Assessment of Back Muscle Activity Using High-Density Surface Electromyography in Patients With Degenerative Thoracolumbar Kyphosis: A Preliminary Study
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
To examine back muscle activity and fatigue behavior in female patients with degenerative thoracolumbar kyphosis (DTK) using high-density surface electromyography (HD-SEMG) and evaluate the effects of using a soft spinal orthosis.
Methods
Seven female participants with DTK (mean age: 73.6 years) were assessed during static standing and a weighted holding task with and without a soft spinal orthosis. HD-SEMG signals were obtained from the lumbar erector spinae using a 64-electrode array. Time to fatigue (TTF), spatial displacement, and directional consistency of muscle activation were analyzed using the spatial center of activity (SCoA) and circular variance (CV).
Results
Orthosis use significantly reduced the sagittal vertical axis and low back pain. TTF was significantly prolonged during the weighted holding task with the orthosis (p=0.012), indicating delayed fatigue onset. SCoA displacement was significantly reduced in both tasks (p<0.001), whereas CV analysis demonstrated improved directional consistency of muscle activation.
Conclusion
HD-SEMG revealed early fatigue onset and unstable muscle activation patterns in patients with DTK, particularly during load-bearing tasks performed without orthotic support. Orthosis used improved endurance and neuromuscular efficiency by reducing spatial and directional variability in muscle recruitment. These findings underscore the utility of HD-SEMG for elucidating the neuromuscular pathophysiology of DTK and support the use of spinal orthoses as a conservative treatment approach.
Degenerative thoracolumbar kyphosis (DTK) is a structural spinal deformity resulting from age-related degenerative changes. It is characterized by a reduction or loss of normal lumbar lordosis (LL) or an increase in thoracic or thoracolumbar kyphosis in the sagittal plane [1]. Among adult spinal deformities, age-related DTK is estimated to affect 20%–40% of older adults and has emerged as a significant medico-social concern in aging populations [2]. The postural deformities associated with DTK impose abnormal mechanical stress on spinal support structures, often resulting in chronic pain [3]. Such pain typically occurs during standing, walking, and other activities of daily living and may significantly impair quality of life [4]. A forward-flexed trunk posture has been reported to exacerbate muscle fatigue and pain [5]. Moreover, spinal extensor muscle atrophy, closely associated with DTK progression [6], is considered a major contributor to low back pain (LBP) [7]. These findings suggest that back muscle dysfunction plays a central role in the structural progression and symptomatic burden of DTK.
High-density surface electromyography (HD-SEMG) is an advanced technique for evaluating motor unit activity. By employing two-dimensionally arranged electrode arrays, it enables the acquisition of detailed physiological data across broad muscle regions [8]. In patients with nonspecific LBP, HD-SEMG has been used to reveal the activation of localized lumbar erector spinae during high-effort, fatiguing tasks such as repetitive lifting [9] and sustained isometric contractions [10].
Given the postural and symptomatic similarities, individuals with DTK may exhibit specific back muscle activity patterns comparable to those observed in patients with nonspecific LBP. However, HD-SEMG has not yet been utilized to evaluate back muscle function in this population. Consequently, the neuromuscular characteristics and fatigue-related behavior of back muscles in individuals with DTK remain poorly understood. This study aimed to investigate the back muscle activity in patients with DTK using HD-SEMG to advance the understanding of its pathophysiology and support the development of conservative treatment strategies.
METHODS
Study design
This observational study evaluated the back muscle activity in participants with DTK using HD-SEMG.
Ethical approval
This study was approved by the Institutional Review Board of Kanazawa University (approval number: 114067). Written informed consent was obtained from all participants in accordance with the principles of the Declaration of Helsinki.
Participants
The study participants comprised female patients with DTK who received outpatient care at our hospital. Female patients diagnosed with DTK (sagittal vertical axis [SVA]≥50° or pelvic tilt [PT]≥25°), aged ≥60 years, and capable of comprehending the study procedures were included in the study. All participants received both verbal and written explanations of the study and subsequently provided written informed consent. Patients with scoliosis with a Cobb angle of ≥20°, who previously underwent spinal surgery, and who were unable to independently wear the orthosis due to paralysis or cognitive impairment were excluded. The study was limited to female participants aged ≥60 years, as previous studies indicated a higher prevalence of adult spinal deformity in this demographic [11]. Ultimately, seven participants were enrolled (mean age: 73.6±7.9 years; height: 151.0±5.8 cm; weight: 53.0±8.7 kg; body mass index: 23.0±2.7 kg/m2).
Procedure
Each participant was evaluated under two conditions: with the orthosis (orthosis condition) and without the orthosis (control condition). In the absence of the orthosis, posture was classified as kyphotic. To minimize potential adverse events and ensure adequate familiarity with the device, the participants were instructed to wear the orthosis for one week prior to data collection.
The orthosis was designed to be worn beneath the clothing, like an undergarment, and included an open posterior section to prevent interference with HD-SEMG (Fig. 1). Trunk extension was facilitated by applying posteriorly directed tension through an integrated strap. When the strap was removed, the orthosis became mechanically inactive and served as the control condition (without the orthosis).
HD-SEMG was employed to assess back muscle activity during two standing tasks, performed under orthosis and control conditions. In accordance with a previously published protocol [12], the participants performed (1) a 1-minute static standing task and (2) a 2-minute standing task while holding a 2-kg weight with both hands.
Spinal parameters and evaluation of LBP
Changes in spinal alignment parameters and LBP were evaluated under both orthosis and control conditions. Spinal alignment was assessed using full-spine radiographs, and the parameters were measured by a spine surgeon experienced in sagittal alignment evaluation. The parameters included SVA, LL, thoracic kyphosis, sacral slope, PT, and pelvic incidence (Fig. 2) [13]. LBP intensity was measured using a visual analog scale (VAS).
HD-SEMG recording
An array of 64 multiple electrodes (1-mm diameter, 8-mm interelectrode distance, GR08MM1305; OT Bioelettronica) was used to record HD-SEMG signals from the right erector spinae muscles. Recordings were limited to the right side to minimize the duration of the experimental procedure and patient burden. Electrode placement was determined according to a previously published protocol [14]. The electrode grids were positioned over the erector spinae muscles, identified through palpation, approximately 2 cm lateral to the spinous processes, and expanded from L5 to T12–L1, depending on trunk length. Prior to placement, the skin was cleansed with alcohol to reduce impedance. Electrodes were secured using a bi-adhesive sheet (KIT08MM1305; OT Bioelettronica) following the application of conductive paste (Elefix Z-181BE; Nihon Koden), in accordance with standardized protocols. A reference electrode was placed on the right acromion. Monopolar, multichannel surface electromyographic signals were amplified (gain ×150) and digitized at a sampling rate of 2,048 Hz per channel using the Quattrocento system (OT Bioelettronica). The signals were subsequently processed offline using a band-pass filter (10–500 Hz) and analyzed with MATLAB software (version 2024b; MathWorks).
To evaluate localized muscular fatigue, time to fatigue (TTF) was defined as the earliest time point at which the mean frequency (MF) of the surface electromyography (EMG) signal decreased to less than 90% of the baseline value. Although the median frequency is a commonly used fatigue indicator, MF was selected for this study. The MF represents the spectral centroid, whereas the median frequency serves as the geometric divider of the power spectrum. As aging [15] and muscle atrophy [16] are known to induce multimodal or irregular spectral shapes, MF provides a more reliable estimate of spectral shifts in populations with compromised muscle quality, such as elderly patients with DTK. MF was calculated at 5-second intervals using power spectral density (PSD) estimation with Welch’s method and averaged across all 64 recording channels to provide a spatially global representation of muscle activity. The PSD was computed to quantify the distribution of EMG power across frequencies. PSD analysis is widely used to assess muscle fatigue due to the characteristic shift of the EMG frequency content toward lower frequencies during sustained contraction. Welch’s method was employed as it is one of the most widely adopted techniques for physiological signal analysis, offering reduced variance through segmention, windowing, and averaging of periodograms [17].
To minimize the effect of transient fluctuations during initial postural stabilization, the baseline MF was defined as the value obtained at the fourth time point (i.e., 15 seconds after task onset). TTF for each condition was subsequently determined as the earliest time point at which MF declined to less than 90% of the established baseline. A 10% reduction threshold was adopted based on previous studies demonstrating that such a relative decrease reflects the physiologically meaningful onset of neuromuscular fatigue during sustained submaximal isometric activity (Fig. 3). This threshold has been widely adopted in EMG-based fatigue analyses, as it provides an optimal balance between sensitivity to fatigue-related changes and robustness against minor signal variability [18,19]. Furthermore, it serves as a standardized criterion that facilitates comparisons across participants and experimental conditions.
To compare TTF between conditions, Kaplan–Meier survival analysis with log-rank testing was employed to assess differences in the probability of fatigue onset between the orthosis and control conditions. Fatigue incidence and latency were also descriptively examined using cumulative survival curves. HD-SEMG was further employed to evaluate the spatial and temporal dynamics of trunk muscle activation during prolonged static standing. Signals were recorded with a 13×5 electrode grid positioned over the lumbar region. Raw EMG data were segmented into 10-second windows, and the PSD was computed for each channel using Welch’s method. Spectral power within the 20–150 Hz frequency band was summed to represent localized muscle activity. Spatial activation maps were subsequently generated from the power distribution. The spatial center of activity (SCoA) was calculated as the two-dimensional weighted centroid of the map:
xc=∑i,jxi,j·Ai,j∑i,jAi,j
yc=∑i,jyi,j·Ai,j∑i,jAi,j
where Ai,j represents the spectral power at electrode (i, j), and xi,j, and yi,j denote the spatial coordinates on the electrode grid. The temporal trajectory of the SCoA was tracked throughout the 60-second trial. Displacement vectors were defined as the differences in the SCoA positions between the initial (0–10 seconds) and last (50–60 seconds) time windows.
To assess the directional consistency of muscle activation, the circular variance (CV) of the angular displacement of the SCoA between the first and last 10-second windows was calculated. Displacement angles were derived from the directional change in the SCoA vector on the two-dimensional grid. CV for each condition was calculated using the following formula:
R=|1n∑k=1neiθk|
CV=1-R
where θk denotes the displacement angle for participant k, and n represents the total number of participants. Values closer to 0 indicate more concentrated (i.e., consistent) directional shifts, whereas values closer to 1 reflect greater directional variability.
Comparisons between orthosis conditions were performed using non-parametric tests due to the bounded nature of CV (range: 0–1).
Statistical analysis
Continuous variables are expressed as means±standard deviations or medians (minimum, maximum), as appropriate. Given the small sample size (n=7), non-parametric tests were employed for all comparisons to ensure statistical robustness without assuming a normal distribution. The Wilcoxon signed-rank test was conducted to compare spinal alignment parameters, LBP scores, and HD-SEMG variables (TTF and SCoA metrics) before and after orthosis use.
TTF was analyzed using the Wilcoxon signed-rank test as the paired design violated the independence assumption required for the log-rank test. Participants who did not exhibit a 10% decrease in the mean frequency were treated as right-censored at 60 or 120 seconds, corresponding to the end of the holding period. Kaplan–Meier curves were generated solely for visualization purposes only and were not used for statistical inference.
A linear mixed-effects model was employed to evaluate the effects of orthosis condition (with vs. without) and task type (static standing vs. weighted holding) on SCoA displacement magnitude. Fixed effects included condition, task, and their interaction (condition×task), whereas the participant identifier was included as a random intercept to account for repeated measures. Post-hoc pairwise comparisons were conducted using estimated marginal means with Tukey-adjusted p-values.
SCoA displacement magnitude (calculated as the Euclidean distance) and angular direction were compared between the orthosis conditions. Directional consistency was evaluated using CV (1-R).
To examine the association between fatigue and spatial redistribution of muscle activity, the relationship between SCoA displacement and TTF was evaluated using repeated-measures correlation, which accounts for within-participant dependency. Correlations were computed separately for the with and without orthosis conditions. Individual regression lines derived from the repeated-measures correlation model were plotted to visualize within-participant trends.
Statistical significance was defined as a p-value of <0.05. All statistical analyses were performed using R (version 4.5.1, R Foundation for Statistical Computing) with the lme4, lmerTest, and rmcorr packages.
RESULTS
Table 1 presents the changes in spinal alignment parameters and LBP under the orthosis and control conditions. Use of the orthosis resulted in significant improvements in both SVA (p<0.01) and LBP, measured using the VAS (p<0.01). No significant differences were observed in other spinal alignment parameters.
Kaplan–Meier survival analysis demonstrated that TTF (defined as the time at which the MF of the surface EMG signal declined below 90% of the baseline) was influenced by both task type and orthosis condition. During the static standing task, the mean TTF durations were 58.6±2.1 seconds with the orthosis and 52.1±8.3 seconds without it. Although the participants wearing the orthosis tended to maintain muscle activation for a longer duration, this difference was not significant (Wilcoxon signed-rank test: p=0.125) (Fig. 4A, 4C). During the weighted holding task, the orthosis condition yielded a significantly longer TTF (57.9±3.8 seconds) compared with the control condition (45.0±8.6 seconds) (p=0.031; Fig. 4B, 4D), indicating that the orthosis effectively delayed the onset of localized muscle fatigue under an external load. Notably, five of seven participants reached the fatigue threshold in the weighted task without orthosis, whereas only two participants did so while wearing the orthosis. These findings support the effectiveness of a lumbar orthosis in improving fatigue resistance, particularly during sustained postural loading.
The LME model revealed the significant main effects of task (p<0.001) and condition (p<0.001), as well as a significant interaction effect (p<0.001), on SCoA displacement magnitude. Displacement was greater during the weighted task, particularly in the absence of the orthosis. Post hoc comparisons using Tukey’s Honestly Significant Difference test indicated significantly lower SCoA displacement in the orthosis condition for both the weighted holding (mean difference=0.806, p<0.001) (Fig. 5) and static standing tasks (mean difference=0.797, p<0.001). All pairwise comparisons among the four groups yielded significant differences (p<0.001). Collectively, these findings indicate that the orthosis significantly reduces lumbar displacement, particularly under increased loading conditions.
Although larger displacement magnitudes were observed in the absence of orthosis, the direction of SCoA displacement remained relatively consistent across participants. During the static standing task, angular analysis demonstrated similarly high directional consistency in both conditions (Fig. 6A). The mean vector lengths (R), reflecting the concentration of directional shifts, were 0.98 with the orthosis and 0.97 without it, indicating comparable spatial consistency. Likewise, CV values were low and similar between conditions (with orthosis: 0.02; without orthosis: 0.03), suggesting that orthosis use exerted minimal impact on the directionality of trunk muscle activation under static postural tasks.
In the weighted holding condition, directional consistency was notably greater with the orthosis (R=0.82) compared with that without the orthosis (R=0.37), with corresponding CV values of 0.19 and 0.85, respectively. These findings suggest that muscle activation trajectories were not only more consistent but also more posteriorly directed when the orthosis was worn. This interpretation is further supported by the polar histogram (Fig. 6B), which shows that most vectors in the orthosis condition clustered around ~180°, corresponding to the posterior direction in the polar coordinate system (0°=head, 180°=feet). By contrast, vectors in the without orthosis condition were more widely distributed across angular directions, indicating reduced directional stability.
To examine the association between fatigue duration and the spatial redistribution of muscle activation, repeated-measures correlations were calculated between the TTF and SCoA. Given that the orthosis condition substantially shifted the group means of both variables, correlations were also computed separately for each condition. Strong within-subject correlations were observed under both conditions (with orthosis: r=0.97, p<0.001; without orthosis: r=0.78, p=0.02) (Fig. 7), supporting the physiological linkage between TTF and spatial redistribution of muscle activity.
DISCUSSION
Participants with DTK exhibited significant improvements in both SVA and LBP, assessed using the VAS, when wearing the orthosis. The orthosis used in this study was a soft type, offering less structural rigidity compared with conventional rigid braces [20,21]. Consistent with the findings of previous studies, which indicated that soft orthoses exert limited corrective effects compared with rigid braces [22], the improvements in spinal alignment were modest. Nevertheless, the orthosis contributed to sagittal plane realignment, evidenced by a reduction in SVA, and this postural correction was accompanied by reduced LBP. These results suggest that even soft orthoses may provide symptomatic relief by providing minimal yet beneficial postural support. Overall, the comparison of postural and electromyographic outcomes with and without orthosis use provides novel insights into paraspinal muscle activity in individuals with DTK.
This study evaluated trunk muscle fatigue and postural adaptation in patients with DTK using HD-SEMG during two standing tasks performed with and without orthotic support. This method enabled the simultaneous assessment of TTF, spatial displacement, and directional control of muscle activation, quantified by the SCoA. Unlike conventional EMG, which provides limited spatial resolution, HD-SEMG facilitated a more comprehensive spatiotemporal analysis of paraspinal recruitment across thoracolumbar levels.
The results demonstrated that participants with DTK exhibited earlier fatigue onset and greater spatial shifts in muscle activation, particularly during loaded standing without orthotic support. TTF was significantly prolonged with orthosis use, suggesting that orthotic support reduced neuromuscular load and delayed fatigue onset. These findings are consistent with the reports of previous studies describing increased fatigue and compensatory trunk strategies in individuals with sagittal spinal deformities [23,24]. Beyond enhancing endurance, orthotic support improved the spatial and directional regulation of trunk muscle activity. Specifically, the displacement magnitude of the SCoA was significantly reduced with orthosis use across both standing tasks, indicating more centralized and stable motor control. Physiologically, shifts in the SCoA during sustained contraction reflect the “spatial redistribution” of muscle activity, as active motor units are recruited to maintain force output [9,10]. This redistribution serves as a protective mechanism to prevent local overloading of fatigued muscles. This stabilizing effect was particularly pronounced during loaded standing. In the absence of orthotic support, participants demonstrated larger and more erratic SCoA displacements, suggestive of maladaptive compensatory mechanisms. Angular analysis further substantiated these findings. During static standing, displacement vectors were relatively consistent and posteriorly directed under both conditions. However, during the weighted task, orthosis use maintained a posterior-focused recruitment strategy, whereas the no-orthosis condition resulted in highly variable trajectories. CV analysis confirmed decreased directional stability without orthotic support, suggesting impaired neuromuscular coordination or uncontrolled activation of adjacent muscle regions. Collectively, these findings extend previous observations [23] by demonstrating that not only muscle weakness but also impaired motor consistency and adaptability contribute to postural inefficiency in kyphotic deformities. The observed reductions in spatial and directional stability may reflect diminished motor variability—a characteristic commonly reported in patients with chronic musculoskeletal conditions [24,25]. Such rigidity in muscle recruitment patterns may hinder compensatory capacity and accelerate postural decline.
Clinically, the present findings support the role of spinal orthoses in enhancing neuromuscular efficiency in individuals with DTK. By reducing paraspinal musculature and improving trunk muscle coordination, orthoses may help preserve functional capacity and mitigate fatigue-related postural collapse. Conservative treatment, particularly the combination of exercise therapy and bracing, has demonstrated effectiveness in patients with DTK [26]. Exercise programs focused on strengthening, stretching, and postural alignment have been shown to improve spinal alignment, muscle strength, and overall quality of life [27-29]. Similarly, orthosis use has been associated with improvements in sagittal alignment and trunk extensor strength [30]. When integrated with targeted therapeutic exercise, orthoses may contribute to restoring adaptive motor strategies, thereby enhancing clinical outcomes in kyphosis management.
This study has some limitations. First, the sample size was small and derived from a single clinical population, which limited the statistical power of the analysis and increased the risk of type II errors. Consequently, the findings should be considered preliminary and exploratory. Although the results suggest significant trends in muscle fatigue and spatial activation, caution is warranted, as these findings may not be fully generalizable without validation in larger, more diverse cohorts. Future studies with adequately powered sample sizes are warranted to confirm these observed effects. Second, the absence of a healthy control group precluded direct comparisons between patients with spinal kyphosis and age-matched normative patterns. Future studies including healthy older adults would facilitate the differentiation between pathological and age-related neuromuscular adaptations. Third, HD-SEMG recordings were restricted to the right paraspinal region due to practical constraints. Although previous studies have reported minimal bilateral asymmetry during static postures [14], lateral imbalances may occur in individuals with postural deformities. Thus, bilateral recordings would provide a more comprehensive assessment of spatial recruitment patterns. Fourth, the standing tasks employed were of short duration. Although sufficient to induce fatigue without orthosis, longer or dynamic tasks (e.g., walking or sit-to-stand) may better capture functional neuromuscular adaptations over time. Finally, although TTF and SCoA provide meaningful surface-level indicators of muscle behavior, they remain indirect measures. Incorporating imaging modalities (e.g., magnetic resonance imaging and ultrasound) or biomechanical modeling in future studies could further validate the observed neuromuscular patterns and identify underlying structural correlates. Despite these limitations, this study is the first to integrate HD-SEMG with survival analysis and spatial trajectory tracking to evaluate fatigue and postural control in patients with spinal kyphosis. The findings provide novel evidence that may inform and refine conservative treatment strategies, including orthotic interventions and exercise-based rehabilitation for DTK.
This study employed HD-SEMG to evaluate trunk muscle fatigue and spatial activation patterns in patients with DTK during static and loaded standing tasks. Orthosis use significantly delayed the onset of fatigue and reduced both the magnitude and variability of spatial muscle displacement. Moreover, it was associated with more consistent directional shifts in the SCoA, suggesting improved neuromuscular coordination and postural stability. These results underscore the utility of HD-SEMG in detecting subtle spatiotemporal changes in muscle activity and highlight the clinical relevance of orthotic support in mitigating neuromuscular fatigue in patients with DTK. By enhancing endurance and stabilizing trunk muscle recruitment, lumbar orthoses may improve posture alignment and functional performance during upright tasks. Collectively, these results provide a physiological basis for incorporating orthotic interventions into the conservative management of DTK.
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING INFORMATION
This work was supported in part by a research grant from the Japanese Society for Musculoskeletal Medicine. The funding organization had no involvement in the study design; data collection, analysis, and interpretation; or manuscript writing.
AUTHOR CONTRIBUTION
Conceptualization: Kato S. Investigation: Kurokawa Y, Nishikawa Y. Methodology: Kurokawa Y, Nishikawa Y, Kato S. Data curation: Kurokawa Y, Nishikawa Y. Formal analysis: Nishikawa Y. Resources: Yokogawa N, Shimizu T, Demura S, Kato S. Project administration: Kato S. Funding acquisition: Kato S. Supervision: Demura S, Kato S. Visualization: Nishikawa Y. Validation: Nishikawa Y. Writing – original draft: Kurokawa Y. Writing – review and editing: Nishikawa Y. Approval of final manuscript: all authors.
ACKNOWLEDGMENTS
The authors would like to thank NIPPON SIGMAX Co., Ltd. (Tokyo, Japan) for providing the orthosis used for this study. The company had no role in the study design, data collection, data analysis, interpretation, or in the writing of this manuscript.
DATA AVAILABILITY
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.
Fig. 1.
Design of the trunk brace used in the study. The brace was worn beneath clothing and featured a posterior opening to accommodate high-density surface electromyography recordings. Posteriorly directed tension was applied to facilitate trunk extension, and the brace became mechanically inactive upon strap removal.
Fig. 2.
Spinal parameters used in this study. The sagittal vertical axis is defined as the horizontal offset from the posterosuperior corner of S1 to the vertebral midbody of C7. Lumbar lordosis was measured from the superior endplate of L1 to the superior endplate of S1. Thoracic kyphosis was measured from the superior endplate of T4 to the inferior endplate of T12. Lumbar lordosis was measured from the superior endplate of L1 to the superior endplate of S1. Pelvic tilt was defined as the angle between the vertical and the line connecting the midpoint of the upper acral endplate to the femoral head axis. The sacral slope is defined as the angle between the horizontal and upper sacral endplate. Pelvic incidence is defined as the angle between the perpendicular to the upper sacral endplate at its midpoint and the line connecting this point to the femoral head axis. This figure was originally created by the authors based on the data and illustrations in Schwab et al. [13]. SVA, sagittal vertical axis; LL, lumbar lordosis; TK, thoracic kyphosis; SS, sacral slope; PT, pelvic tilt; PI, pelvic incidence.
Fig. 3.
Time-series changes in the mean frequency of surface electromyography signals during static standing identification 1. Blue and red lines indicate represent the orthosis and no-orthosis conditions, respectively. Dashed horizontal lines denote 10% drop thresholds, defined as a 10% reduction from the mean frequency at 15 seconds. In this participant, baseline values were nearly identical between conditions, resulting in overlapping thresholds. Dotted vertical lines indicate the time to fatigue, identified as the first instance the signal dropped below the threshold. Mean frequency was calculated every 5 seconds using power spectral analysis across all channels.
Fig. 4.
Time to fatigue (TTF) with and without orthosis. (A, B) Kaplan–Meier curves illustrating TTF during (A) static standing and (B) weighted holding. Participants who did not reach the fatigue criterion (10% reduction in mean frequency) were treated as right-censored at the task limit (60 or 120 seconds). (C, D) Individual paired TTF values for (C) static condition and (D) weighted holding condition. Statistical comparisons were performed using the Wilcoxon signed-rank test. *p<0.05.
Fig. 5.
Comparison of spatial center of activity (SCoA) displacement magnitudes across postural tasks and orthosis conditions. Individual data points (n=7) are presented for static standing (blue) and weighted holding (purple) tasks. Displacement magnitude, expressed in arbitrary units, was calculated as the Euclidean distance between the SCoA in the initial (0–10 seconds) and final (50–60 seconds) 10-second windows. Orthosis use significantly reduced SCoA displacement in both tasks (p<0.001, Tukey’s HSD test following linear mixed-effects analysis), indicating enhanced postural stability. Horizontal bars indicate group means; asterisks (*) denote significant differences.
Fig. 6.
Directional distribution of spatial center of activity (SCoA) shifts during static standing and weighted holding tasks. Polar histograms illustrate the angular displacement of the SCoA between the initial and final 10-second intervals, reflecting shifts in muscle activation direction. In static standing (A), both conditions exhibited consistent posterior shifts (i.e., downward on the polar plot). By contrast, the weighted holding task (B) demonstrated tightly clustered posterior shifts with orthosis use, and more dispersed, variable shifts without orthosis use, indicating reduced directional stability. Each bin represents 30°; angular orientation is displayed clockwise from the top (0°=head, 180°=feet). Directional shifts reflect the electrode grid placement on the right lower back (2 cm lateral to the spinous processes from T12 to L5), where posterior shifts appear downward and lateral deviations toward the right side of the plot.
Fig. 7.
(A) Repeated-measures correlations between spatial center of activity (SCoA) displacement and time to fatigue under orthosis and (B) no-orthosis conditions. Each point represents an observation from a single participant across the two task conditions, whereas colored lines denote participant-specific regression fits estimated by the repeated-measures correlation model.
Table 1.
Comparison of spinal parameters and low back pain with and without orthosis use
2. Kado DM, Huang MH, Karlamangla AS, Barrett-Connor E, Greendale GA. Hyperkyphotic posture predicts mortality in older community-dwelling men and women: a prospective study. J Am Geriatr Soc 2004;52:1662-7.
3. Burger H, Van Daele PL, Grashuis K, Hofman A, Grobbee DE, Schütte HE, et al. Vertebral deformities and functional impairment in men and women. J Bone Miner Res 1997;12:152-7.
4. Miyakoshi N, Itoi E, Kobayashi M, Kodama H. Impact of postural deformities and spinal mobility on quality of life in postmenopausal osteoporosis. Osteoporos Int 2003;14:1007-12.
5. Takahashi I, Kikuchi S, Sato K, Iwabuchi M. Effects of the mechanical load on forward bending motion of the trunk: comparison between patients with motion-induced intermittent low back pain and healthy subjects. Spine (Phila Pa 1976) 2007;32:E73-8.
6. Lee JC, Cha JG, Kim Y, Kim YI, Shin BJ. Quantitative analysis of back muscle degeneration in the patients with the degenerative lumbar flat back using a digital image analysis: comparison with the normal controls. Spine (Phila Pa 1976) 2008;33:318-25.
7. Hides J, Gilmore C, Stanton W, Bohlscheid E. Multifidus size and symmetry among chronic LBP and healthy asymptomatic subjects. Man Ther 2008;13:43-9.
9. Falla D, Gizzi L, Tschapek M, Erlenwein J, Petzke F. Reduced task-induced variations in the distribution of activity across back muscle regions in individuals with low back pain. Pain 2014;155:944-53.
10. Sanderson A, Martinez-Valdes E, Heneghan NR, Murillo C, Rushton A, Falla D. Variation in the spatial distribution of erector spinae activity during a lumbar endurance task in people with low back pain. J Anat 2019;234:532-42.
11. McAviney J, Roberts C, Sullivan B, Alevras AJ, Graham PL, Brown BT. The prevalence of adult de novo scoliosis: a systematic review and meta-analysis. Eur Spine J 2020;29:2960-9.
13. Schwab F, Patel A, Ungar B, Farcy JP, Lafage V. Adult spinal deformity-postoperative standing imbalance: how much can you tolerate? An overview of key parameters in assessing alignment and planning corrective surgery. Spine (Phila Pa 1976) 2010;35:2224-31.
14. Serafino F, Trucco M, Occhionero A, Cerone GL, Chiarotto A, Vieira T, et al. Understanding regional activation of thoraco-lumbar muscles in chronic low back pain and its relationship to clinically relevant domains. BMC Musculoskelet Disord 2021;22:432.
15. Habenicht R, Ebenbichler G, Bonato P, Kollmitzer J, Ziegelbecker S, Unterlerchner L, et al. Age-specific differences in the time-frequency representation of surface electromyographic data recorded during a submaximal cyclic back extension exercise: a promising biomarker to detect early signs of sarcopenia. J Neuroeng Rehabil 2020;17:8.
16. Kupa EJ, Roy SH, Kandarian SC, De Luca CJ. Effects of muscle fiber type and size on EMG median frequency and conduction velocity. J Appl Physiol (1985) 1995;79:23-32.
17. Welch P. The use of fast Fourier transform for the estimation of power spectra: A method based on time averaging over short, modified periodograms. IEEE Trans Audio Electroacoust 1967;15:70-3.
20. Meccariello L, Muzii VF, Falzarano G, Medici A, Carta S, Fortina M, et al. Dynamic corset versus three-point brace in the treatment of osteoporotic compression fractures of the thoracic and lumbar spine: a prospective, comparative study. Aging Clin Exp Res 2017;29:443-9.
21. Shariatzadeh H, Modaghegh BS, Mirzaei A. The effect of dynamic hyperextension brace on osteoporosis and hyperkyphosis reduction in postmenopausal osteoporotic women. Arch Bone Jt Surg 2017;5:181-5.
22. Gutman G, Benoit M, Joncas J, Beauséjour M, Barchi S, Labelle H, et al. The effectiveness of the SpineCor brace for the conservative treatment of adolescent idiopathic scoliosis. Comparison with the Boston brace. Spine J 2016;16:626-31.
23. Greig AM, Bennell KL, Briggs AM, Wark JD, Hodges PW. Balance impairment is related to vertebral fracture rather than thoracic kyphosis in individuals with osteoporosis. Osteoporos Int 2007;18:543-51.
24. Żurawski AŁ, Kiebzak WP, Kowalski IM, Śliwiński G, Śliwiński Z. Evaluation of the association between postural control and sagittal curvature of the spine. PLoS One 2020;15:e0241228.
25. Lund JP, Donga R, Widmer CG, Stohler CS. The pain-adaptation model: a discussion of the relationship between chronic musculoskeletal pain and motor activity. Can J Physiol Pharmacol 1991;69:683-94.
28. Katzman WB, Sellmeyer DE, Stewart AL, Wanek L, Hamel KA. Changes in flexed posture, musculoskeletal impairments, and physical performance after group exercise in community-dwelling older women. Arch Phys Med Rehabil 2007;88:192-9.
29. Pawlowsky SB, Hamel KA, Katzman WB. Stability of kyphosis, strength, and physical performance gains 1 year after a group exercise program in community-dwelling hyperkyphotic older women. Arch Phys Med Rehabil 2009;90:358-61.
30. Pfeifer M, Begerow B, Minne HW. Effects of a new spinal orthosis on posture, trunk strength, and quality of life in women with postmenopausal osteoporosis: a randomized trial. Am J Phys Med Rehabil 2004;83:177-86.
Assessment of Back Muscle Activity Using High-Density Surface Electromyography in Patients With Degenerative Thoracolumbar Kyphosis: A Preliminary Study
Fig. 1. Design of the trunk brace used in the study. The brace was worn beneath clothing and featured a posterior opening to accommodate high-density surface electromyography recordings. Posteriorly directed tension was applied to facilitate trunk extension, and the brace became mechanically inactive upon strap removal.
Fig. 2. Spinal parameters used in this study. The sagittal vertical axis is defined as the horizontal offset from the posterosuperior corner of S1 to the vertebral midbody of C7. Lumbar lordosis was measured from the superior endplate of L1 to the superior endplate of S1. Thoracic kyphosis was measured from the superior endplate of T4 to the inferior endplate of T12. Lumbar lordosis was measured from the superior endplate of L1 to the superior endplate of S1. Pelvic tilt was defined as the angle between the vertical and the line connecting the midpoint of the upper acral endplate to the femoral head axis. The sacral slope is defined as the angle between the horizontal and upper sacral endplate. Pelvic incidence is defined as the angle between the perpendicular to the upper sacral endplate at its midpoint and the line connecting this point to the femoral head axis. This figure was originally created by the authors based on the data and illustrations in Schwab et al. [13]. SVA, sagittal vertical axis; LL, lumbar lordosis; TK, thoracic kyphosis; SS, sacral slope; PT, pelvic tilt; PI, pelvic incidence.
Fig. 3. Time-series changes in the mean frequency of surface electromyography signals during static standing identification 1. Blue and red lines indicate represent the orthosis and no-orthosis conditions, respectively. Dashed horizontal lines denote 10% drop thresholds, defined as a 10% reduction from the mean frequency at 15 seconds. In this participant, baseline values were nearly identical between conditions, resulting in overlapping thresholds. Dotted vertical lines indicate the time to fatigue, identified as the first instance the signal dropped below the threshold. Mean frequency was calculated every 5 seconds using power spectral analysis across all channels.
Fig. 4. Time to fatigue (TTF) with and without orthosis. (A, B) Kaplan–Meier curves illustrating TTF during (A) static standing and (B) weighted holding. Participants who did not reach the fatigue criterion (10% reduction in mean frequency) were treated as right-censored at the task limit (60 or 120 seconds). (C, D) Individual paired TTF values for (C) static condition and (D) weighted holding condition. Statistical comparisons were performed using the Wilcoxon signed-rank test. *p<0.05.
Fig. 5. Comparison of spatial center of activity (SCoA) displacement magnitudes across postural tasks and orthosis conditions. Individual data points (n=7) are presented for static standing (blue) and weighted holding (purple) tasks. Displacement magnitude, expressed in arbitrary units, was calculated as the Euclidean distance between the SCoA in the initial (0–10 seconds) and final (50–60 seconds) 10-second windows. Orthosis use significantly reduced SCoA displacement in both tasks (p<0.001, Tukey’s HSD test following linear mixed-effects analysis), indicating enhanced postural stability. Horizontal bars indicate group means; asterisks (*) denote significant differences.
Fig. 6. Directional distribution of spatial center of activity (SCoA) shifts during static standing and weighted holding tasks. Polar histograms illustrate the angular displacement of the SCoA between the initial and final 10-second intervals, reflecting shifts in muscle activation direction. In static standing (A), both conditions exhibited consistent posterior shifts (i.e., downward on the polar plot). By contrast, the weighted holding task (B) demonstrated tightly clustered posterior shifts with orthosis use, and more dispersed, variable shifts without orthosis use, indicating reduced directional stability. Each bin represents 30°; angular orientation is displayed clockwise from the top (0°=head, 180°=feet). Directional shifts reflect the electrode grid placement on the right lower back (2 cm lateral to the spinous processes from T12 to L5), where posterior shifts appear downward and lateral deviations toward the right side of the plot.
Fig. 7. (A) Repeated-measures correlations between spatial center of activity (SCoA) displacement and time to fatigue under orthosis and (B) no-orthosis conditions. Each point represents an observation from a single participant across the two task conditions, whereas colored lines denote participant-specific regression fits estimated by the repeated-measures correlation model.
Graphical abstract
Fig. 1.
Fig. 2.
Fig. 3.
Fig. 4.
Fig. 5.
Fig. 6.
Fig. 7.
Graphical abstract
Assessment of Back Muscle Activity Using High-Density Surface Electromyography in Patients With Degenerative Thoracolumbar Kyphosis: A Preliminary Study
Without-orthosis
With-orthosis
p-value
r
SVA
110.4±71.1
92.3±69.3
<0.01**
0.86
LL
-8.3±17.6
-6.4±17.7
0.13
0.68
TK
23.9±12.5
24.0±16.0
0.91
0.10
SS
0.2±10.0
-1.1±10.1
0.17
0.60
PT
39.2±7.3
40.3±7.6
0.16
0.57
PI
39.7±7.5
39.8±8.2
0.98
0.03
LBP (VAS)
75.9±6.1
45.3±20.8
<0.01**
0.82
Table 1. Comparison of spinal parameters and low back pain with and without orthosis use
Spinal parameters (SVA, LL, TK, SS, PT, and PI) are expressed in degrees (°).
Pearson’s correlation coefficient; r indicates post-hoc effect size.
, Yuichi Nishikawa, PT, PhD3
