Clinical Efficacy of an Innovative Multidimensional Traction Therapy in Moderate Adolescent Idiopathic Scoliosis

Summary

This protocol reports a randomized clinical trial that uses the major-curve Cobb angle as the primary structural outcome to evaluate the efficacy of multidimensional traction combined with bracing and mirror-corrective exercises in moderate adolescent idiopathic scoliosis.

Abstract

This protocol aims to determine whether a patient-specific, multidimensional traction (MDT regimen), delivered alongside standard bracing and mirror-corrective exercises, can improve structural correction in moderate adolescent idiopathic scoliosis (AIS). The study is a pilot randomized clinical trial (RCT) enrolling skeletally immature adolescents (Cobb 20 - 45°) and allocating them 1:1 to (i) brace plus mirror-corrective exercises or (ii) the same program augmented by MDT. Traction vectors are individualized to the curve pattern and flexibility. Interventions are provided over a 3-month period. The MDT arm receives supervised sessions 2x a week, and all participants complete standardized daily exercises and wear braces according to guideline-based prescriptions. The primary outcome is the change in the major-curve Cobb angle from baseline to post-treatment, measured on standardized standing radiographs using consistent end vertebrae and blinded dual reading. Safety, adherence, and adverse events are monitored throughout. In results, adding MDT to bracing and exercises yielded a greater reduction in the major-curve Cobb angle. This protocol operationalizes a reproducible, clinic-ready workflow for parameterized multi-vector traction as an adjunct to conventional nonoperative care, offering a framework to maximize immediate in-brace correction and promote follow-up curve regression. This trial provides methodology and preliminary evidence to support larger, longer follow-up studies through skeletal maturity.

Introduction

AIS is a three-dimensional structural deformity of the spine with unknown etiology, characterized by axial rotation, coronal curvature, and sagittal imbalance (typically thoracic hypokyphosis or lumbar hyperlordosis)¹ . AIS primarily occurs in skeletally immature children and adolescents who are otherwise healthy, most commonly during the adolescent growth spurt. In severe cases, it can lead to chronic back pain, limitation of cardiopulmonary function, and substantial psychological burden, thereby impairing growth, development, and quality of life².

Currently, the diagnosis of scoliosis relies primarily on standing spine radiographs; the core criterion is the Cobb angle, measured as the angle between the endplate lines of the superior and inferior end vertebrae of the curve. A Cobb angle ≥ 10° defines scoliosis; curves of approximately 20° - 45° are classified as moderate AIS according to international grading³. In Chinese adolescents, epidemiologic data estimates suggest that roughly 19.5% of diagnosed cases are moderate, highlighting the clinical relevance of this cohort⁴. Management of moderate AIS is primarily non-operative, emphasizing prevention from development to severe deformity (and the attendant possibility of surgery) while striving for the greatest feasible three-dimensional correction and preservation of global spinal alignment⁵.

Brace therapy represents a standard, non-invasive treatment for moderate AIS. However, when evaluated from the perspectives of correction and stabilization, long-term follow-up shows that its primary effect is to prevent curve development, with only a limited proportion of patients achieving significant and sustained angle reduction^6,7,8. Furthermore, delayed development may occur after discontinuation⁹. Therefore, under standardized bracing prescription and adherence, it is essential to integrate novel evidence-based non-operative strategies to strictly control development and maximize immediate/long-term curve improvement^8,10.

Traction therapy aims to improve the extensibility of the spine and surrounding soft tissues. Currently, halo-pelvic traction is commonly applied preoperatively in patients with severe scoliosis to reduce spinal deformity, improve pulmonary function, and lower surgical risk^11,12. A 2022 systematic review including 24 studies and 694 patients reported that, compared with pre-traction values, mean coronal Cobb angle reduction was 27.66° after traction and 47.43° after surgery, while sagittal Cobb angle reduction was 27.23° after traction and 36.77° after surgery; forced vital capacity (FVC) increased by 8.44%¹³. In addition, a 2023 systematic review comprising 8 studies and 210 patients with severe scoliosis demonstrated that preoperative halo-pelvic traction significantly reduced both coronal and sagittal Cobb angles and improved FVC and forced expiratory volume in 1 s (FEV₁)¹².

However, such traction techniques are primarily used as preoperative adjuncts and have not been effectively applied as conservative treatments for patients with moderate scoliosis. Moreover, these approaches generally lack the ability to individualize traction strategies according to patient-specific curve characteristics¹¹. Building upon the above concept, a multidimensional traction paradigm guided by the patient's curve pattern and flexibility is advocated. Instead of applying longitudinal distraction forces along the vertical axis, traction vectors are oriented according to curve morphology, integrating targeted coronal displacement and transverse anti-rotation torque^14,15.

MDT is intended for adolescents with mild to moderate AIS (Cobb angle 10°-45°) and a Risser sign < 5, and is applied as an adjunct to brace treatment and scoliosis-specific corrective exercises to maximize curve correction within a nonoperative framework¹⁶. The procedure requires dedicated traction equipment and trained therapists, with individualized traction patterns evaluated before each session using mirror-corrective exercises and adjusted as needed. This approach is not suitable for patients with severe pulmonary impairment, prior spinal surgery, or secondary scoliosis, including neuromuscular scoliosis; during traction, patient tolerance must be continuously monitored, and traction force should be reduced or the procedure discontinued immediately if symptoms such as dyspnea or dizziness occur¹⁶.

This article presents a pilot study that provides a systematic and standardized description of a multidimensional traction protocol for moderate AIS. Using a small-sample exploratory design, the study aimed to preliminarily assess feasibility, safety, and potential therapeutic effects, thereby informing the design of future large-scale, systematic cohort studies. The protocol covers patient selection and management, preparation of equipment and materials, adjustment of traction device parameters, and the sequence of corrective maneuvers.

A three-tier methodological framework was implemented, comprising randomized allocation, clinician role segregation, and blinded outcome assessment. Randomization was conducted using a sequentially numbered, opaque, sealed envelope (SNOSE) method to balance baseline confounders, including age and skeletal maturity. In addition, strict functional separation between treating physicians and radiographic assessors minimized subjective bias and ensured the integrity of the primary outcome measure, the Cobb angle.

Protocol

This protocol outlines a clinical trial employing multidimensional traction combined with bracing and mirror-corrective exercises for the correction of moderate scoliosis. The study protocol was approved by the Ethics Committee of Human Research at the Ninth People's Hospital of Wuxi, affiliated with Soochow University (approval number KS2025036). Before the trial commenced, written informed consent was secured from each participant. From February to May 2025, the Department of Rehabilitation at the Ninth People's Hospital of Wuxi recruited 24 adolescents with AIS who met the inclusion criteria.

1. Experimental setup

1. Randomize eligible participants into two parallel arms: the control group (Brace combined with mirror-corrective exercises) and the treatment group (Brace combined with mirror-corrective exercises and multidimensional traction therapy). Assign 12 patients to each group¹⁷.
   1. Create 20 identical cards (5 cm x 5 cm) labeled Brace or MDT (10 each), seal them individually in sequentially numbered, black-lined opaque envelopes, and shake them in a container for 1 min to randomize the order.
   2. Store the envelopes in a locked cabinet and extract the next envelope in numerical sequence only after a participant has signed the informed consent and completed all baseline assessments. Open the envelope in a private setting, log the assigned group in a master record, and conceal the allocation from the outcome assessors throughout the study.
2. Ensure all rehabilitation physicians administering the treatment are familiar with spinal anatomy and the specific scoliosis traction correction techniques employed in this study.
3. Prescribe each patient (in both the Brace and MDT) a brace tailored to individual parameters by senior orthotists. Require the patients to wear the brace for at least 8 h daily⁸, as recommended by clinical guidelines⁵. Ensure the entire treatment period lasts for 3 months.
4. Provide structured training and instruction to parents. Instruct patients to perform mirror-corrective exercises at home under parental supervision or in front of a mirror, completing approximately 100 repetitions per day divided into 2-3 sets over the entire 3 months treatment period. Reassess and adjust the exercise program weekly by a therapist.
5. Apply traction 2x per week and adjust the traction configuration weekly based on therapist assessment.
6. Patients in both groups received a custom-designed, hypercorrective Boston scoliosis night brace (Camp Scandinavia AB)⁵.

2. Recruitment of patients

1. Verify the inclusion criteria are met, as follows: Confirm skeletal immaturity with a Risser sign between 0 and 4, ensure the age is between 10 and 18 years old³, and confirm a diagnosis of idiopathic scoliosis with a structural thoracic or thoracolumbar curve where the apical vertebra is between T3 to L3. Determine the Cobb angle on standing radiographs, which is between 20° and 45°, and assess normal cognitive function and the ability to understand instructions and cooperate with corrective exercises.
2. Exclude participants who present with any of the following criteria: Identify patients with severe cardiopulmonary disease, rib deformity, or a history of prior spine surgery. Additionally, exclude patients diagnosed with neuromuscular scoliosis and any other form of secondary scoliosis.

3. Radiographic assessment

1. Acquire radiographs using a digital radiography system with a typical tube voltage of 80-100 kV and automatic exposure control, following routine full-spine imaging protocols¹⁸. Standardize the source-to-image distance and patient positioning across all examinations.
2. Define the primary structural outcome as the change difference in the main-curve Cobb angle (ΔCobb) from baseline to post-treatment. Measure ΔCobb on standardized standing posteroanterior (PA) radiographs, where ΔCobb equals baseline Cobb angle minus post-treatment Cobb angle, ensuring positive values indicate greater correction. Obtain all post-treatment measurements using the same upper and lower end vertebrae identified at baseline and record the average of measurements obtained by two blinded readers.
3. Utilize standing Posterior-Anterior (PA) radiographs as the primary assessment instrument. Measure the Cobb angle from these radiographs using a digital X-ray system equipped with PACS software.
4. Instruct patients to stand barefoot in an upright position on the X-ray platform, feet shoulder-width apart, arms relaxed at the sides. Have patients wear a properly fitted lead apron to minimize radiation exposure, ensuring that the shielding does not alter posture or obscure anatomical landmarks. Make sure the patient maintains a natural standing posture without intentional correction.
5. Acquire a standing PA radiograph of the entire spine. Import radiographs into PACS software. Identify the upper and lower end vertebrae of the major curve. Draw a line along the superior endplate of the upper end vertebra and the inferior endplate of the lower end vertebra. Draw perpendicular lines from these reference lines. Measure the angle of intersection (Cobb angle).
6. Repeat measurements and use the average value from the two independent raters. Evaluate treatment effectiveness based on the change in the major curve Cobb angle between pre- and post-treatment assessments

4. Exercise procedure

1. Tailor exercise prescriptions to each patient's curve pattern and flexibility, following the principle of mirror correction⁸. Instruct each patient to complete about 100 standardized repetitions daily, allowing division into 2-3 sets. Reassess weekly and progressively adjust exercise patterns and training dosage to balance safety and adherence¹⁵.
2. Begin with standing spinal radiographs and surface examination to determine the curve pattern (single- or multi-segment), apical vertebra, and compensatory features. Select the corrective exercise sequence according to the dominant deformity component after fully exposing the back and marking key anatomical landmarks. Perform derotation first when vertebral rotation predominates, and then coronal translation and side-bending. Perform translation and side-bending first when coronal translation predominates, then add derotation. Complete each session in the prescribed order under supervision.
   NOTE: The accuracy of the mirror-corrective posture can be assessed by exposing the patient's spine and visually comparing the spinal contour before and during correction to determine whether the curve appears more linear. Radiographic verification can further be used, with a mirror-corrective posture achieving a Cobb angle reduction of ≥50% considered appropriate for subsequent application.
3. Adopt a controlled terminology for recording movement patterns to ensure accurate documentation and standardized execution as follows: Object moved-Thorax (T) or Lumbar (L); Movement type-Rotation (R) or Translation (Tr); Axes of movement-x (mediolateral), y (cranial-caudal), z (anterior-posterior; Figure 1A); Direction + denotes left/cranial/anterior and − denotes right/caudal/posterior. Notation takes the form [sign][movement][axis][region]; for example, Corrective Exercise Left Translation of Thorax was recorded using the abbreviation +TxT to facilitate standardized data management (Table 1).
4. Ask the patient to fold their arms across the chest and grasp the opposite shoulders. Ensure that the intercostal and interscapular regions are fully expanded and stabilize the movement pattern prior to starting the exercise.
5. Translate the rib cage to the left and right in the coronal plane using a rib-driven thoracic motion pattern. Avoid rotation and lateral flexion. Maintain natural breathing throughout and hold the end position for 3s (Figure 1B,C).
   NOTE: Engage the ribcage to drive thoracic movements. Avoid recruiting other body regions during exercise.
6. Use the rib-thoracic coupled activation pattern to perform left-right lateral flexion in the coronal plane. Avoid axial rotation and maintain natural breathing. Hold the end position for 3s (Figure 1D,E).
7. Engage the ribcage to drive thoracic rotation in the horizontal plane. Avoid any side translation or side-bending. Maintain natural breathing during the movement and hold the end range for 3s (Figure 1F,G).
8. Ask the patient to engage the core via deliberate abdominal contraction to initiate lumbar movement. Translate the trunk left and right in the coronal plane without any rotation or side-bending. Breathe naturally throughout the task and hold the end range for 3s (Figure 2A,B).
   1. Apply gentle manual perturbations to the patient in a standing or seated position and instruct the patient to maintain the posture without displacement to facilitate perception and activation of core engagement.
9. Ask the patient to engage the core via deliberate abdominal contraction to initiate lumbar-driven motion. Side-bend to both sides in the coronal plane. Do not allow axial rotation; keep the pelvis neutral and maintain natural breathing. Hold for 3s at each terminal range (Figure 2C,D).
10. Ask the patient to engage the core via abdominal contraction, then rotate left and right in the horizontal (transverse) plane using lumbar motion. Avoid coronal translation and side-bending. Breathe naturally, and hold the terminal position for 3s (Figure 2E,F)

5. Traction procedure

1. Arrange the following equipment for traction: Traction frame (Figure 3A), Foam roller (Figure 3B), Fixation strap (Figure 3C), Pull strap (Figure 3D).
2. Administer the comprehensive intervention to the treatment group, consisting of bracing, mirror-corrective exercises, and multi-vector traction. Provide patients in the treatment group with traction therapy twice a week for a total duration of 3 months, with an interval of at least 2 days between sessions.
   1. Monitor the patient's subjective response continuously during the initial traction session. Reduce the traction force to a level at which only a tolerable tension sensation is perceived if the patient experiences obvious pain or discomfort in addition to traction tension.
3. Position the patient comfortably in the traction chair, ensuring the trunk remains upright. Place foam pads at both the anterior superior iliac spine and the iliac crest margin on the concave side of the curve to buffer strap pressure and reduce local discomfort.
   1. Place a foam roller or cushion under the feet per patient's height and leg length to enhance plantar support, maintain pelvic neutrality, and prevent excessive forward tilt or sitting instability.
4. Wrap the fixation strap around the anterior superior iliac spine and iliac crest on the concave side, with foam padding placed beneath the strap, secure the oblique buckles on both sides of the strap to the buckle located at the bottom of the traction frame on the convex side, then tighten the pelvic fixation strap to maintain the pelvis in a neutral position and check strap tension and symmetry (Figure 4A).
5. Move the roller on the concave side of the lumbar curve to the corresponding hole on the traction column that is level with the lumbar apical vertebra. Identify the apical level using surface anatomical landmarks and maintain the patient in the customized mirror-corrective posture throughout the adjustment (Figure 4A).
   1. Localize the lumbar vertebrae by using consistent surface landmarks. Specifically, identify the umbilicus, as it usually corresponds to the vertebral level of L3 - L4.
6. Position the wide central portion of the traction strap over the apical region of the lumbar curve, secure both side buckles into the corresponding locking holes on the pull bar, then rotate the adjustment wheel on the concave-side traction column clockwise to apply sufficient traction tension without causing obvious patient discomfort (Figure 4A).
   1. Determine traction tension based on the patient's self-reported tension level using a 1-10 scale, where 10 represents the physiological tolerance limit. Maintain traction at a tension level of approximately 6 during the initial session and increase to a level of 8-9 in subsequent sessions, provided that the patient remains comfortable and no adverse symptoms occur.
7. Move the roller on the concave side of the thoracic curve to the corresponding hole on the traction column that is one level superior to the thoracic apical vertebra. Identify the apical level using surface anatomical landmarks and keep the patient in the customized mirror-corrective posture throughout the adjustment (Figure 4A).
   1. Identify the key anatomical landmarks for rib localization to facilitate assessment. Note that the acromion is approximately aligned with the second rib, the spine of the scapula corresponds to the third rib, and the inferior angle of the scapula aligns approximately with the seventh rib.
8. Position the wide central portion of the traction strap at the level of the two ribs immediately inferior to the apical vertebra of the thoracic curve, secure both side buckles into the corresponding locking holes on the pull bar, then rotate the adjustment wheel on the concave-side traction column clockwise to apply sufficient traction tension without causing obvious patient discomfort (Figure 4B).
   1. Ensure effective transmission of traction to the target thoracic segment and align traction vectors along the oblique orientation of the ribs rather than strictly along the spinal axis. Apply segment-specific contact over the corresponding rib arc. Direct traction forces through the rib-vertebra articulation to the target thoracic segment.
9. Place a foam pad in the axillary-intercostal region on the concave side. Route the fixation strap through the axilla along the intercostal space and secure both sides buckle obliquely to the fixation buckle on the traction column on the convex side of the thoracic curve. Adjust the strap to an appropriate tension that allows the patient to maintain the prescribed mirror-corrective posture under traction without the strap force disrupting the posture.
10. Follow a bottom-up sequence when setting up the traction system, starting from the lumbar spine and progressing to the thoracic spine for patients with multi-segment scoliosis. Secure all fixation and traction straps and perform a final traction adjustment to ensure moderate tension without causing patient discomfort.
11. Begin the initial session with a duration of 5-8 min. For subsequent sessions, increase the duration by 3-5 min, aiming for an optimal treatment time of 15-20 min. Do not exceed 20 min in any session.
12. Maintain therapist presence throughout the entire traction session and instruct the participant to report any discomfort immediately. In sessions following the initial treatment, assess the participant's adaptation to traction tension 5 min after initiation. Adjust the tension using the adjustment wheel according to the participant's subjective feedback. Target a perceived tightness of 8-9 on a 10-point scale to achieve optimal therapeutic effect.
13. Rotate the adjustment wheel counterclockwise to gradually reduce the traction force. Release the traction strap from the pull bar and the fixation strap from their buckles. Once the entire traction apparatus is removed, instruct the participant to remain seated and breathe calmly for 1 min. Then, have the participant stand up slowly and perform gentle movements. For patients with multi-segment scoliosis, release the traction in a top-to-bottom sequence.

6. Statistical analysis

1. Present the continuous variables as mean ± standard deviation (SD). Use the independent sample t-test to compare the differences between the two groups. All statistical tests were two-sided, and p < 0.05 was considered statistically significant. All statistical procedures and data visualizations were performed using GraphPad Prism 8.0.
2. Pre-specify the Last Observation Carried Forward (LOCF) method as the contingency approach for handling any missing follow-up data.

Results

A total of 12 patients in the control group and 12 patients in the treatment group were included in the final statistical analysis. The patient demographic is provided in Table 2.

In the treatment group, one participant reported mild lower-back soreness following traction therapy, had no further soreness, and no impact on daily activities or follow-up adherence after physician-guided adjustment of movement patterns. No significant adverse events occurred in the control group, ...

Discussion

In this randomized controlled trial, 24 adolescents (male and female, all fulfilling the inclusion criteria) were randomly assigned in a 1:1 ratio to two groups: (1) brace combined with corrective exercise; (2) brace combined with corrective exercise plus individualized multidimensional traction. The primary outcome was the change in Cobb angle (from baseline to endpoint), measured under standardized radiographic conditions and evaluated by blinded assessors. Results showed that, on the basis of bracing to suppress devel...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Digital Radiography (DR) System	Philips Medical (Suzhou) Co., Ltd.	DigitalDiagnostC50	Imaging Examination
Fixation strap	Aspine	Traction Device	Traction Device is used for the correction of scoliosis and restoration of physiological spinal curvatures.
Foam roller	Aspine	Foam roller	Traction Device is used for the correction of scoliosis and restoration of physiological spinal curvatures.
Pull strap	Aspine	Traction Device	Traction Device is used for the correction of scoliosis and restoration of physiological spinal curvatures.
Traction Device	Aspine	Traction Device	Traction Device is used for the correction of scoliosis and restoration of physiological spinal curvatures.