IntroductionSpinal orthoses have undergone significant evolution over decades, progressing from rigid immobilization devices to sophisticated soft exoskeletons designed to enhance functionality and patient outcomes [1]. Early iterations primarily used rigid structures to apply corrective or limiting forces, proving effective for postsurgical stabilization, injury protection, and the management of spinal deformities such as scoliosis and kyphosis [2]. However, the long-term adverse effects of rigid immobilization—including muscle atrophy and decreased bone mineral density—highlighted the need for alternative approaches [3]. This led to the development, by the late 1990s, of the first biomechanically informed orthoses incorporating semirigid and soft structures for kyphosis management [4].
Today, semirigid thoracolumbar orthoses are a standard conservative intervention for reducing thoracic kyphosis and improving muscle strength in postural and age-related hyperkyphosis [1]. While clinically effective, evidence suggests that they may not represent the most efficient conservative method for either kyphosis reduction or strengthening back musculature [5]. Consequently, orthotic design has continued to advance. A key development was the 2004 introduction of the Spinomed orthosis, which demonstrated improved outcomes [6]. Throughout this progression, the integration of leaf springs has been a notable feature, associated with significant clinical effectiveness [7,8]. Nonetheless, research has largely focused on therapeutic end points rather than on optimizing the material properties of the springs themselves [1].
Our previous work investigated leaf springs fabricated from AISI (American Iron and Steel Institute) 1075 carbon steel, a common and widely available material [8]. Although effective, the intervention produced a mean reduction in thoracic kyphosis of 8.14° (SD 10.5°), which was less than the 11.76° (10.4°) reduction achieved with the Spinomed orthosis [6]. Both orthotic interventions were, in turn, outperformed by supervised exercise training, as identified in a meta-analysis [5]. This performance gap underscores the need for further innovation in orthotic design.
To enhance the efficacy of semirigid thoracolumbar orthoses, we propose integrating a programmable local vibration system. Local vibration is a recognized intervention for mitigating muscle soreness and fatigue [9] and can promote muscle hypertrophy [10]. Although prior studies have explored vibration as a postural reminder in young populations with hyperkyphosis [11], its application for muscle strengthening in age-related hyperkyphosis—a condition strongly correlated with decreased back extensor strength [12]—remains underexplored. Commonly used displacement probes in muscular rehabilitation lack consistent evidential support [13], and concerns persist regarding potential muscle weakness from improper use. A programmable system could address these safety concerns, ensuring controlled application to protect muscle and skin integrity.
Spinal orthoses have evolved significantly from rigid, immobilizing braces toward semirigid and soft exoskeletons that aim to support function while mitigating the muscle atrophy and bone density loss associated with rigid designs [1,14,15]. While effective for stabilization, rigid structures often lead to discomfort and restricted mobility. Semirigid thoracolumbar orthoses now represent a standard conservative intervention for age-related hyperkyphosis, aiming to reduce the thoracic kyphosis angle (TKA) and improve back extensor strength [1]. Key innovations, such as the Spinomed orthosis and the integration of biomechanically tuned leaf springs, have enhanced corrective outcomes and trunk support [7,16]. However, despite these advancements, orthotic interventions generally remain less effective than supervised exercise programs, which meta-analyses have identified as producing superior kyphosis reduction through targeted strengthening [1,15-17].
Current orthotic research emphasizes clinical outcomes over material optimization [18], leaving potential for enhanced spring designs to close the efficacy gap with exercises [5]. Leaf spring materials for orthoses have been studied, with focus on optimizing flexibility, strength, and weight [7,8,16]. Common materials used in leaf springs include AISI 1075 carbon steel and composite polymers such as carbon fiber–reinforced plastics. Carbon fiber composites exhibit superior stiffness-to-weight ratios and fatigue resistance compared to traditional steel springs, leading to lighter orthoses with improved energy return and durability. Studies comparing steel and composite leaf springs report that composites exhibit up to ~30% higher stiffness and significantly lower stress under load, which may be beneficial for sustained corrective forces while reducing the risk of material fatigue. Thermoplastics with adjustable molding properties also allow customizable fit without compromising spring performance [19,20]. These material advances enable tunable mechanical properties in leaf springs for tailored spinal support devices [14].
Comparing the Spinomed orthosis versus exercise outcomes in hyperkyphosis, several randomized studies and meta-analyses find both interventions to be effective in reducing kyphosis angle and improving back extensor strength [21]. The Spinomed orthosis improves spinal alignment and muscle endurance by providing postural support and facilitating muscle activation, with kyphosis angle reductions reported between 7° and 12° after months of use [16,22]. However, recent randomized controlled trials of semirigid thoracolumbar orthoses in older adults with hyperkyphosis report improvements in TKA, trunk muscle strength, and balance after orthosis use for 6 to 12 weeks. One trial showed significant kyphosis reduction and increased isometric trunk extensor and flexor strength using a semirigid backpack-style orthosis compared to no treatment. Another randomized controlled trial demonstrated that Spinomed orthosis improved back muscle maximal voluntary contraction and balance parameters, highlighting muscle facilitation effects. However, trials underscore that orthoses often provide modest benefit relative to targeted exercise regimens, and none included programmable local vibration systems integrated with springs to date [8,16,23]. Meta-analyses highlight no significant difference between Spinomed orthosis and posture training support for immediate improvements, but supervised exercise tends to surpass orthoses in long-term outcomes [17]. Exercise programs, particularly those focusing on back extensor strengthening and postural training, often achieve greater kyphosis reductions and functional gains than orthoses alone, although adherence and safety vary [5,21].
Biomechanical analyses of leaf spring placement and force transmission in spinal supports emphasize critical design parameters, including spring position relative to spinal curvature, attachment points, and force vector alignment with anatomical axes. Optimal positioning on the posterior thoracolumbar region allows the application of assistive or resistive forces tailored to patient posture and muscle capacity. Modeling studies using finite element analysis and musculoskeletal simulations show that leaf springs modulate spinal loading by offloading vertebral structures and enhancing muscle activation through controlled resistance. Variations in spring material, thickness, and length affect stiffness and corrective moment magnitude, influencing comfort and clinical efficacy. These biomechanical insights guide the engineering of orthotic designs that dynamically interact with trunk biomechanics for improved kyphosis management [14,19,24]. Leaf springs function by selectively assisting or resisting trunk movements to offload bony and soft tissues or augment muscular performance. Their configuration—including positioning, attachment, material type, thickness, and length—dictates the magnitude and direction of the applied force, making them versatile for both exoskeletal and orthotic applications [14]. In exoskeletons, leaf springs are primarily protective, engineered to reduce lumbar loading during lifting or minimize strain during repetitive tasks [25]. They are typically offset from the body to align centers of rotation, generating corrective momentum while mitigating mechanical stress on anatomy and skin irritation from rigid contact [26,27]. In spinal orthoses, however, leaf springs are used chiefly to induce corrective spinal positioning. Positioned parallel and posterior to the spine, they apply controlled resistive or assistive forces directly to the torso to improve muscle function and, consequently, spinal alignment [6,8]. Unlike in exoskeletons, they often maintain direct body contact, with spacing adjustments used to modulate force or permit limited postural adjustment without rigid immobilization.
Whole-body vibration training and localized vibration systems have been shown to increase muscle mass and strength, counteracting sarcopenia and muscle fatigue. Randomized trials demonstrate that vibration frequencies around 60 Hz, with controlled amplitude, significantly enhance lower limb muscle activation and strength gains in seniors. Local vibration therapy, while established for muscle hypertrophy and fatigue reduction in sarcopenic older adults [28], remains underexplored in spinal orthoses for hyperkyphosis-related back extensor weakness. Prior applications focused on postural cues in young participants rather than strength gains in older individuals with hyperkyphosis [11,29,30]. Investigations integrating local vibration with spinal orthoses have primarily used parameters (eg, frequency, force amplitude) designed to enhance proprioception and postural awareness, rather than to induce muscular strengthening. To date, the application of local vibration within spinal supports has been limited primarily to serving as a postural reminder or neuromuscular stimulator in young, healthy participants [11,30]. Prior to the data derived from this study [31], no research on spinal orthotics had explicitly investigated local vibration with the primary goal of improving muscle strength. Using vibration parameters intended for strength adaptation in a wearable device, without a programmed and controlled application, carries a risk of adverse effects, including potential damage to skin and muscle tissues. This highlights the novelty of programmable vibration integration for safe, controlled muscle stimulation.
Programmable vibration devices allow safe, adjustable stimulation intensity addressing concerns about muscle or skin injury, making them a promising adjunct in geriatric orthopedic rehabilitation [31]. This study details the technical development and feasibility evaluation of a novel orthosis, based on the results from the first author’s doctoral research. It also specifically focuses on the technical specifications and development process of a novel semirigid thoracolumbar spinal orthosis equipped with 2 leaf springs and a programmable local vibration system, designed to reduce TKA and improve trunk muscle strength.
DiscussionOverviewThis technical note describes the development and feasibility evaluation of a semirigid thoracolumbar orthosis integrating a programmable local vibration system. The study had 3 phases: (1) bench testing confirming technical specifications (8 Hz, 10-mm free displacement, 9.5 N/cm² on-body contact pressure); (2) short-term assessment in 10 participants showing acceptable comfort (7.6 out of 10) and appearance (7.95 out of 10) with no skin safety concerns; and (3) a 4-week single-participant pilot demonstrating 100% adherence, no adverse events, and descriptive pre-post changes in kyphosis angle and muscle function outcomes.
Technical FeasibilityThe bench testing confirmed that the electromechanical system met its design specifications. The 8-Hz vibration frequency was reliably achieved, and the on-body contact pressure of 9.5 N/cm² (95 kPa) was lower than the free-state pressure (20.0 N/cm²) due to tissue compliance and padding. This reduction is consistent with previous reports of wearable vibration devices [32] and suggests that the system delivers mechanical stimulation within a range that is likely safe for soft tissue, though formal tissue tolerance studies would be needed to confirm.
The choice of 8 Hz frequency warrants discussion. While higher frequencies (30‐60 Hz) are often used for proprioceptive training [37,38], lower frequencies (8‐15 Hz) have been shown to improve muscle strength with potentially lower risk of tissue irritation [39,40]. Since participants in this study received vibration during seated isometric postural holding, 8 Hz was selected as a conservative starting point that balances potential neuromuscular effects with safety. The 10-second vibration bursts with 5-minute rest intervals were based on neurophysiological evidence indicating that muscle spindle afferents require approximately 6 seconds to return to baseline activity following mechanical stimulation [41].
Short-Term Acceptability and Safety (Phase 2)The short-term assessment in 10 participants with hyperkyphosis revealed acceptable scores for appearance (7.95/10) and comfort (7.6/10). Among these participants, 3 noted specific concerns: 1 reported the orthosis was “heavy,” 1 reported it was “hard to wear,” and 1 reported no specific concerns. No skin-related adverse events were observed after 30 minutes of wear. These findings suggest that the orthosis is acceptable for short-term use, though the “hard to wear” comment highlights the need for improved donning instructions or design modifications.
Four-Week Single-Participant Pilot (Phase 3)FeasibilityThe participant completed all prescribed sessions with a mean daily wear time of 1.4 (SD 0.3) hours, meeting the prescribed target of 1 to 2 hours. No technical failures, battery depletions, or adverse events occurred. These feasibility metrics support the safe application of the device for home use in future larger studies.
Thoracic Kyphosis AngleThe observed reduction of 1.43° (from 47.24° to 45.81°) did not exceed the MDC=4.62° established for photogrammetry [34]. Therefore, this change cannot be interpreted as a clinically meaningful improvement. Possible explanations for the modest change include the short intervention duration (4 wk) or the mild baseline kyphosis (<52°). Future studies with longer intervention periods and participants with more severe kyphosis may be needed to detect meaningful spinal alignment changes.
Muscle Function Outcomes Based on Descriptive FindingsPre-post changes in muscle function are reported descriptively (Table 3) and should be interpreted with caution, given the single-participant design. Several observations are noted, including the following: isometric extensor peak torque increased by 5 Nm (from 19 to 24 Nm), whereas isometric flexor peak torque decreased by 9 Nm (from 34 to 25 Nm). Isokinetic concentric peak torque increased for both flexors (from 6 to 20 Nm) and extensors (from 12 to 20 Nm). Time-to-peak torque decreased for both flexors and extensors across several test types.
While these descriptive changes are notable, they cannot be attributed to the intervention with certainty due to the absence of a control group, repeated measures, or inferential statistics. The patterns observed may reflect true neuromuscular changes, practice effects, measurement variability, or a combination of factors. Importantly, no electromyography or mechanistic measurements were collected; therefore, proposed mechanisms such as altered cocontraction, reciprocal inhibition, or motor unit recruitment remain speculative hypotheses requiring direct investigation in future studies.
Previous studies of local vibration in spinal orthoses have primarily used vibration as a postural reminder (eg, 30‐60 Hz vibration motors) rather than as a therapeutic stimulus for muscle strengthening [11,29,30]. These studies reported reductions in kyphosis angle of approximately 8° after 4 days of use [30], which exceeds the change observed in this study. The difference may be due to the younger population studied (mean age 28.06, SD 6.06 y), the type of vibration (high-frequency, low-amplitude vibration motors vs 8 Hz solenoid drivers), or the vibration protocol (continuous vs intermittent).
In contrast, this study used a lower frequency (8 Hz) with higher force (20 N free, 9.5 N worn) delivered intermittently (10 s on, 5 min off). This protocol was designed to prioritize safety for older adults and to avoid potential overstimulation. The trade-off may be a slower onset of therapeutic effects, which could require longer intervention periods to detect meaningful changes.
This feasibility study provides the technical foundation for the trial and demonstrates that the device is safe and usable before proceeding to larger controlled studies. Based on descriptive results, we hypothesized mechanisms that require future investigation. The following mechanisms are speculative hypotheses derived from the descriptive findings. They were not directly measured in this study and hence require confirmation through electromyography, neurophysiological, or mechanistic studies. The descriptive pattern of muscle function changes—increased isometric extensor torque but decreased isometric flexor torque, alongside increased isokinetic torque in both muscle groups—might be explained by several hypothetical mechanisms:
Reduced antagonistic cocontraction: a decrease in flexor torque during isometric testing might indicate reduced cocontraction of antagonist muscles, which could improve postural efficiency and reduce spinal compressive loads [42]. However, without electromyography data, this remains speculative.
Task-specific neuromuscular adaptation: the contrasting results between isometric (flexor torque decrease) and isokinetic (flexor torque increase) tests might suggest that the intervention affects motor control strategies differently depending on the contraction type and speed [43]. This hypothesis requires direct testing.
Vibration-induced neural effects: local vibration is known to influence muscle spindle afferents and may alter cortical and spinal excitability [44,45]. These effects could theoretically lead to improved motor unit recruitment efficiency, as suggested by a decreased time to peak torque [46]. However, no neurophysiological measurements were collected in this study.
Future studies should incorporate surface electromyography to quantify agonist-antagonist activation ratios, cocontraction indices, and recruitment patterns during functional tasks. Mechanistic studies using transcranial magnetic stimulation or H-reflex testing are needed to confirm central versus peripheral adaptations.
LimitationsThis study has several important limitations as discussed in the following
Single-participant pilot (phase 3): Only 1 participant completed the 4-week intervention. Therefore, the findings cannot be generalized to the broader population of individuals with hyperkyphosis. Pre-post changes are reported descriptively without inferential statistics (no P values, CIs, or effect sizes), as such tests are not valid for N=1.
No control group: The absence of a control or comparison group means that observed pre-post changes cannot be attributed to the intervention with certainty. Practice effects, natural history, placebo effects, or measurement variability may account for some or all of the observed changes.
Short intervention duration: Four weeks may be insufficient to detect meaningful changes in spinal alignment, particularly in individuals with mild baseline kyphosis (47.24°). Longer intervention periods (eg, 12‐24 wk) may be needed for structural spinal changes.
No mechanistic measurements: electromyography, neurophysiological (H-reflex, transcranial magnetic stimulation), or imaging studies were not conducted. Therefore, the proposed mechanisms involving cocontraction, motor unit recruitment, or neural adaptation remain speculative and should be interpreted as hypotheses requiring future testing.
Modest kyphosis change: The 1.43° reduction in TKA did not exceed the MDC (4.62°), indicating that this change was not clinically meaningful. This finding should not be interpreted as evidence of clinical effectiveness.
Single site, single orthotist: All assessments and fittings were conducted by the same research team at one institution, which may limit reproducibility and generalizability.
Participant selection: Phase 2 participants (N=10) had a mean age of 62.2 (SD 7.83; range 45‐77) years and a mean kyphosis of 55.53° (SD 1.69°). The phase 3 participant was 45 years old with milder kyphosis (47.24°), limiting the applicability to older adults or those with more severe hyperkyphosis. Outcomes were assessed immediately after the 4-week intervention. The durability of any observed changes and long-term safety remain unknown.
Clinical Implications (Preliminary)At present, this device should be considered an investigational tool rather than a clinically proven intervention. Based on this feasibility study alone, the orthosis cannot be recommended for clinical use to improve kyphosis angle or trunk muscle strength. The primary contributions of this study are as follows: (1) demonstrating that a programmable vibration system can be successfully integrated into a semirigid spinal orthosis, (2) providing bench-tested technical specifications, and (3) establishing preliminary safety and adherence data to justify larger controlled trials. Clinicians and researchers may use these findings to guide the design of future studies but should not interpret the descriptive pre-post changes as evidence of clinical effectiveness.
ConclusionA semirigid thoracolumbar orthosis with integrated programmable vibration (8 Hz, 20 N free force, 9.5 N/cm² on-body contact pressure) was successfully developed and tested. Bench testing confirmed reliable technical performance. Short-term assessment in 10 participants with hyperkyphosis indicated acceptable comfort and appearance (scores 7.6‐7.95 out of 10), with no immediate skin safety concerns. In a 4-week single-participant pilot, adherence was 100%, daily wear time met the prescribed target (mean 1.4, SD 0.3 h), and no adverse events were reported. The observed 1.43° reduction in TKA did not exceed the MDC (4.62°) and, therefore, was not clinically meaningful. Descriptive pre-post changes in muscle function outcomes (eg, isokinetic peak torque increases of 8‐14 Nm) are reported but cannot be attributed to the intervention without a control group or inferential statistics. This study demonstrates technical feasibility and safety, justifying progression to larger controlled studies with adequate sample sizes, control groups, and mechanistic measurements (eg, electromyography) to determine whether this device improves clinical outcomes in hyperkyphosis.