical adjustment for several factors frequently addressed as potential sources of variability in BIA measurements (i.e., hydration, fasting duration, body temperature, and menstrual status in females) ( 18 , 25 ), indicates that these parameters exerted minimal influence on potential informative value of regional raw BIA variables in youth. Consequently, regional BIA appears suitable for large-scale population-based assessments without the need for stringent pre-assessment controls that might otherwise hinder data collection. Nevertheless, more rigorously controlled investigations are required to substantiate these observations further. From a practical perspective, these findings provide a basis for exploring the use of regional raw BIA to identify morphological and functional asymmetries between limbs. Such applications may be particularly relevant during growth and maturation, when developmental imbalances frequently emerge, as well as in young athletic populations, in whom asymmetrical profiles can directly influence performance and are closely linked to injury risk assessment and management ( 43 ). Thus, advancing this line of research may assist the integration of regional BIA variables into diverse field settings (e.g., sports, rehabilitation, and schools). Despite the novelty of this investigation, several limitations should be acknowledged. The findings are generalizable only to healthy youth or those with minor health conditions, and future work is needed to determine whether similar patterns emerge in clinical populations or with other BIA devices capable of regional assessment. Although maturation was estimated using validated population-specific models ( 44 , 45 ), these do not fully replace gold-standard approaches such as radiographic assessments or Tanner staging. This also applies to hydration measurement (i.e., salivary osmolality), as recent guidelines aimed at optimizing hydration-related research recommend using at least two methods (ideally three) to enable more robust conclusions ( 46 ). Nevertheless, several strengths reinforce the robustness of the present findings, including the recruitment of a large and heterogenous cohort, the use of DXA for body composition assessment, the control of hydration, fasting status, temperature, and menstruation at the time of BIA assessments, the derivation of multiple raw BIA indices derived from both series and parallel models and normalized to stature and segment length, and the adoption of a mixed cross-sectional and prospective research design. Conclusion In both cross-sectional and prospective analyses, positive associations were observed for regional PhA and Cap with regional LST and BMC, whereas negative associations were observed for R- and Xc-related variables. R-related indices consistently showed the highest informative value, without a clear advantage of regional normalization (RI-specific) over stature normalization. However, these findings indicate that segment-specific BIA metrics may provide modest additional insight for regional LST and BMC assessment, particularly in the lower limbs. Further research is needed to confirm the applicability of regional data across different BIA devices and clinical populations, and to assess their relevance to regional functional characteristics. Abbreviations BIA Bioelectrical impedance analysis BIVA Bioelectrical impedance vector analysis BMC Bone mineral content BMI Body mass index Cap Capacitance CV Coefficient of variation DXA Dual-energy X-ray absorptiometry FFM Fat-free mass FM Fat mass LST Lean soft tissue PhA Phase angle R Resistance RI Resistance index (normalized to stature) RI-specific Resistance index (normalized to segment length) Xc Capacitive reactance XcI Capacitive reactance index (normalized to stature) XcI-specific Capacitive reactance (normalized to segment length) Xcp Capacitive reactance in parallel model XcpI Capacitive reactance in parallel model (normalized to stature) XcpI-specific Capacitive reactance in parallel model (normalized to segment length) Z Impedance Declarations Acknowledgements We acknowledge Megan Hetherington-Rauth and Tomás Carneiro for their contributions to project preparation; Alexandre Moreira, Jerca Kavcic, Katja Ponikvar, Teresa Ourique, and Viviana Staiano for field data collection; Catarina Nunes, Estela Oliveira, Filipe Jesus, Matilde Ourique, and Vanessa Santos for laboratory data collection; Dulce Batista, Hemi Kondo, Ivo Rosa, Jesselin, Juliana Lameiro, Kevin, Magda Pereira, Mariana Lameiro, and Sofia Boavida for data organization; and Maria Isabel Fragoso for scientific advisory. We also express our gratitude to the Oeiras Valley Municipality and Agrupamento de Escolas de Carnaxide for their commitment and collaboration throughout the project. Funding Statement This work was supported by the Portuguese Foundation for Science and Technology [grant to GBR: 2020.07856.BD; grant to IRC: SFRH/BD/149394/2019; grant to AVB: 2025.02275.BD]. This work is also financed by the Portuguese Foundation for Science and Technology within the unit I&D 472 [grant number UIDB/00447/2020]. Competing Interests If there are no conflicts of interest to declare. Author Contribution GBR and LBS conceived and designed the research protocol. AVB, GBR, IRC, MP, and RF conducted research and data analysis. AVB, GBR, JPM, and RF wrote the first draft of the manuscript. AMS, GBR, HCL, and LBS wrote the second draft of the manuscript. All authors read, commented, and approved the final version of the manuscript. Data Availability Statement The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. References Niu C, Zhang P, Zhang C, Dong J, Liang H, Xiao D, et al. Evolution of research trends and emerging hotspots in bioelectrical impedance analysis over the last two decades: a bibliometric analysis. J Int Soc Sports Nutr. 2025;22(1):2523381. Ward L, Brantlov S. Bioimpedance basics and phase angle fundamentals. Rev Endocr Metab Disord. 2023;24(3):381–91. Lukaski H. Evolution of bioimpedance: a circuitous journey from estimation of physiological function to assessment of body composition and a return to clinical research. Eur J Clin Nutr. 2013;67(1):S2-S9. Sun S, Chumlea W, Heymsfield S, Lukaski H, Schoeller D, Friedl K, et al. Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. Am J Clin Nutr. 2003;77(2):331–40. Barbosa-Silva M, Barros A. Bioelectrical impedance analysis in clinical practice: a new perspective on its use beyond body composition equations. Curr Opin Clin Nutr Metab Care. 2005;8(3):311–7. Bellido D, Garcia-Garcia C, Talluri A, Lukaski H, Garcia-Almeida J. Future lines of research on phase angle: strengths and limitations. Rev Endocr Metab Disord. 2023;24(3):563–83. Barnett A, Bagno S. The physiological mechanisms involved in the clinical measure of phase angle. Am J Physiol - Legacy Content. 1935;114(2):366–82. Martins P, de Lima L, Berria J, Petroski E, da Silva A, Silva D. Association between phase angle and isolated and grouped physical fitness indicators in adolescents. Physiol Behav. 2020;217:112825. De Marco J, de Lima T, Tanimoto P, Miranda C, Bim M, Pelegrini A. Association between phase angle and body composition components in adolescent athletes: a cross-sectional study. Physiol Meas. 2025;46(9). Rosa G, Francisco R, Silva A, Lukaski H, Bernardino A, Sardinha L. Raw bioelectrical impedance parameters as informative markers of body composition in youth. Clin Nutr. 2025;51:266–77. Rosa GB, Francisco R, Silva AM, Lukaski HC, Sardinha LB. Changes in Raw Bioelectrical Impedance Parameters as Markers of Body Composition in Youth: A Prospective Cohort Investigation. Nutrition. 2026:113089. Langer R, da Costa K, Bortolotti H, Fernandes G, de Jesus R, Gonçalves E. Phase angle is associated with cardiorespiratory fitness and body composition in children aged between 9 and 11 years. Physiology & Behavior. 2020;215:112772. Ballarin G, Licenziati M, Alicante P, Di Vincenzo O, Valerio G, Scalfi L. Bioelectrical impedance analysis-derived phase angle and body composition are predictors of health-related fitness in children and adolescents with obesity. Children (Basel). 2022;9(12):1943. Yanchis D, Chandrakumar A, So S, Patterson C, Belza C, Silva C, et al. Utility of phase angle in children with intestinal failure. Clin Nutr. 2025;52:94–102. da Silva B, Orsso C, Gonzalez M, Sicchieri J, Mialich M, Jordao A, et al. Phase angle and cellular health: inflammation and oxidative damage. Rev Endocr Metab Disord. 2023;24(3):543–62. Wilhelm-Leen E, Hall Y, Horwitz R, Chertow G. Phase angle, frailty and mortality in older adults. J Gen Intern Med. 2014;29(1):147–54. Cornish B, Jacobs A, Thomas B, Ward L. Optimizing electrode sites for segmental bioimpedance measurements. Physiol Meas. 1999;20(3):241–50. Dupertuis Y, Jimaja W, Beardsley Levoy C, Genton L. Bioelectrical impedance analysis instruments: how do they differ, what do we need for clinical assessment? Curr Opin Clin Nutr Metab Care. 2025;28(5):379–87. Campa F, Coratella G, Cerullo G, Noriega Z, Francisco R, Charrier D, et al. High-standard predictive equations for estimating body composition using bioelectrical impedance analysis: a systematic review. J Transl Med. 2024;22(1):515. Chula de Castro JA, Lima TRd, Silva DAS. Body composition estimation in children and adolescents by bioelectrical impedance analysis: A systematic review. Journal of Bodywork and Movement Therapies. 2018;22(1):134–46. Pouillet C. Annales de Chimie et de Physique. Paris: Chez Crochard; 1828. Sardinha L, Rosa G, Hetherington-Rauth M, Correia I, Magalhaes J, Silva A, et al. Development and validation of bioelectrical impedance prediction equations estimating regional lean soft tissue mass in middle-aged adults. Eur J Clin Nutr. 2023;77(2):202–11. Silva A, Silva T, Judice P, Rosa G, Bernardino A, Magalhaes J, et al. Regional fat mass estimation using bioelectrical impedance analysis in healthy adults. Br J Nutr. 2025;133(9):1202–10. Fuller N, Fewtrell M, Dewit O, Elia M, Wells J. Segmental bioelectrical impedance analysis in children aged 8–12 y: 2. The assessment of regional body composition and muscle mass. International Journal of Obesity. 2002;26(5):692–700. Brantlov S, Ward LC, Jodal L, Rittig S, Lange A. Critical factors and their impact on bioelectrical impedance analysis in children: a review. J Med Eng Technol. 2017;41(1):22–35. Nyboer J. Workable volume and flow concepts of bio-segments by electrical impedance plethysmography. TIT J Life Sci. 1972;2(1):1–13. Francisco R, Jesus F, Nunes C, Carvalho A, Alvim M, Campa F, et al. Prediction of body water compartments by raw bioelectrical impedance parameters in athletes: Comparison between series and parallel measurements. Scand J Med Sci Sports. 2023;33(10):1998–2008. Green SB. How Many Subjects Does It Take To Do A Regression Analysis. Multivariate Behav Res. 1991;26(3):499–510. Ling CH, de Craen AJ, Slagboom PE, Gunn DA, Stokkel MP, Westendorp RG, et al. Accuracy of direct segmental multi-frequency bioimpedance analysis in the assessment of total body and segmental body composition in middle-aged adult population. Clin Nutr. 2011;30(5):610–5. De Rui M, Veronese N, Bolzetta F, Berton L, Carraro S, Bano G, et al. Validation of bioelectrical impedance analysis for estimating limb lean mass in free-living Caucasian elderly people. Clin Nutr. 2017;36(2):577–84. Kyle U, Earthman C, Pichard C, Coss-Bu J. Body composition during growth in children: limitations and perspectives of bioelectrical impedance analysis. Eur J Clin Nutr. 2015;69(12):1298–305. Kriemler S, Puder J, Zahner L, Roth R, Braun-Fahrlander C, Bedogni G. Cross-validation of bioelectrical impedance analysis for the assessment of body composition in a representative sample of 6- to 13-year-old children. Eur J Clin Nutr. 2009;63(5):619–26. Gutierrez-Marin D, Luque V, Ferre N, Fewtrell MS, Williams JE, Wells JCK. Associations of age and body mass index with hydration and density of fat-free mass from 4 to 22 years. Eur J Clin Nutr. 2019;73(10):1422–30. Rosa G, Lukaski H, Sardinha L. The science of bioelectrical impedance-derived phase angle: insights from body composition in youth. Rev Endocr Metab Disord. 2025;26(4):603–24. Gabriel C, Gabriel S, Corthout E. The dielectric properties of biological tissues: I. Literature survey. Phys Med Biol. 1996;41(11):2231–49. Novotny SA, Warren GL, Hamrick MW. Aging and the muscle-bone relationship. Physiology (Bethesda). 2015;30(1):8–16. Wolff J. Concerning the interrelationship between form and function of the individual parts of the organism. By Julius Wolff, 1900. Clin Orthop Relat Res. 1988(228):2–11. Pollock NK, Laing EM, Baile CA, Hamrick MW, Hall DB, Lewis RD. Is adiposity advantageous for bone strength? A peripheral quantitative computed tomography study in late adolescent females. Am J Clin Nutr. 2007;86(5):1530–8. Bosy-Westphal A, Danielzik S, Dörhöfer R, Piccoli A, Müller MJ. Patterns of bioelectrical impedance vector distribution by body mass index and age: implications for body-composition analysis. Am J Clin Nutr. 2005;82(1):60–8. Kryst Ł, Woronkowicz A, Kowal M, Sobiecki J. Intergenerational changes in limb circumferences in children and adolescents aged 3–18 from Kraków (Poland) from 1983 to 2010. American Journal of Human Biology. 2018;30(5):e23165. de Oliveira N, Iniesta R, Vlachopoulos D, Barker A, Matias C, Gonçalves E, et al. Can regional phase angle reflect changes in physical fitness and body composition from late childhood to adolescence? A cohort study. Nutrition. 2025:113035. Dittmar M, Reber H. Validation of different bioimpedance analyzers for predicting cell mass against whole-body counting of potassium (40K) as a reference method. Am J Hum Biol. 2004;16(6):697–703. Atkins SJ, Bentley I, Hurst HT, Sinclair JK, Hesketh C. The Presence of Bilateral Imbalance of the Lower Limbs in Elite Youth Soccer Players of Different Ages. J Strength Cond Res. 2016;30(4):1007–13. Moore SA, McKay HA, Macdonald H, Nettlefold L, Baxter-Jones AD, Cameron N, et al. Enhancing a Somatic Maturity Prediction Model. Med Sci Sports Exerc. 2015;47(8):1755–64. Sherar LB, Mirwald RL, Baxter-Jones AD, Thomis M. Prediction of adult height using maturity-based cumulative height velocity curves. J Pediatr. 2005;147(4):508–14. Francisco R, Armstrong LE, Silva AM. Recommendations for Optimizing Research Regarding the Effects of Dehydration on Athletic Performance. Sports Medicine. 2025. Tables Tables are available in the Supplementary Files section. Additional Declarations There is NO conflict of interest to disclose. Supplementary Files Tables12.docx Table 1-2 Tables3.docx Table 3 SupplementaryTable1.docx Supplementary Table 1. SupplementaryTable2.docx Supplementary Table 2. SupplementaryTable3.docx Supplementary Table 3. SupplementaryTables47.docx Supplementary Table 4-7. SupplementaryTable8.docx Supplementary Table 8. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 20 Apr, 2026 Editor assigned by journal 13 Apr, 2026 Submission checks completed at journal 10 Apr, 2026 First submitted to journal 02 Apr, 2026 Unknown event 02 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9224431","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":622287273,"identity":"7cdcb42f-371b-4663-8396-f0e872215306","order_by":0,"name":"Gil Rosa","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYDACZuYGECXHwJAApAokGAwIa2EEazGGaDGAaUnApweiJbEBooWBsBZ5d8a2Dx/+2KX3s+cYMPMYWNibszc/YC78gVuL4WHG5pkz25JzZ/a8AWmRSNzZc8yAeQYeWwybGZuZeRuYczfcANsikWBwIwHIIKSF5099uj1Ui73BjfQPeLXIM4O0sB1OMJCAaGGEWIdHiwFQC+PMtuOGM848Kzg4B+yXMwWHZ6ThsaX/8GGGD3+q5fnbkzc+eFNRBwyx9o2PC2zw2HIAiQNnH8atAWhLAzZRZnxaRsEoGAWjYMQBANvGSheYejGNAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-9358-4820","institution":"Faculdade de Motricidade Humana, Universidade de Lisboa","correspondingAuthor":true,"prefix":"","firstName":"Gil","middleName":"","lastName":"Rosa","suffix":""},{"id":622287274,"identity":"969800e7-efc0-47f7-a74f-808a52aaf1c6","order_by":1,"name":"Ruben Francisco","email":"","orcid":"https://orcid.org/0000-0002-6413-5268","institution":"Faculty of Human Kinetics, University of Lisbon","correspondingAuthor":false,"prefix":"","firstName":"Ruben","middleName":"","lastName":"Francisco","suffix":""},{"id":622287276,"identity":"d0b1eb2e-f187-4809-a9bf-ea143c971d24","order_by":2,"name":"Inês Correia","email":"","orcid":"https://orcid.org/0000-0002-3246-8965","institution":"Faculdade de Motricidade Humana, Universidade de Lisboa","correspondingAuthor":false,"prefix":"","firstName":"Inês","middleName":"","lastName":"Correia","suffix":""},{"id":622287281,"identity":"5fd04092-92c2-4c77-aaa9-20a561e26627","order_by":3,"name":"Ana Bernardino","email":"","orcid":"","institution":"Faculty of Human Kinetics, University of Lisbon","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"","lastName":"Bernardino","suffix":""},{"id":622287282,"identity":"d44378fd-833b-4ae0-bc56-ed7dfad71409","order_by":4,"name":"João Magalhães","email":"","orcid":"","institution":"Faculdade de Motricidade Humana, Universidade de Lisboa","correspondingAuthor":false,"prefix":"","firstName":"João","middleName":"","lastName":"Magalhães","suffix":""},{"id":622287284,"identity":"835262a0-0ae6-46cd-9ea5-b752b4014e01","order_by":5,"name":"Martim Pires","email":"","orcid":"","institution":"Faculdade de Motricidade Humana, Universidade de Lisboa","correspondingAuthor":false,"prefix":"","firstName":"Martim","middleName":"","lastName":"Pires","suffix":""},{"id":622287285,"identity":"caefff20-5768-46c1-974a-acc29f8cd033","order_by":6,"name":"Henry Lukaski","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Henry","middleName":"","lastName":"Lukaski","suffix":""},{"id":622287286,"identity":"e5db9783-b2e9-456f-b055-7580931d693d","order_by":7,"name":"Analiza Silva","email":"","orcid":"https://orcid.org/0000-0002-8984-8600","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Analiza","middleName":"","lastName":"Silva","suffix":""},{"id":622287287,"identity":"99c81ac8-1136-46cb-94e1-a338cc620d63","order_by":8,"name":"Luis Sardinha","email":"","orcid":"","institution":"Faculdade de Motricidade Humana, Universidade de Lisboa","correspondingAuthor":false,"prefix":"","firstName":"Luis","middleName":"","lastName":"Sardinha","suffix":""}],"badges":[],"createdAt":"2026-03-25 14:31:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9224431/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9224431/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109249528,"identity":"21efcf52-4b31-4af5-a2b8-78bdd7818b36","added_by":"auto","created_at":"2026-05-14 08:55:24","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":3880862,"visible":true,"origin":"","legend":"\u003cp\u003eSex- and limb-specific cross-sectional associations (adjusted) between regional BIA parameters and DXA-derived lean soft tissue (A), fat mass (B), and bone mineral content (C), displayed as radar charts (n=613).\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/0aa58e495388adcec41573c3.jpg"},{"id":108383929,"identity":"d71aecfd-33fb-4846-9642-d8ba55c828c0","added_by":"auto","created_at":"2026-05-04 05:50:27","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":3770663,"visible":true,"origin":"","legend":"\u003cp\u003eSex- and limb-specific prospective associations (adjusted) between changes in regional BIA parameters and changes in DXA-derived lean soft tissue (A), fat mass (B), and bone mineral content (C), displayed as radar charts (n=562).\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/0c4abcf1bf462759f95fdd4c.jpg"},{"id":109250096,"identity":"78fad524-f26c-4df5-a79d-ed845a6907c0","added_by":"auto","created_at":"2026-05-14 09:06:32","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2719540,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/6e9dd7ad-8640-45d6-bec1-567bec55cc75.pdf"},{"id":108383927,"identity":"8f2c9eac-9b84-433f-af01-d9381708c369","added_by":"auto","created_at":"2026-05-04 05:50:27","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":437398,"visible":true,"origin":"","legend":"Table 1-2","description":"","filename":"Tables12.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/9c516911c159e78ca515b6a4.docx"},{"id":108493262,"identity":"eb0acc3d-b2f4-40bd-b056-2d502ba16105","added_by":"auto","created_at":"2026-05-05 09:59:48","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":21038,"visible":true,"origin":"","legend":"Table 3","description":"","filename":"Tables3.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/9b7be8b7d40570834ecaa6fa.docx"},{"id":108383931,"identity":"70f8347e-b131-4cd9-8087-fd8a14c031fe","added_by":"auto","created_at":"2026-05-04 05:50:27","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":18402,"visible":true,"origin":"","legend":"Supplementary Table 1.","description":"","filename":"SupplementaryTable1.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/f9bfbabdb334086c5d1f2b17.docx"},{"id":108493444,"identity":"73016b9e-4e74-4d84-ad98-1b4cb317b8b6","added_by":"auto","created_at":"2026-05-05 10:00:24","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":373762,"visible":true,"origin":"","legend":"Supplementary Table 2.","description":"","filename":"SupplementaryTable2.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/206af36b61912a5240617135.docx"},{"id":108804504,"identity":"14c40c41-13fb-4858-826c-425612a40efd","added_by":"auto","created_at":"2026-05-08 15:20:58","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":33273,"visible":true,"origin":"","legend":"Supplementary Table 3.","description":"","filename":"SupplementaryTable3.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/c834dcb97ade7e355b204e7a.docx"},{"id":108804477,"identity":"920b40e1-e2e8-4c64-a589-a1346a6842d0","added_by":"auto","created_at":"2026-05-08 15:20:40","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":168737,"visible":true,"origin":"","legend":"Supplementary Table 4-7.","description":"","filename":"SupplementaryTables47.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/eada9e436f1298775928a9c5.docx"},{"id":108383934,"identity":"962d0a22-77eb-4c7d-9ae6-caf33e6e4307","added_by":"auto","created_at":"2026-05-04 05:50:27","extension":"docx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":20822,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Table 8.\u003c/p\u003e","description":"","filename":"SupplementaryTable8.docx","url":"https://assets-eu.researchsquare.com/files/rs-9224431/v1/dc3abc0d6801e0f40b2b361d.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose.","formattedTitle":"Advancing the Assessment of Regional Body Composition in Children and Adolescents Using Regional Raw Bioelectrical Impedance Data","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBioelectrical impedance analysis (BIA) has become increasingly relevant across clinical and research fields due to its ability to characterize and discriminate indicators of health, nutritional status, body composition, physical function, and even mortality (1). As a practical alternative to costly and less accessible methods, BIA has traditionally been implemented through prediction equations to estimate health-related outcomes (2, 3). However, these equations are susceptible to latent sources of error arising from population and BIA device specificity, hydration variability, and multi-step modeling assumptions (4, 5).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOver the last two decades, there has been a growing interest in the direct interpretation of raw BIA variables (1), which may provide more physiologically established information (5, 6). Among raw BIA variables, particular attention has been given to whole-body phase angle (PhA), which represents the physiological ratio between capacitive reactance (Xc, the delay of alternating current penetrating cell membrane) and resistance (R, the direct opposition to current flow) (7). PhA is expressed as the arctangent of Xc/R x 180/\u0026pi;. In pediatric populations, whole-body PhA measured mostly at 50kHz has been shown to be useful for informing body composition mostly measured at the molecular level [e.g., lean soft tissue (LST), fat-free mass (FFM), and fat mass (FM)] (8, 9, 10, 11) and physical fitness (8, 12, 13, 14). Owing to its consistent associations with morphofunctional characteristics across diverse health contexts in youth, PhA is gaining recognition as a holistic indicator of health (15, 16).\u003c/p\u003e\n\u003cp\u003eBeyond whole-body information, modern regional BIA devices now allow the acquisition of raw BIA parameters from individual body segments. This approach is based on the principle of equipotentials (17), which allows automatic selection of multiple current pathways without the need for electrode repositioning (18). Particularly among children and adolescents, there is a biological rationale for exploring the usefulness of these regional raw BIA parameters, as rapid growth and maturation lead to limb-specific morphological and functional changes that whole-body measures may not capture. These adaptations include longitudinal bone elongation, regional LST or FM accretion, and progressive improvements in neuromuscular performance. Accordingly, the extent to which regional PhA and other raw BIA parameters explain such limb-specific domains of body composition and physical function remains insufficiently investigated in youth (19, 20).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMoreover, the physiological interpretation of regional BIA measurements not only on their direct acquisition but also on how raw BIA parameters are normalized to reflect segment-specific conductive pathways. To date, most evidence has relied on standardizing raw BIA variables to stature, under the rationale that stature normalization accounts for the conductive pathway length (i.e., following Pouillet\u0026apos;s equation (21)), thus improving interpretability relative to absolute values (10). Accordingly, raw BIA stature-normalized data [e.g., impedance (Z) and R] have been used to develop predictive equations for regional LST and FM (22, 23). However, this approach assumes that total body height adequately represents the electrically relevant conductive pathway, an assumption that may not hold when the objective is true segmental assessment and may attenuate physiological specificity and obscure region-specific associations. From a biophysical standpoint, the conductive pathway of interest in regional evaluation is more plausibly defined by the anatomical length of the specific limb assessed rather than overall stature (24). Therefore, adjusting raw BIA variables to segment length may provide a more anatomically coherent representation of regional characteristics. Whether segmental length adjustment of raw BIA variables yields greater physiological relevance than conventional stature normalization in a regional approach remains to be tested.\u003c/p\u003e\n\u003cp\u003eTo further expand the potential applications of regional BIA in youth, the present aim of this investigation is to determine whether distinct regional raw BIA parameters provide cross-sectional and prospective information on limb-specific body composition in a large cohort of children and adolescents. In addition, it also aims to examine whether normalization of regional R to limb segment length (RI-specific) offers greater informational value than traditional stature-normalized adjustment (RI).\u003c/p\u003e"},{"header":"Materials \u0026 Methods","content":"\u003cp\u003e\u003cem\u003eSample and study design\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA mixed cross-sectional and prospective design was used to assess 681 children and adolescents (49% females) enrolled in the PRO-BODY project (10). The participants were assessed between September 2022 and July 2023 (Moment 1, baseline) and between September 2023 and July 2024 (Moment 2, follow-up), and were excluded if they had sensory, motor, or intellectual impairments, were institutionalized, or presented a mismatch between chronological age and biological maturation status. Additional information on the investigation protocol, including detailed inclusion and exclusion criteria and ethical approval, has been reported elsewhere (10, 11). Written informed consent was obtained from legal guardians, and verbal assent was provided by all participants.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBioelectrical impedance analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eA phase-sensitive BIA device (BIA 101 BIVA PRO, Akern, Florence, Italy) with eight-point Ag/AgCI low Z [i.e., \u0026lt;30 ohms (Ω)] electrodes (BIATRODES\u003csup\u003e\u0026reg;\u003c/sup\u003e, Akern Srl, Florence, Italy) was used to measure PhA (\u0026ordm;), R (Ω), and Xc (Ω). The BIA device was calibrated each morning using the manufacturer\u0026rsquo;s standard control circuit (R = 380 Ω; Xc = 45 Ω). Following verification of pre-test conditions (i.e., minimum 4-hour fast, no structured, vigorous-intensity physical activity in the previous 12 hours), and after instructing participants to empty their bladder and to remove all metal accessories, the technician cleaned the electrode sites with 70% isopropyl alcohol and applied four emitting and four sensing electrodes according to standardized procedures (25). Assessments were performed after a 5-minute resting period using an alternating current of 245 \u0026mu;A RMS at 50 kHz, delivered through the manufacturer-defined regional reading channels. The participants were lying supine on a non-conductive surface, with a leg opening of 45\u0026deg; and arms positioned at 30\u0026deg; from the trunk, in a room free from electromagnetic interference and with an ambient temperature between 22\u0026ordm;C and 25\u0026ordm;C. In the present investigation, the coefficients of variation (CV) for the dominant and non-dominant arms and legs PhA measures ranged from 1.8% to 2.3% in females (n=6) and from 1.6% to 2.5% in males (n=6).\u003c/p\u003e\n\u003cp\u003eIn addition to the abovementioned raw BIA parameters derived from series models, parallel equivalents of Xc (i.e., Xcp; Ω) and Cap (pF) were obtained using basic converting formulas described elsewhere (26). Parallel R and Z were not included due to redundancy with series R (27). Series R and both series- and parallel-Xc values were subsequently normalized to stature squared (m\u003csup\u003e2\u003c/sup\u003e), yielding the R index (RI), the Xc index (XcI), and the Xcp index (XcpI). These variables were also normalized to each segment length (m\u003csup\u003e2\u003c/sup\u003e), producing RI-specific, XcI-specific, and XcpI-specific, respectively.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAnthropometry\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eBody mass (kg) and stature (cm) were measured to the nearest 0.01 kg and 0.1 cm, respectively, using an electronic scale with an integrated stadiometer (SECA 796, Hamburg, Germany), with participants wearing minimal clothing. Body mass index (BMI) was calculated as body mass divided by stature squared (kg/m\u003csup\u003e2\u003c/sup\u003e). Sitting height was measured to the nearest 0.1 cm with participants seated upright on a 40-cm bench, whereas total upper limb length was measured to the nearest 0.1 cm from the acromion to the tip of the middle finger with the arm fully extended. Total lower limb length was calculated by subtracting trunk length (i.e., sitting height - bench height) from stature. The CVs for stature, sitting height, and arm length measurements remained under 0.3% in both sexes (females, n=6; males, n=6).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eBody Composition\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBone mineral content (BMC, kg), FM (kg), and FFM (kg) were measured using a dual-energy X-ray absorptiometry (DXA) fan-beam densitometer (Hologic Explorer-W, fan-beam densitometer, software QDR for Windows version 12.4, Waltham, MA, USA). Additional information on DXA procedures has been previously described (10). Regional information was derived using Hologic APEX software (Version 3.3.0.1). Regional LST (kg) was obtained by subtracting BMC from the FFM of each limb. The CVs for the dominant and non-dominant arms and legs FM, FFM, and BMC measures ranged from 0.1% to 0.8% in both sexes (females, n=6; males, n=6).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eHydration Status\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA salivary sample was collected from each participant using a non-stimulated passive drool method. Salivary osmolality was subsequently measured by freezing-point depression using a semi-micro osmometer (K-7400, KNAUER, Berlin, Germany), according to procedures described elsewhere (10). In this investigation, the CV for salivary osmolality assessment was 1.1% and 1.3% for 8 and 7 participants, respectively.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eStatistics\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDescriptive characteristics were presented as means and standard deviations for each analytical sample. For prospective analyses, absolute differences between baseline and follow-up as well as proportional changes relative to baseline were expressed as percentages [e.g., (PhA at follow-up - PhA at baseline / PhA at baseline) x 100]. Within-participant changes over time were evaluated using paired t-tests (or Wilcoxon signed-rank tests when normality assumptions were not met).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMultiple linear regression was used to examine the cross-sectional and prospective associations between regional raw BIA parameters and regional body components. In cross-sectional analysis, both unadjusted (M1) and adjusted (M2) models were applied separately for each sex, adjusting for ethnicity, presence of chronic diseases (described in \u003cstrong\u003eSTable 1\u003c/strong\u003e), BMI, salivary osmolality, fasting time, body temperature, and menstruation status (in females). In prospective analysis, the changes (\u0026Delta;) in regional raw BIA, and body composition parameters were expressed as the proportion of increase or decrease relative to baseline. In addition, models were run unadjusted (M1) and adjusted (M2) for ethnicity, presence of chronic diseases, BMI, age at baseline, each baseline raw BIA variable, and each body composition measure. To ensure reliable estimations of the predictive value of regional raw BIA parameters, while avoiding model overfitting and instability in the standardized coefficients, several statistical procedures were considered:\u003c/p\u003e\n\u003cul class=\"decimal_type\"\u003e\n \u003cli\u003eSample size calculation required a minimum sample size of n\u0026ge; 112 for females and n \u0026ge;111 for males in the cross-sectional analysis, and a minimum of n\u0026ge; 111 for both females and males in the prospective analysis (based on a 5% type I error and 80% power) (28).\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMulticollinearity of independent variables was verified using the variance inflation factor (i.e., above 5). This assessment led to the removal of age from the cross-sectional models due to collinearity with raw BIA parameters (e.g., PhA and R), particularly in males. Residuals\u0026apos; linearity, homoscedasticity, and normality were verified through diagram plot analysis.\u003c/li\u003e\n \u003cli\u003eThe Benjamini-Hochberg False Discovery Rate (FDR) test was applied across all raw BIA variables within each regional outcome to avoid inflation of type I error resulting from multiple testing (i.e., only associations that remained significant after FDR adjustment were considered valid).\u003c/li\u003e\n \u003cli\u003eTo compare the explanatory contribution of RI and RI-specific across adjusted models (semi-partial R\u003csup\u003e2\u003c/sup\u003e; R\u003csup\u003e2\u003c/sup\u003ePart), Z-Steiger\u0026rsquo;s Z test and nonparametric bootstrap resampling were used, respectively.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eAll statistical analyses were performed using SPSS software (version 30.0, IBM Corp., Armonk, NY, USA), considering a significance level of \u0026alpha;=0.05.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eDescriptive characteristics of anthropometric, raw BIA variables, and body composition of cross-sectional (n = 613, 13.9 ± 2.1y; 52% females) and prospective analytical samples (n = 562, 13.4 ± 2.1y at baseline, 14.4 ± 2.1y at follow-up; 52% females) are provided in \u003cstrong\u003eSTables 2 and 3\u003c/strong\u003e, respectively.\u003c/p\u003e\n\u003ch3\u003eRegional body composition\u003c/h3\u003e\n\u003cp\u003e\u003cem\u003eLean soft tissue (LST).\u003c/em\u003e The adjusted cross-sectional associations showed a highly consistent pattern across sex and limb segments (Fig. 1, \u003cstrong\u003epanel A\u003c/strong\u003e), with RI and RI-specific displaying the strongest inverse associations with LST (\u003cem\u003eβ\u003c/em\u003e = -0.879 to -0.598; p \u0026lt; 0.001), whereas Xc-related parameters in series (XcI and XcI-specific; \u003cem\u003eβ\u003c/em\u003e = -0.499 to -0.209; p \u0026lt; 0.001) and in parallel (XcpI and XcpI-specific; \u003cem\u003eβ\u003c/em\u003e = -0.828 to -0.504; p \u0026lt; 0.001; Table 1) showed smaller effect sizes. Positive associations were also observed for regional PhA and Cap (\u003cem\u003eβ\u003c/em\u003e = 0.283 to 0.838; p \u0026lt; 0.001). In the prospective analyses, ΔRI and ΔRI-specific retained the highest associations with ΔLST across sexes and limb segments (females: \u003cem\u003eβ\u003c/em\u003e = -0.554 to -0.263; males: \u003cem\u003eβ\u003c/em\u003e = -0.593 to -0.463; p \u0026lt; 0.001) (Fig. 2, \u003cstrong\u003epanel A;\u003c/strong\u003e Table 2). Among Xc-related parameters, ΔXcpI and ΔXcpI-specific showed higher prospective associations with regional ΔLST (females: \u003cem\u003eβ\u003c/em\u003e = -0.381 to -0.217; males: \u003cem\u003eβ\u003c/em\u003e = -0.522 to -0.411; p \u0026lt; 0.001) than XcI and XcI-specific. Among ΔPhA and ΔCap, positive associations were observed (females: \u003cem\u003eβ\u003c/em\u003e \u0026lt; 0.323; males: \u003cem\u003eβ\u003c/em\u003e \u0026lt; 0.551; p \u0026lt; 0.001), with ΔCap emerging as a relevant long-term informative marker of ΔLST in males (\u003cem\u003eβ\u003c/em\u003e = 0.444 to 0.551; p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eFat mass (FM).\u003c/em\u003e The adjusted cross-sectional associations with regional FM were weaker and more variable than those observed for LST (Fig. 1, \u003cstrong\u003epanel B\u003c/strong\u003e). Although RI and RI-specific showed only very small positive associations in females (\u003cem\u003eβ\u003c/em\u003e \u0026lt; 0.163; p \u0026lt; 0.001; in the arms), higher associations were observed in males (\u003cem\u003eβ\u003c/em\u003e = 0.346 to 0.490; p \u0026lt; 0.001; in the arms and legs). A similar trend was observed for parallel-transformed Xc variables, with XcpI and XcpI-specific being associated with FM of the arms and legs in females (\u003cem\u003eβ\u003c/em\u003e \u0026lt; 0.197) and males (\u003cem\u003eβ\u003c/em\u003e = 0.370 to 0.494; both p \u0026lt; 0.001). Likewise, for PhA and Cap, females exhibited a lower magnitude of associations than males (females: \u003cem\u003eβ\u003c/em\u003e = -0.223 to -0.149; males: \u003cem\u003eβ\u003c/em\u003e = -0.583 to -0.368; both p \u0026lt; 0.001). At the prospective level, all associations between changes in raw BIA variables and ΔFM became negligible (p \u0026gt; 0.05; Fig. 2, \u003cstrong\u003epanel B\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eBone mineral content (BMC).\u003c/em\u003e Both RI and RI-specific showed the strongest inverse associations with BMC across all limbs in females (\u003cem\u003eβ\u003c/em\u003e = -0.659 to -0.544) and in males (\u003cem\u003eβ\u003c/em\u003e = -0.772 to -0.726; both p \u0026lt; 0.001; Fig.\u0026nbsp;1, \u003cstrong\u003epanel C\u003c/strong\u003e). Among Xc-related variables, the highest associations were observed between parallel-transformed variables (females: \u003cem\u003eβ\u003c/em\u003e = -0.544 to -0.398; males: \u003cem\u003eβ\u003c/em\u003e = -0.708 to -0.642; both p \u0026lt; 0.001) rather than in raw series variables (females: \u003cem\u003eβ\u003c/em\u003e = -0.261 to -0.162; males: \u003cem\u003eβ\u003c/em\u003e = -0.415 to -0.363; both p \u0026lt; 0.001). Regarding PhA and Cap, positive associations with BMC were observed, with higher values in males (\u003cem\u003eβ\u003c/em\u003e = 0.423 to 0.721) than in females (\u003cem\u003eβ\u003c/em\u003e \u0026lt; 0.332; both p ≤ 0.05). In the prospective analysis, ΔR-related variables continue to show the highest associations with ΔBMC of all limbs in both sexes (females: \u003cem\u003eβ\u003c/em\u003e = -0.303 to -0.136; males: \u003cem\u003eβ\u003c/em\u003e = -0.484 to -0.331; both p \u0026lt; 0.001) (Fig.\u0026nbsp;2, \u003cstrong\u003epanel C\u003c/strong\u003e). With the exception of dominant leg ΔXcI in females (p \u0026gt; 0.05), all ΔXc-related parameters were negatively associated with ΔBMC (females: \u003cem\u003eβ\u003c/em\u003e = -0.241 to -0.096; males: \u003cem\u003eβ\u003c/em\u003e = -0.419 to -0.201; both p \u0026lt; 0.001). In terms of ΔPhA and ΔCap, most of the previously observed positive associations become trivial in females, although they remained valid, particularly for ΔCap in males (\u003cem\u003eβ\u003c/em\u003e = 0.110 to 0.375; p \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003eThe explained variance of each raw BIA variable, by outcome domain, sex, limb involvement, and dominance, is displayed in Figs. 1–2 and \u003cstrong\u003eSTables 4–7.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRegarding the comparison of explained variance between RI and RI-specific, no major differences were observed in the variables' ability to explain regional body components in the cross-sectional analysis (Table\u0026nbsp;3). Nevertheless, in prospective analyses, RI-specific explained variance was approximately twice that of RI for lower-limb ΔLST and ΔBMC in females. In contrast, RI did not outperform RI-specific in any model or sex-specific analysis (all p \u0026gt; 0.05). Although no consistent statistical advantage of ΔRI-specific over ΔRI was observed, sensitivity analyses suggested that this pattern became more evident with advancing biological maturation, with ΔRI-specific showing larger effect sizes than ΔRI in older youth (i.e., after peak height velocity), particularly for non-dominant leg ΔLST in female adolescents and for both leg ΔLST and ΔBMC in males (\u003cstrong\u003eSTable 8)\u003c/strong\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present investigation provides the first comprehensive evaluation of regional body composition through multiple regional raw BIA variables in youth, encompassing both series and parallel models and indices normalized to stature and segment length. By examining all limb segments (arms and legs) and integrating both cross-sectional and prospective analyses, we demonstrate that most regional raw BIA variables were associated with regional body composition (LST and BMC) in youth, with generally stronger associations observed in males. In addition, the comparison between conventional RI and RI-specific did not reveal a consistent advantage of regional adjustment in explaining regional body composition. Nevertheless, some improvements were observed in specific contexts, particularly in prospective analyses, among females, and among more biologically mature males.\u003c/p\u003e \u003cp\u003eWhile most evidence in the field has focused on whole-body BIA variables, either for prediction equation development or direct physiological interpretation, a concomitant line of work has advanced single- and multifrequency BIA devices capable of providing isolated segmental measurements. Instead of relying on a single current loop (e.g., wrist-ankle), these devices create multiple independent circuits, each isolating a specific body region by controlling where current is injected and where voltage is sensed (i.e., equipotential principle; using four to eight electrode configurations) (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). A reference investigation (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e) exploring the potential of regional BIA applications (i.e., to date, the tenth-most-cited article in the BIA field (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e)) has shown its predictive value for estimating regional LST in middle-aged adults. Since then, only a few investigations, involving predominantly adults and older populations, have developed regional BIA-derived equations with high concordance with DXA-measured LST (r\u0026thinsp;\u0026gt;\u0026thinsp;0.88) and FM (r\u0026thinsp;\u0026gt;\u0026thinsp;0.40 ) (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn youth, an opposite trend is observed, with predictive equations based on regional BIA devices remaining largely absent (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e). Only a limited number of investigations have explored models incorporating combined scores of regional BIA variables (e.g., the sum of raw BIA variables), showing some potential for estimating whole-body FFM of children (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). According to the authors, the 8-electrode configuration provided more accurate FFM predictions than 4-electrode systems, suggesting a greater ability to capture intrinsic properties of regional body composition distribution (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). However, despite these preliminary applications of regional BIA devices in children, no equations currently exist to directly estimate regional body components from raw regional BIA variables (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCurrently, most research in pediatric populations primarily focuses on the direct interpretation of raw BIA parameters in relation to body composition, thus minimizing model-dependent assumptions and reducing potential sources of estimation error observed in predictive equations (5). In one of the earliest investigations, Fuller et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) documented a valid relationship between regional raw BIA variables (i.e., Z and Z normalized to segment length) and regional body components (i.e., FFM and FM) in youth. According to the authors, both regional Z indices were more strongly associated with regional FFM than with FM, thus supporting the rationale that the availability of highly metabolizable components with higher water content (FFM) (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) increases regional conductivity, thereby reducing R (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e). Although a higher proportion of cell membranes would theoretically decrease Xc due to greater capacitive properties (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e), the influence of Xc on Z is known to be comparatively small. As a result, R remains the dominant biophysical determinant of Z, explaining the stronger relationship between Z and FFM observed in their study (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e). Our findings further supported this idea by showing higher cross-sectional and prospective associations between regional raw BIA variables, especially R indices, and regional LST rather than with regional FM.\u003c/p\u003e \u003cp\u003eInteresting findings were also documented for the regional BMC. Despite this molecular body component traditionally being considered a poor electrical conductor (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e) and, therefore, expected to show positive associations with R and negative associations with Xc, an opposite trend emerged. This finding is consistent with the close interrelationship between muscle and bone that develops during growth, in which both components adapt in a coordinated, mechanically driven manner (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e). In youth, this relationship often follows allometric scaling (\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e), such that increases in skeletal muscle mass are proportionally matched by increases in bone size, strength, and mineral content. As a result, greater BMC in healthy youth coexists with higher proportions of metabolically active, water-rich components, producing association patterns that more closely resemble those of LST than of non-conducting components (e.g., FM) (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e). These observations underscore that, particularly in healthy youth, the electrical behavior of BMC cannot be interpreted in isolation but must be considered in the context of its coupling with other biological components (e.g., LST, body cell mass, skeletal muscle tissue). Accordingly, associations between raw BIA parameters and BMC are likely indirect and should therefore be interpreted with caution.\u003c/p\u003e \u003cp\u003eAnother finding was that normalizing R to segment length (RI-specific) did not produce meaningful advantages in regional body composition over the conventional approach of normalizing these variables to stature (RI), which remains the most commonly used method in regional BIA (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e). This finding may be interpreted considering that, in youth, R is primarily modulated by segment cross-sectional area (CSA) rather than segment length. Although body segments are typically shorter in length (favoring R), they are characterized by smaller CSA, which contributes to the relatively higher R values that are observed in early ages (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e). This also implies that the assumption of constant tissue resistivity may not fully hold across growth, as changes in body composition (e.g., FFM and LST hydration fraction) may act as moderating factors. Therefore, while additional adjustment for segment length yielded limited physiological gain relative to normalization to stature, which may be advantageous in large-scale studies where time efficiency is relevant, further research is needed to better understand the role of regional CSA in modulating regional R-related indices.\u003c/p\u003e \u003cp\u003eEven so, our findings show that these may not apply to all limbs similarly. For instance, we have observed a slight advantage of RI-specific relative to RI when explaining LST and BMC of the lower limbs (2\u0026ndash;4% higher R\u003csup\u003e2\u003c/sup\u003ePart), both in females and in more biologically mature males. One possible explanation is that the relevance of length adjustment may vary across body segments according to their morphology. While the CSA of the arms remains relatively small (forearm circumference: 20cm at 10y; 23-26cm at 17y; arm circumference: 21cm at 10y; 25-27cm at 17y), meaning that R is predominantly driven by a limited conductive area, the circumferences of the lower limb (calf and thigh) are expected to increase up to ~\u0026thinsp;35cm and to ~\u0026thinsp;55cm, respectively, at 17y, making segment length a more physiologically relevant component of the conductive pathway (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e). This idea is theoretically supported by the closer adherence to Pouillet\u0026rsquo;s law, in which R increases proportionally with the length of the region of interest (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Nevertheless, this concept warrants further investigation in older populations and with alternative BIA methodologies (e.g., bioelectrical impedance vector analysis), where segment-specific adjustments to regional impedance variables may provide greater physiological interpretability.\u003c/p\u003e \u003cp\u003eBeyond regional R indices, the association patterns observed for regional PhA, Cap, and Xc-related variables followed previous whole-body observations in youth (i.e., higher associations in males) (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e), in which PhA and Cap tended to show positive associations with LST, and negative associations with FM, whereas Xc indices displayed opposite trends. Moreover, we observed that PhA, Cap, and both stature- and segment-normalized Xcp was more consistently associated with regional body composition than series-derived Xc, for which most associations were absent. While these findings reinforce the relevance of PhA as a marker whose regional application can reflect both regional (current investigation) and whole-body composition (i.e., highly metabolically active components) (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e), they also provide the first evidence of the value of examining regional parallel-transformed and derived raw BIA variables (e.g., Xcp and Cap), which exhibited greater alignment with the body's physiological organization (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e), and appear to offer greater physiological insight than traditional series measures. Nevertheless, future research should extend the application of these raw BIA parameters to regional body composition at higher physiological levels (e.g., cellular-level, body cell mass; tissue-level, skeletal muscle mass).\u003c/p\u003e \u003cp\u003eFinally, the near-identical association patterns identified across dominant and non-dominant limbs in both the upper- and lower-body reinforce the methodological stability of regional raw BIA and its ability to capture morphological and functional characteristics independent of habitual limb use. Furthermore, the consistency of the associations observed in the present investigation, even after statistical adjustment for several factors frequently addressed as potential sources of variability in BIA measurements (i.e., hydration, fasting duration, body temperature, and menstrual status in females) (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e), indicates that these parameters exerted minimal influence on potential informative value of regional raw BIA variables in youth. Consequently, regional BIA appears suitable for large-scale population-based assessments without the need for stringent pre-assessment controls that might otherwise hinder data collection. Nevertheless, more rigorously controlled investigations are required to substantiate these observations further.\u003c/p\u003e \u003cp\u003eFrom a practical perspective, these findings provide a basis for exploring the use of regional raw BIA to identify morphological and functional asymmetries between limbs. Such applications may be particularly relevant during growth and maturation, when developmental imbalances frequently emerge, as well as in young athletic populations, in whom asymmetrical profiles can directly influence performance and are closely linked to injury risk assessment and management (\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e). Thus, advancing this line of research may assist the integration of regional BIA variables into diverse field settings (e.g., sports, rehabilitation, and schools).\u003c/p\u003e \u003cp\u003eDespite the novelty of this investigation, several limitations should be acknowledged. The findings are generalizable only to healthy youth or those with minor health conditions, and future work is needed to determine whether similar patterns emerge in clinical populations or with other BIA devices capable of regional assessment. Although maturation was estimated using validated population-specific models (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e), these do not fully replace gold-standard approaches such as radiographic assessments or Tanner staging. This also applies to hydration measurement (i.e., salivary osmolality), as recent guidelines aimed at optimizing hydration-related research recommend using at least two methods (ideally three) to enable more robust conclusions (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e). Nevertheless, several strengths reinforce the robustness of the present findings, including the recruitment of a large and heterogenous cohort, the use of DXA for body composition assessment, the control of hydration, fasting status, temperature, and menstruation at the time of BIA assessments, the derivation of multiple raw BIA indices derived from both series and parallel models and normalized to stature and segment length, and the adoption of a mixed cross-sectional and prospective research design.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn both cross-sectional and prospective analyses, positive associations were observed for regional PhA and Cap with regional LST and BMC, whereas negative associations were observed for R- and Xc-related variables. R-related indices consistently showed the highest informative value, without a clear advantage of regional normalization (RI-specific) over stature normalization. However, these findings indicate that segment-specific BIA metrics may provide modest additional insight for regional LST and BMC assessment, particularly in the lower limbs. Further research is needed to confirm the applicability of regional data across different BIA devices and clinical populations, and to assess their relevance to regional functional characteristics.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eBIA\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Bioelectrical impedance analysis\u003c/p\u003e\n\u003cp\u003eBIVA\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Bioelectrical impedance vector analysis\u003c/p\u003e\n\u003cp\u003eBMC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Bone mineral content\u003c/p\u003e\n\u003cp\u003eBMI\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Body mass index\u003c/p\u003e\n\u003cp\u003eCap\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Capacitance\u003c/p\u003e\n\u003cp\u003eCV\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Coefficient of variation\u003c/p\u003e\n\u003cp\u003eDXA\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Dual-energy X-ray absorptiometry\u003c/p\u003e\n\u003cp\u003eFFM\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Fat-free mass\u003c/p\u003e\n\u003cp\u003eFM\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Fat mass\u003c/p\u003e\n\u003cp\u003eLST\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Lean soft tissue\u003c/p\u003e\n\u003cp\u003ePhA\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Phase angle\u003c/p\u003e\n\u003cp\u003eR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Resistance\u003c/p\u003e\n\u003cp\u003eRI\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Resistance index (normalized to stature)\u003c/p\u003e\n\u003cp\u003eRI-specific\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Resistance index (normalized to segment length)\u003c/p\u003e\n\u003cp\u003eXc\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Capacitive reactance\u003c/p\u003e\n\u003cp\u003eXcI\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Capacitive reactance index (normalized to stature)\u003c/p\u003e\n\u003cp\u003eXcI-specific\u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Capacitive reactance (normalized to segment length)\u003c/p\u003e\n\u003cp\u003eXcp\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Capacitive reactance in parallel model\u003c/p\u003e\n\u003cp\u003eXcpI\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Capacitive reactance in parallel model (normalized to stature)\u003c/p\u003e\n\u003cp\u003eXcpI-specific \u0026nbsp; \u0026nbsp;Capacitive reactance in parallel model (normalized to segment length)\u003c/p\u003e\n\u003cp\u003eZ \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Impedance\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe acknowledge Megan Hetherington-Rauth and Tom\u0026aacute;s Carneiro for their contributions to project preparation; Alexandre Moreira, Jerca Kavcic, Katja Ponikvar, Teresa Ourique, and Viviana Staiano for field data collection; Catarina Nunes, Estela Oliveira, Filipe Jesus, Matilde Ourique, and Vanessa Santos for laboratory data collection; Dulce Batista, Hemi Kondo, Ivo Rosa, Jesselin, Juliana Lameiro, Kevin, Magda Pereira, Mariana Lameiro, and Sofia Boavida for data organization; and Maria Isabel Fragoso for scientific advisory. We also express our gratitude to\u0026nbsp;the Oeiras Valley Municipality and \u003cem\u003eAgrupamento de Escolas de Carnaxide\u003c/em\u003e for their commitment and collaboration throughout the project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Portuguese Foundation for Science and Technology [grant to GBR: 2020.07856.BD; grant to IRC:\u0026nbsp;SFRH/BD/149394/2019; grant to AVB: 2025.02275.BD]. This work is also financed by the Portuguese Foundation for Science and Technology within the unit I\u0026amp;D 472 [grant number UIDB/00447/2020].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIf there are no conflicts of interest\u0026nbsp;to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGBR and LBS conceived and designed the research protocol. AVB, GBR, IRC, MP, and RF conducted research and data analysis. AVB, GBR, JPM, and RF wrote the first draft of the manuscript. AMS, GBR, HCL, and LBS wrote the second draft of the manuscript. All authors read, commented, and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eNiu C, Zhang P, Zhang C, Dong J, Liang H, Xiao D, et al. Evolution of research trends and emerging hotspots in bioelectrical impedance analysis over the last two decades: a bibliometric analysis. J Int Soc Sports Nutr. 2025;22(1):2523381.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWard L, Brantlov S. Bioimpedance basics and phase angle fundamentals. Rev Endocr Metab Disord. 2023;24(3):381\u0026ndash;91.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLukaski H. Evolution of bioimpedance: a circuitous journey from estimation of physiological function to assessment of body composition and a return to clinical research. Eur J Clin Nutr. 2013;67(1):S2-S9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun S, Chumlea W, Heymsfield S, Lukaski H, Schoeller D, Friedl K, et al. Development of bioelectrical impedance analysis prediction equations for body composition with the use of a multicomponent model for use in epidemiologic surveys. Am J Clin Nutr. 2003;77(2):331\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbosa-Silva M, Barros A. Bioelectrical impedance analysis in clinical practice: a new perspective on its use beyond body composition equations. Curr Opin Clin Nutr Metab Care. 2005;8(3):311\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBellido D, Garcia-Garcia C, Talluri A, Lukaski H, Garcia-Almeida J. Future lines of research on phase angle: strengths and limitations. Rev Endocr Metab Disord. 2023;24(3):563\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarnett A, Bagno S. The physiological mechanisms involved in the clinical measure of phase angle. Am J Physiol - Legacy Content. 1935;114(2):366\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartins P, de Lima L, Berria J, Petroski E, da Silva A, Silva D. Association between phase angle and isolated and grouped physical fitness indicators in adolescents. Physiol Behav. 2020;217:112825.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Marco J, de Lima T, Tanimoto P, Miranda C, Bim M, Pelegrini A. Association between phase angle and body composition components in adolescent athletes: a cross-sectional study. Physiol Meas. 2025;46(9).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosa G, Francisco R, Silva A, Lukaski H, Bernardino A, Sardinha L. Raw bioelectrical impedance parameters as informative markers of body composition in youth. Clin Nutr. 2025;51:266\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosa GB, Francisco R, Silva AM, Lukaski HC, Sardinha LB. Changes in Raw Bioelectrical Impedance Parameters as Markers of Body Composition in Youth: A Prospective Cohort Investigation. Nutrition. 2026:113089.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLanger R, da Costa K, Bortolotti H, Fernandes G, de Jesus R, Gon\u0026ccedil;alves E. Phase angle is associated with cardiorespiratory fitness and body composition in children aged between 9 and 11 years. Physiology \u0026amp; Behavior. 2020;215:112772.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBallarin G, Licenziati M, Alicante P, Di Vincenzo O, Valerio G, Scalfi L. Bioelectrical impedance analysis-derived phase angle and body composition are predictors of health-related fitness in children and adolescents with obesity. Children (Basel). 2022;9(12):1943.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYanchis D, Chandrakumar A, So S, Patterson C, Belza C, Silva C, et al. Utility of phase angle in children with intestinal failure. Clin Nutr. 2025;52:94\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eda Silva B, Orsso C, Gonzalez M, Sicchieri J, Mialich M, Jordao A, et al. Phase angle and cellular health: inflammation and oxidative damage. Rev Endocr Metab Disord. 2023;24(3):543\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilhelm-Leen E, Hall Y, Horwitz R, Chertow G. Phase angle, frailty and mortality in older adults. J Gen Intern Med. 2014;29(1):147\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCornish B, Jacobs A, Thomas B, Ward L. Optimizing electrode sites for segmental bioimpedance measurements. Physiol Meas. 1999;20(3):241\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDupertuis Y, Jimaja W, Beardsley Levoy C, Genton L. Bioelectrical impedance analysis instruments: how do they differ, what do we need for clinical assessment? Curr Opin Clin Nutr Metab Care. 2025;28(5):379\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCampa F, Coratella G, Cerullo G, Noriega Z, Francisco R, Charrier D, et al. High-standard predictive equations for estimating body composition using bioelectrical impedance analysis: a systematic review. J Transl Med. 2024;22(1):515.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChula de Castro JA, Lima TRd, Silva DAS. Body composition estimation in children and adolescents by bioelectrical impedance analysis: A systematic review. Journal of Bodywork and Movement Therapies. 2018;22(1):134\u0026ndash;46.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePouillet C. Annales de Chimie et de Physique. Paris: Chez Crochard; 1828.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSardinha L, Rosa G, Hetherington-Rauth M, Correia I, Magalhaes J, Silva A, et al. Development and validation of bioelectrical impedance prediction equations estimating regional lean soft tissue mass in middle-aged adults. Eur J Clin Nutr. 2023;77(2):202\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva A, Silva T, Judice P, Rosa G, Bernardino A, Magalhaes J, et al. Regional fat mass estimation using bioelectrical impedance analysis in healthy adults. Br J Nutr. 2025;133(9):1202\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFuller N, Fewtrell M, Dewit O, Elia M, Wells J. Segmental bioelectrical impedance analysis in children aged 8\u0026ndash;12 y: 2. The assessment of regional body composition and muscle mass. International Journal of Obesity. 2002;26(5):692\u0026ndash;700.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrantlov S, Ward LC, Jodal L, Rittig S, Lange A. Critical factors and their impact on bioelectrical impedance analysis in children: a review. J Med Eng Technol. 2017;41(1):22\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNyboer J. Workable volume and flow concepts of bio-segments by electrical impedance plethysmography. TIT J Life Sci. 1972;2(1):1\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrancisco R, Jesus F, Nunes C, Carvalho A, Alvim M, Campa F, et al. Prediction of body water compartments by raw bioelectrical impedance parameters in athletes: Comparison between series and parallel measurements. Scand J Med Sci Sports. 2023;33(10):1998\u0026ndash;2008.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreen SB. How Many Subjects Does It Take To Do A Regression Analysis. Multivariate Behav Res. 1991;26(3):499\u0026ndash;510.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLing CH, de Craen AJ, Slagboom PE, Gunn DA, Stokkel MP, Westendorp RG, et al. Accuracy of direct segmental multi-frequency bioimpedance analysis in the assessment of total body and segmental body composition in middle-aged adult population. Clin Nutr. 2011;30(5):610\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Rui M, Veronese N, Bolzetta F, Berton L, Carraro S, Bano G, et al. Validation of bioelectrical impedance analysis for estimating limb lean mass in free-living Caucasian elderly people. Clin Nutr. 2017;36(2):577\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyle U, Earthman C, Pichard C, Coss-Bu J. Body composition during growth in children: limitations and perspectives of bioelectrical impedance analysis. Eur J Clin Nutr. 2015;69(12):1298\u0026ndash;305.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKriemler S, Puder J, Zahner L, Roth R, Braun-Fahrlander C, Bedogni G. Cross-validation of bioelectrical impedance analysis for the assessment of body composition in a representative sample of 6- to 13-year-old children. Eur J Clin Nutr. 2009;63(5):619\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGutierrez-Marin D, Luque V, Ferre N, Fewtrell MS, Williams JE, Wells JCK. Associations of age and body mass index with hydration and density of fat-free mass from 4 to 22 years. Eur J Clin Nutr. 2019;73(10):1422\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRosa G, Lukaski H, Sardinha L. The science of bioelectrical impedance-derived phase angle: insights from body composition in youth. Rev Endocr Metab Disord. 2025;26(4):603\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGabriel C, Gabriel S, Corthout E. The dielectric properties of biological tissues: I. Literature survey. Phys Med Biol. 1996;41(11):2231\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNovotny SA, Warren GL, Hamrick MW. Aging and the muscle-bone relationship. Physiology (Bethesda). 2015;30(1):8\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolff J. Concerning the interrelationship between form and function of the individual parts of the organism. By Julius Wolff, 1900. Clin Orthop Relat Res. 1988(228):2\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePollock NK, Laing EM, Baile CA, Hamrick MW, Hall DB, Lewis RD. Is adiposity advantageous for bone strength? A peripheral quantitative computed tomography study in late adolescent females. Am J Clin Nutr. 2007;86(5):1530\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBosy-Westphal A, Danielzik S, D\u0026ouml;rh\u0026ouml;fer R, Piccoli A, M\u0026uuml;ller MJ. Patterns of bioelectrical impedance vector distribution by body mass index and age: implications for body-composition analysis. Am J Clin Nutr. 2005;82(1):60\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKryst Ł, Woronkowicz A, Kowal M, Sobiecki J. Intergenerational changes in limb circumferences in children and adolescents aged 3\u0026ndash;18 from Krak\u0026oacute;w (Poland) from 1983 to 2010. American Journal of Human Biology. 2018;30(5):e23165.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Oliveira N, Iniesta R, Vlachopoulos D, Barker A, Matias C, Gon\u0026ccedil;alves E, et al. Can regional phase angle reflect changes in physical fitness and body composition from late childhood to adolescence? A cohort study. Nutrition. 2025:113035.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDittmar M, Reber H. Validation of different bioimpedance analyzers for predicting cell mass against whole-body counting of potassium (40K) as a reference method. Am J Hum Biol. 2004;16(6):697\u0026ndash;703.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAtkins SJ, Bentley I, Hurst HT, Sinclair JK, Hesketh C. The Presence of Bilateral Imbalance of the Lower Limbs in Elite Youth Soccer Players of Different Ages. J Strength Cond Res. 2016;30(4):1007\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoore SA, McKay HA, Macdonald H, Nettlefold L, Baxter-Jones AD, Cameron N, et al. Enhancing a Somatic Maturity Prediction Model. Med Sci Sports Exerc. 2015;47(8):1755\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSherar LB, Mirwald RL, Baxter-Jones AD, Thomis M. Prediction of adult height using maturity-based cumulative height velocity curves. J Pediatr. 2005;147(4):508\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrancisco R, Armstrong LE, Silva AM. Recommendations for Optimizing Research Regarding the Effects of Dehydration on Athletic Performance. Sports Medicine. 2025.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"european-journal-of-clinical-nutrition","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"ejcn","sideBox":"Learn more about [European Journal of Clinical Nutrition](http://www.nature.com/ejcn/)","snPcode":"41430","submissionUrl":"https://mts-ejcn.nature.com/cgi-bin/main.plex","title":"European Journal of Clinical Nutrition","twitterHandle":"@ejcneditor","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Bioimpedance, Dual-energy X-ray absorptiometry, Pediatric, Phase angle, Regional analysis","lastPublishedDoi":"10.21203/rs.3.rs-9224431/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9224431/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e \u003cb\u003eBackground/Objectives\u003c/b\u003e \u003c/p\u003e \u003cp\u003eRegional bioelectrical impedance analysis (BIA) offers a practical yet unexplored approach to characterizing limb-specific physiological properties in youth. Thus, we aimed to examine the cross-sectional and prospective associations between regional raw BIA parameters, including stature- and limb length-normalized indices, and regional body composition in children and adolescents.\u003c/p\u003e \u003cp\u003e \u003cb\u003eSubjects/Methods\u003c/b\u003e \u003c/p\u003e \u003cp\u003eA total of 681 participants underwent regional BIA at 50kHz to obtain phase angle (PhA), resistance index (RI), RI normalized to limb length (RI-specific), reactance index (XcI), XcI normalized to limb length (XcI-specific), parallel XcI (XcpI), XcpI normalized to limb length (XcpI-specific), and capacitance (Cap). Lean soft tissue (LST), fat mass (FM), and bone mineral content (BMC) were measured with dual-energy X-ray absorptiometry. Multiple regression models with false-positive detection were considered.\u003c/p\u003e \u003cp\u003e \u003cb\u003eResults\u003c/b\u003e \u003c/p\u003e \u003cp\u003eRegional BIA variables showed weak-to-strong cross-sectional associations with LST and BMC (\u003cem\u003eβ\u003c/em\u003e = |0.143| to |0.921|). RI-specific consistently exhibited the strongest relationships (\u003cem\u003eβ\u003c/em\u003e = -0.879 to -0.544) outperforming RI and all Xc-derived indices. PhA and Cap were positively associated with LST and BMC but showed lower explanatory power (\u003cem\u003eβ\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.116 to 0.838). For FM, the associations were mostly weak (\u003cem\u003eβ\u003c/em\u003e \u0026lt; |0.583|), with PhA showing the highest associations. In prospective analyses, effect sizes were attenuated overall; however, changes in RI-related indices remained consistently associated with LST and BMC (\u003cem\u003eβ\u003c/em\u003e = -0.593 to -0.136), while no associations were observed for FM.\u003c/p\u003e \u003cp\u003e \u003cb\u003eConclusions\u003c/b\u003e \u003c/p\u003e \u003cp\u003eRegional raw BIA variables enhance the characterization of limb composition in youth, with R-related indices offering the greatest sensitivity to regional LST and BMC.\u003c/p\u003e","manuscriptTitle":"Advancing the Assessment of Regional Body Composition in Children and Adolescents Using Regional Raw Bioelectrical Impedance Data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-04 05:50:21","doi":"10.21203/rs.3.rs-9224431/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-04-20T18:12:34+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-13T12:35:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-10T15:27:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"European Journal of Clinical Nutrition","date":"2026-04-02T14:49:53+00:00","index":"","fulltext":""},{"type":"checksFailed","content":"","date":"2026-04-02T14:36:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"european-journal-of-clinical-nutrition","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"ejcn","sideBox":"Learn more about [European Journal of Clinical Nutrition](http://www.nature.com/ejcn/)","snPcode":"41430","submissionUrl":"https://mts-ejcn.nature.com/cgi-bin/main.plex","title":"European Journal of Clinical Nutrition","twitterHandle":"@ejcneditor","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"fb55bf19-52bf-4eea-a599-ac060c917764","owner":[],"postedDate":"May 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66220881,"name":"Health sciences/Health care/Nutrition"},{"id":66220882,"name":"Health sciences/Health care/Paediatrics"}],"tags":[],"updatedAt":"2026-05-04T05:50:22+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-04 05:50:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9224431","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9224431","identity":"rs-9224431","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}