Acupuncture Combined with Core Stability Exercise for Pain, Function, and Postural Control Impairments in Chronic Low Back Pain: A Randomized Controlled Trial Protocol

Abstract

Background

Chronic low back pain (cLBP) is a major cause of disability, productivity loss, and healthcare burden. Its recurrence has been associated with impaired postural control, particularly altered anticipatory postural adjustments (APAs). Core stability exercise (CSE) may improve compensatory postural adjustments (CPAs), but its effects on APAs are limited. Acupuncture is effective for cLBP; however, whether it modulates APA-related feedforward control remains unclear.

Objective

This exploratory randomised controlled trial aims to evaluate the feasibility and preliminary effectiveness of acupuncture combined with CSE and Self-Compassion Training (SCT) for improving pain, function, pain-related psychological factors, and postural control in patients with cLBP.

Patients and Methods

This assessor-blinded, parallel-group trial will enrol 60 patients with cLBP, randomised 1:1 to a study group (SG) or control group (CG). The CG will receive CSE plus SCT, whereas the SG will receive Yaosan acupuncture in addition to CSE and SCT for 4 weeks. Primary outcomes include APA activation timing, APA activation amplitude, CPA activation amplitude, and centre-of-pressure sway area and path length, assessed at baseline and post-intervention. Secondary outcomes include pain intensity, disability/function, pain catastrophizing, fear-avoidance beliefs, and pain self-efficacy, assessed using the VAS, RMDQ, PCS, FABQ, and PSEQ at baseline, post-intervention, and 3-month follow-up. Safety outcomes include acupuncture-related adverse events, blood pressure, pulse, and post-exercise muscle soreness. Analyses will be performed using SPSS 25.0, with mixed-effects models as the primary analytical approach for repeated continuous outcomes.

Conclusion

This trial will provide preliminary evidence on whether adjunctive acupuncture improves pain, function, psychological outcomes, and postural control and modulates APA-related feedforward mechanisms in cLBP.

Trial Registration

ITMCTR2026000520.

Introduction

Low back pain (LBP) is typically defined as pain located between the inferior margin of the 12th rib and the superior border of the gluteal folds. It is a major contributor to the global burden of disease and one of the leading causes of years lived with disability (YLDs) worldwide.^1,2 The prevalence of LBP is approximately 12% in the general adult population, increases markedly among individuals aged 40 years and older, and reaches a lifetime prevalence of up to 40%.³ In China, data from 2016 indicated that approximately 67.3 million people were affected by LBP, with an increasing incidence and a trend towards younger age groups.⁴ Globally, 619 million people were affected by LBP in 2020, and this number is projected to increase to 843 million by 2050 owing to lifestyle changes and population ageing.⁵ The high prevalence of LBP results in substantial productivity loss and socioeconomic burden. Approximately 15.4% of workers in the United States lose an average of 10.5 workdays per year due to LBP, resulting in a considerable number of lost workdays annually.^5,6 Data from Brazil between 2012 and 2016 showed that LBP resulted in 59 million missed workdays and direct societal costs of USD 2.2 billion.⁷ Overall, LBP has become an increasingly serious global public health issue.

Chronic low back pain (cLBP) is generally defined as low back pain persisting for more than three months. Notably, approximately 90% of cLBP cases are non-specific. The diagnosis is primarily based on symptom presentation and the exclusion of specific causes, rather than on identifiable structural pathology. Specific aetiologies, such as infection, tumour, and fracture, should be excluded during diagnostic evaluation. This makes treatment challenging, underscoring the need for more effective, mechanism-based therapeutic approaches.³

Postural control refers to the ability to maintain body stability by regulating the centre of pressure (COP) and the spatial orientation of body segments. Its core processes include anticipatory postural adjustments (APAs) and compensatory postural adjustments (CPAs).^8,9 APAs represent a feedforward control mechanism by which the central nervous system pre-activates core musculature to minimise predictable perturbations before they occur. In contrast, CPAs are feedback-driven regulatory responses that occur after perturbations, restoring balance. These processes work synergistically to maintain postural stability.

Notably, APAs and CPAs represent functionally distinct but complementary aspects of postural control. APAs are proactive, subcortically mediated feedforward processes that depend on prior experience and sensorimotor prediction, whereas CPAs are reactive, feedback-driven responses to actual perturbations. This dissociation has important clinical implications: impairments in feedforward APAs cannot be fully compensated by reactive CPAs because CPAs occur only after perturbations have already compromised spinal stability. Thus, specifically targeting APAs may be essential for addressing the core postural control deficit in cLBP.^8–10

Previous studies suggest that abnormalities in postural control are important contributors to the onset and recurrence of LBP.¹⁰ In addition, a systematic review and meta-analysis by Knox et al confirmed that patients with chronic LBP exhibit delayed muscle onset in response to both expected and unexpected perturbations, indicating that both APAs and CPAs are altered in this population.¹¹ Patients with cLBP may exhibit characteristic impairments, including delayed APA activation and CPA dysfunction, such as delayed onset of the transversus abdominis, multifidus, and other core postural muscles during rapid arm movements and external perturbations.^10,11 Therefore, targeted improvement of APA and CPA functions may provide new avenues for the management of LBP.

Currently, first-line rehabilitation interventions for cLBP primarily focus on exercise therapy. However, their mechanisms largely target feedback pathways represented by CPAs, with relatively limited effects on feedforward APAs.¹² For instance, a randomised controlled trial by Vasseljen et al found that 8 weeks of core stability exercise failed to improve feedforward activation of the deep abdominal muscles in patients with chronic LBP.¹³ Similarly, Lomond et al reported that 6-week stabilisation or movement-system-impairment-based treatment did not significantly improve APAs.¹⁴ These findings suggest that standard exercise protocols, although effective at enhancing reactive CPAs, do not adequately address deficits in feedforward APA.

Accordingly, exploring interventions that specifically modulate APAs and combining them with conventional training that improves CPAs to establish a “feedforward–feedback” bidirectional regulatory strategy has become an important research direction. As a traditional Chinese therapeutic modality, acupuncture has demonstrated benefits in treating cLBP, including pain relief, restoration of lumbar muscle function, and improved coordination and stability.¹⁵ Acupuncture is thought to modulate somatosensory afference and central sensorimotor integration, both of which are critically involved in feedforward postural control.^16,17 However, whether acupuncture exerts its therapeutic effects by modulating APAs, a specific neuromuscular control mechanism, has not been directly tested.

Literature Review

Patients with cLBP exhibit characteristic abnormalities in APAs and CPAs, which are closely associated with persistent pain.¹⁸ For example, during endogenous postural perturbation tasks, such as bilateral arm elevation, patients show significant delays in APA activation of the external oblique and the transversus abdominis/internal oblique muscles.¹⁹ Yu et al²⁰ further found that, under exogenous perturbations, APA activation timing in the erector spinae was prolonged; under endogenous perturbations, activation amplitudes of both APAs and CPAs were higher than in healthy individuals. In addition, under cognitive load, APAs in certain trunk muscles are not only delayed but also characterised by altered co-activation patterns. Moreover, the degree of APA delay is positively correlated with impairment in daily function.²¹ Xiao et al²² further revealed that when exposed to various perturbations, core muscles, such as the transversus abdominis and multifidus, exhibit delayed APA activation and enhanced CPA activation. Muscle fatigue further exacerbates APA abnormalities in the transversus abdominis, suggesting a potential association with LBP risk.

However, standard rehabilitation protocols, such as core stability exercises, have limited effects on improving APAs. For instance, 6-week or 8-week core stability training did not significantly improve APAs in patients with LBP.^14,23 A systematic review by Van Oosterwijck et al further confirmed that although sensorimotor control training may improve transversus abdominis APA onset, the evidence remains conflicting and the methodological quality of current studies is low.²⁴ In contrast, such training appears to effectively enhance CPAs: after 8 weeks, CPA activation amplitude of the right transversus abdominis/internal oblique increased,¹² and after 10 weeks, CPAs of the erector spinae were also significantly enhanced.²⁵ These findings suggest that current rehabilitation models may primarily improve postural stability by strengthening feedback-mediated CPAs, yet fail to effectively modulate impaired feedforward APA mechanisms. Therefore, identifying interventions that specifically improve APAs and integrating them with training strategies that enhance CPAs to achieve bidirectional coupled modulation represent key directions for optimising rehabilitation outcomes in cLBP.

Importantly, the pThe proposed combination of APA-targeted acupuncture with CPA-targeted core stability training has not previously been tested. This bidirectional, mechanism-matched approach is novel because it moves beyond traditional “one-size-fits-all” exercise prescription toward a stratified rehabilitation strategy that aligns specific interventions with distinct components of postural control impairment. By integrating traditional Chinese acupuncture with modern motor control theory, this study aims to provide a translational bridge between mechanistic understanding and clinical application in cLBP. The combination of Shenshu (BL23), Dachangshu (BL25), and Weizhong (BL40) is commonly used and has demonstrated clinical benefits in acupuncture for LBP.²⁶ Clinical effectiveness has been reported in terms of pain relief, improved lumbar function, and enhanced postural control.²⁷ The potential mechanisms are multifaceted. Peripherally, acupuncture stimulation may act on the lateral recesses of the lumbar spine and perineural regions, alleviate muscle tension, reduce inflammatory oedema of the nerve roots, and improve microcirculation, thereby optimising lumbar proprioceptive input and the readiness of the core musculature.²⁸ Centrally, functional MRI studies suggest that the analgesic effect of the “lumbar three-needle” approach is closely related to functional modulation of the mesencephalon–cingulate gyrus–thalamus–prefrontal cortex pathway,²⁹ which also participates in sensorimotor integration and anticipatory control.

Taken together, the above evidence indicates that acupuncture may treat cLBP by facilitating peripheral sensory input and central neural integration.^28,29 Notably, patients with cLBP often present with sensory abnormalities and central integration dysfunction, and these abnormalities are associated with impairments in anticipatory postural adjustments.^30,31 However, no studies have directly investigated whether acupuncture can modulate APAs, thereby exerting therapeutic effects by improving sensory input and central integration. Whether APAs can be regulated through these pathways to improve cLBP remains to be confirmed. Therefore, an in-depth investigation of the effects of acupuncture on APAs may help elucidate potential mechanisms and provide key scientific evidence to develop a new rehabilitation model that integrates Chinese and Western medicine to achieve bidirectional regulation of postural control.

Based on this theoretical framework, we hypothesise that acupuncture can improve impaired anticipatory postural adjustment function in cLBP by modulating peripheral sensory afference, optimising the timing and magnitude of motor unit recruitment, and promoting synergistic integration of sensorimotor cortical and spinal neural pathways. When combined with core stability exercise (CSE), which strengthens compensatory postural adjustment capacity, this intervention is expected to achieve bidirectional “feedforward–feedback” coupled modulation of APAs and CPAs, thereby comprehensively improving postural control and ultimately alleviating LBP and related functional impairments.

Specifically, after the intervention, the study group is expected to show significantly greater improvements in APA activation amplitude, APA activation timing, and CPA activation amplitude than the control group. Therefore, this exploratory randomised controlled trial aims to evaluate the feasibility and preliminary efficacy of adjunctive acupuncture, in addition to CSE and self-compassion training, for improving postural control, pain, function, and pain-related psychological outcomes in patients with cLBP. Furthermore, this study aims to explore whether adjunctive acupuncture modulates APA-related feedforward postural control mechanisms.

Materials and Methods

Design and Setting

This study will be conducted at the Department of Rehabilitation Medicine at Renhe Hospital, an affiliated hospital of China Three Gorges University. The study protocol is written in accordance with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2025 reporting guidelines.³² This is a prospective, randomised, assessor-blinded, parallel-group controlled trial designed to evaluate the effectiveness and safety of acupuncture combined with core stability exercise in improving impairments in anticipatory/compensatory postural adjustment in patients with cLBP. Patients with cLBP will be randomised in a 1:1 ratio to either acupuncture combined with CSE or CSE alone. The trial report will be prepared in accordance with the Consolidated Standards of Reporting Trials (CONSORT) 2025 guidelines.³³ This study was approved and supervised by the Ethics Committee of the Renhe Hospital of China Three Gorges University (RHLL-2025-07) and registered in the International Traditional Medicine Clinical Trial Registry (ITMCTR2026000520). The trial flowchart of this study is presented in Figure 1.

Figure 1.

Flow diagram of the study.

Abbreviations: CSE, core stability exercise; SCT, self‑compassion training; APAs, anticipatory postural adjustments; CPAs, compensatory postural adjustments; COP, centre of pressure; VAS, visual analogue scale; PCS, pain catastrophizing scale; FABQ, fear-avoidance beliefs questionnaire; RMDQ, roland-morris disability questionnaire; PSEQ, pain self-emcacy questionnaire.

Study enrollment is planned from April 2025 to March 2026. Data collection continues for 1 year after study enrollment and is expected to conclude in March 2027, with results anticipated by the end of 2027. Refer to Figure 1 for details of the timeline for enrolled participants.

Study Objectives

The primary objective is to assess APAs, CPAs, and postural control between groups at baseline and after 4 weeks of intervention.

• a)APAs and CPAs: The arm-raising task (ART), an instantaneous endogenous postural perturbation task, will be used to assess APA activation timing, APA activation amplitude, and CPA activation amplitude in patients with cLBP. These indices will be derived from bilateral surface electromyography (sEMG) signals recorded from core postural muscles, including the rectus abdominis (RA), transversus abdominis/internal oblique (TrA/IO), erector spinae (ES), and multifidus (MF). Participants will stand barefoot on a force platform with both arms relaxed at their sides. A visual cue for arm raising, consisting of a black dot on a white background, will be displayed on a monitor positioned 2 m in front of the participant at eye level. Upon stimulus presentation, participants will be instructed to raise the right upper limb forward and upward to shoulder level as rapidly as possible and maintain that position. When the stimulus disappears, they will lower the upper limb to the starting position and relax. Perturbation onset (T0) during the ART will be defined as the time point at which the acceleration signal from the inertial measurement unit (IMU) attached to the dominant wrist reaches 5% of the maximum acceleration amplitude. Following Yu et al,³⁴ the APA time window will be defined as −250 ms to +50 ms relative to T0, whereas the CPA time window will be defined as +50 ms to +350 ms relative to T0. To capture the temporal dynamics of postural adjustments, both epochs will be further divided into two sub-epochs: APA1, −250 to −100 ms; APA2, −100 to +50 ms; CPA1, +50 to +200 ms; and CPA2, +200 to +350 ms. Muscle activation amplitude, expressed as normalised integrated EMG amplitude (iEMG), will be calculated for each epoch using the following formulas: for the full APA and CPA epochs, iEMG = (a − 2b)/2b, where a is the raw iEMG for the epoch and b is the baseline iEMG calculated over −600 to −450 ms; for sub-epochs, which have the same duration as the baseline epoch (150 ms), iEMG = (a − b)/b.³⁵ sEMG data will be recorded using a Trigno™ wireless biofeedback system (Delsys Incorporated, USA), with one sEMG channel and nine inertial measurement unit (IMU) channels per sensor, sampled at 2000 Hz.
• b)Postural control: Static postural control will be assessed using centre-of-pressure (COP) metrics, including sway area (mm²) and path length (mm), under four conditions: bipedal stance with eyes open, bipedal stance with eyes closed, right single-leg stance with eyes open, and left single-leg stance with eyes open. Sway area will be calculated using the fitted 85% ellipse area, and path length will be defined as the total sway distance over time in two directions.³⁶ Each static postural task will last 30s and will be repeated three times, with a 2-min rest interval between trials. COP during standing will be measured using a Wii Balance Board (WBB; Nintendo Co., Ltd., Kyoto, Japan). The WBB samples at 100 Hz are portable, easy to operate, and support wireless data transmission. The validity and reliability of the WBB for assessing standing balance have been systematically evaluated. A systematic review by Clark et al,³⁷ which included 25 studies, concluded that the WBB is a reliable and valid tool for assessing standing balance in both healthy and clinical populations. Lesch et al³⁸ further confirmed that the WBB demonstrates excellent reliability for postural balance assessment, supporting its utility as a cost-effective tool for clinical balance assessment.

Secondary objectives are to assess pain intensity, pain catastrophising, pain-related fear, functional impairment related to LBP, and pain self-efficacy at baseline, at 4 weeks, and at 3 months post-intervention.

• a)Pain intensity: Assessed using the visual analogue scale (VAS),³⁹ ranging from 0 to 10; 0 indicates no pain, and 10 indicates unbearable severe pain, with higher scores reflecting greater pain intensity.
• b)Pain catastrophising: Assessed using the Pain Catastrophizing Scale (PCS).⁴⁰ The scale comprises three dimensions—helplessness (items 1–5, 12), rumination (items 8–11), and magnification (items 6, 7, 13)—for a total of 13 items. Each item is scored 0–4, where 0 indicates “not at all” and 4 indicates “always”. The total score ranges from 0 to 52, with higher scores indicating greater catastrophising.
• c)Pain-related fear: Assessed using the Fear-Avoidance Beliefs Questionnaire (FABQ).⁴¹ The scale includes two dimensions—physical activity (items 1–5) and work-related (items 6–16)—for a total of 16 items. Each item is scored 0–6, where 0 indicates “strongly disagree” and 6 indicates “strongly agree”. The total score ranges from 0 to 96, with higher scores indicating stronger fear-avoidant beliefs.
• d)Functional impairment related to LBP: Assessed using the Roland–Morris Disability Questionnaire (RMDQ),⁴² which contains 24 items scored 0–1, where 0 indicates “no” and 1 indicates “yes”. The total score ranges from 0 to 24, with higher scores indicating greater impact of LBP on daily functioning.
• e)Pain self-efficacy: Assessed using the Pain Self-Efficacy Questionnaire (PSEQ),⁴³ which includes 10 items scored 0–6, where 0 indicates “no confidence at all” and 6 indicates “complete confidence”. The total score ranges from 0 to 60, with higher scores indicating greater confidence in self-managing pain and engaging in daily activities.

Safety outcomes include acupuncture-related adverse events (local needling pain, bleeding, infection, syncope), fluctuations in blood pressure and pulse, and post-exercise muscle soreness (assessed by VAS), collected throughout the intervention period.

Participants

Sample Size Calculation

Given that this is an exploratory study, no formal sample size calculation was performed. With a 20% attrition rate, we plan to enrol 30 participants per group, for a total of 60.

Inclusion Criteria

• a)Age between 18 and 59 years;
• b)Pain located in the lumbar region, between the level below the 12th rib and above the buttocks;
• c)Chronic or recurrent LBP persisting for more than 3 months, with pain episodes in the past 3–12 months;
• d)Right-handed;
• e)VAS score ≥ 3;
• f)BMI: 18.5–29.9kg/m².

Exclusion Criteria

• a)Pelvic or spinal surgery within the past two years;
• b)Any specific lumbar pathology (eg, spinal tumours, vertebral fractures, lumbar spinal stenosis, spondylolisthesis, rheumatoid arthritis, joint ankylosis) and severe or progressive scoliosis;
• c)Radicular symptoms or neurological radiation symptoms to both lower limbs;
• d)Receiving other types of LBP treatment within the past three months;
• e)Planning pregnancy in the near future or currently pregnant;
• f)Severe dysfunction of major organs (eg, heart, lungs, kidneys);
• g)Severe visual or auditory impairment;
• h)Cognitive or psychological disorders.

Study Procedures

Screening, Inclusion and Randomisation

All patients with cLBP presenting to the outpatient clinic of the Department of Rehabilitation Medicine at Renhe Hospital, affiliated with China Three Gorges University, will undergo systematic screening by study physicians with assistance from clinical research coordinators. If a patient preliminarily meets the inclusion criteria, the study physician will explain the study purpose, procedures, and potential risks and benefits in detail. After fully understanding the study and voluntarily agreeing to participate, patients will sign a written informed consent form.

Baseline assessments will be conducted immediately after enrolment. These will include demographic data (sex, age, height, weight, BMI, and educational level), pain-related scales (VAS, PCS, FABQ, RMDQ, and PSEQ), and postural control metrics (APAs activation timing, APAs activation amplitude, CPAs activation amplitude, COP sway area, and path length). All baseline assessments must be completed before randomisation.

Randomisation will be performed by study personnel not involved in recruitment using Excel to generate the allocation sequence. Patients will be numbered in order of enrolment, random numbers will be generated with the RAND function and sorted using the RANK function, then assigned in a 1:1 ratio to two groups using the MOD function. After randomisation, patients will be assigned to the study group or the control group. The first intervention should begin within 24 hours after randomisation whenever possible, and all treatments will be administered strictly according to the protocol-defined procedures.

Blinding Process

The random allocation sequence will be generated in Excel by an independent statistician not involved in participant recruitment or assessment. The generated allocation information will be placed sequentially into opaque, sealed, and consecutively numbered envelopes. The envelopes will be kept by designated research coordinators at each centre who are not involved in participant enrolment or assessment. After a participant completes all baseline assessments and formally signs the informed consent form, the coordinator will open the next sealed envelope in sequence and inform the therapist of the participant’s assigned group (study or control) based on the allocation inside. The therapist will then implement the corresponding intervention. This process ensures that personnel responsible for recruitment and assessment cannot predict group allocation before participant enrolment.

Due to the nature of the interventions, blinding will not be applied to participants or therapists. To ensure objective outcome assessment, assessor blinding will be implemented: all evaluators responsible for data collection will remain unaware of participant group allocation. Assessments will be conducted in a space independent of the treatment area. Evaluators will not participate in any treatment activities, and participants will be instructed to avoid disclosing the treatments they have received during assessments.

Each sealed envelope will contain a backup copy of the allocation information, also sealed. Unblinding of the backup envelope will be authorised by the principal investigator only in the event of a serious adverse event (eg, severe injury induced by training) when the attending physician deems knowledge of the specific intervention essential for urgent medical management. Any unblinding event, along with its reason and date, must be documented immediately.

Interventions

All study personnel underwent standardised training and passed competency assessments to ensure protocol fidelity and consistency. Based on our previous work in patients with chronic low back pain, all participants will receive a structured Self‑Compassion Training (SCT) programme as a standardised background psychological self-management component. This programme was adapted from established self-compassion interventions and from our prior study, in which SCT was shown to improve psychological outcomes in patients with chronic low back pain.⁴⁴ In the present study, SCT is not designed as a group-specific experimental intervention, but will be delivered uniformly to all participants to provide comparable psychological support and self-management training across study groups. SCT will be delivered by qualified counsellors and will comprise four weekly face-to-face group sessions, each lasting 2 hours, together with daily structured home practice. The curriculum will focus on bodily awareness, emotional awareness, stress regulation, self-kindness, and sustained self-care skills tailored to common psychological and pain-related features of chronic low back pain. Techniques will include body-scan meditation, affectionate breathing, self-compassion writing, and brief loving-kindness meditation, among other mindfulness and self-kindness practices. Home practice will be supported via WeChat using instructional materials and daily reminders.

Control group: CSE will follow a two-stage progressive protocol—low-load exercises (Mat) and high-load exercises (Stability Ball)⁴⁵—with four sessions per week, each lasting 40 minutes. Weeks 1–2 (Low-load, mat-based): Warm-up (Lateral Bend Stretch; Supine Trunk Twist); Exercise training (Plank; Modified Side Plank; Cat-Cow Stretch; Double-Leg Abdominal Press; Bear Crawl Hold or Quadruped Position; Glute Bridge with Leg Lift; Dead Bug; Superman; Single-Leg Romanian Deadlift or Single-Leg Balance; Alternating Superman/Swimming); Cool-down (Supine Knee-to-Chest Pose; Kneeling Child’s Pose). Weeks 3–4 (High-load, stability ball): Warm-up (Side Flexion on the Ball; Pelvic Clock on the Ball); Exercise training (Plank on the Ball; Seated Single-Leg Raise on the Ball; Back Extension on the Ball; Superman on the Ball; Seated Arm and Leg Lift on the Ball; Bridge with Arm Swing on the Ball; Kneeling Ball Roll-Out; Prone Single-Leg Extension on the Ball; Reverse Bridge with Twist on the Ball; Pike on the Ball); Cool-down (Supine Trunk Twist; Seated Forward Bend). It follows a “warm-up–training–cool-down” structure, emphasising movement quality and control to safely and systematically improve deep core muscle function, motor control, and overall postural stability in cLBP (Table 1).

Table 1.

Core Stability Exercise

Week	Core Stability Exercise
1–2	Low-load exercises (mat exercises): 1. Pre-Session Warm-up Lateral Bend Stretch; Supine Trunk Twist 2. Exercise Training Plank; Side Plank (Modified); Cat-Cow Stretch; Double-Leg Abdominal Press; Bear Crawl Hold (or Quadruped Position); Glute Bridge with Leg Lift; Dead Bug; Superman; Single-Leg Romanian Deadlift (or Single-Leg Balance); Alternating Superman (Swimming) 3. Post-Training Cool-down Supine Knee-to-Chest Pose; Kneeling Child’s Pose
3–4	High-load exercises (Stability Ball Exercises): 1. Pre-Session Warm-up Side Flexion on the Ball; Pelvic Clock Exercise on the Ball 2. Exercise Training Plank on the Ball; Seated Single-Leg Raise on the Ball; Back Extension on the Ball; Superman Exercise on the Ball; Seated Arm and Leg Lift on the Ball; Bridge with Arm Swing on the Ball; Kneeling Ball Roll-Out; Prone Single-Leg Extension on the Ball; Reverse Bridge with Twist on the Ball; Pike Exercise on the Ball 3. Post-Training Cool-down Supine Trunk Twist; Seated Forward Bend

Study group: The study group will receive acupuncture in addition to the control group’s CSE protocol. As in previous studies,⁴⁶ we will use the classic “Yaosan acupuncture” for LBP, including Shenshu (BL23), Dachangshu (BL25), and Weizhong (BL40). Acupoint localisation: a) Shenshu (BL23), prone, 1.5 cun lateral to the Governor Vessel below the L2 spinous process; b) Dachangshu (BL25), prone, 1.5 cun lateral to the Governor Vessel below the L4 spinous process, approximately aligned with the iliac crest; c) Weizhong (BL40), prone with the knee flexed, midpoint of the popliteal crease between the biceps femoris and semitendinosus tendons. Instruments include an acupoint locator probe, disposable acupuncture needles (0.25 × 50 mm; Shengpeng Medical Equipment Co., China), and 75% alcohol swabs. Patients will be positioned prone to fully expose the lumbodorsal region. Before needling, the therapist will explain the purpose of the treatment, procedures, and possible sensations (eg, mild soreness or distension) to reduce anxiety and facilitate cooperation. A probe will be used to press and locate tender points; after routine skin disinfection and drying, needles will be inserted perpendicularly to a depth of 1.0–1.5 cun. Upon achieving deqi, the acupoint stimulator (LH202H; Beijing Huayun Ante Technology Co., China) will be connected. Electrical stimulation parameters will be a continuous direct‑current waveform at 2 Hz with a 0.5 ms pulse width, and the current will be adjusted to the maximum tolerable level (2–5 mA). One pair of electrodes will be connected to ipsilateral BL23 and BL25, and one pair to bilateral BL40; needles will be removed after 30 minutes of stimulation. Treatment will be 4 sessions per week, 30 minutes each, for 4 consecutive weeks. All acupuncture interventions will be performed by licensed acupuncturists with at least 5 years of clinical experience in Traditional Chinese Medicine acupuncture.⁴⁷ Prior to the trial, all participating acupuncturists will undergo standardised training on the study protocol, including acupoint localisation, needling techniques, and electrical stimulation parameters, to ensure consistency of the intervention across practitioners.

Follow-Up

To evaluate the long-term effectiveness of the interventions, all patients will be followed up 3 months after the interventions end. Follow-up assessments, including the VAS, PCS, FABQ, RMDQ, and PSEQ, will be conducted using online questionnaires by independent assessors blinded to group allocation. Proactive reminders will be used to increase follow-up rates and reduce attrition bias. Data collection and management will follow the study protocol to ensure objective outcome assessment and data integrity.

Treatment Discontinuation

Clear criteria for treatment discontinuation have been established, including temporary interruption and permanent discontinuation. Temporary interruption applies to reversible conditions requiring short-term evaluation and management, such as acupuncture-related adverse events (eg, marked syncope, local haematoma), acute injury after exercise, acute pain exacerbation, and short-term acute illness; treatment may resume after resolution and investigator evaluation. Permanent discontinuation applies in cases of serious adverse events, participant withdrawal of consent, loss to follow-up, or investigator judgment that continuation is unsafe. Investigators must document all discontinuations in detail.

Data Collection

Study data will be collected and managed via a dedicated system across baseline, treatment, and all follow-up time points. Blinded assessors will collect data in a dedicated assessment room; therapists will record each treatment session and adverse events. The study coordinator and data manager will independently double-enter all data within 48 hours of collection. The system will automatically perform logical and range checks and generate queries for any missing, abnormal, or inconsistent data. The final database will be locked under blinded conditions; all data access and modifications will be audit-trailed to ensure integrity, accuracy, and traceability.

Data Analyses

The intention-to-treat (ITT) population will include all randomised participants. For participants who withdraw or are lost to follow-up, missing data will be imputed using the last observation carried forward method and included in the analysis to minimise bias from missing data. The per-protocol (PP) population will exclude participants who withdraw from the study or are lost to follow-up.⁴⁸ Statistical analyses will be conducted separately in the ITT and PP populations to assess the sensitivity and robustness of the results. In the event of discrepant findings, the ITT findings will form the basis of the primary conclusion in the final report, with the PP results reported alongside them.

Statistical analyses will be performed using SPSS version 25.0. Categorical variables will be summarised as frequencies and percentages. Continuous variables will be presented as means and standard deviations if normally distributed, or as medians and interquartile ranges if non-normally distributed. Repeatedly measured primary continuous outcomes will be analysed using mixed-effects models as the primary longitudinal analytical approach, given their repeated measurements over time. The primary outcomes include APA activation timing, APA activation amplitude, CPA activation amplitude, COP sway area, and COP path length. The secondary outcomes include VAS, PCS, FABQ, RMDQ, and PSEQ.

For consistency with the original statistical analysis plan, supplementary analyses will be conducted as prespecified comparisons. For primary outcomes, independent-sample t-tests or Mann–Whitney U-tests will be used for between-group comparisons at each time point, and paired-sample t-tests or Wilcoxon signed-rank tests will be used for within-group changes, depending on the data distribution. For secondary outcomes, repeated-measures ANOVA will be performed when model assumptions are satisfied; if these assumptions are not satisfied, generalised estimating equations will be used. The two-sided significance level will be set at α = 0.05. As no sample size calculation was performed, a post hoc power analysis of the primary outcomes will be conducted descriptively using G*Power 3.1.9.7 based on the observed effect sizes after completion of the analyses.

Patient and Public Involvement

Patients or public representatives were not formally involved in the design or development of this protocol. Eligible participants will be recruited into this randomised controlled trial through outpatient clinics or public advertisements. All enrolled participants will receive all protocol-specified interventions and assessments free of charge. After completing the study, participants may request their individual trial results.

Ethics and Dissemination

This study will adhere to the ethical principles of the Declaration of Helsinki. The study protocol and informed consent were approved by the Ethics Committee of Renhe Hospital, affiliated with China Three Gorges University (Approval number: RHLL-202-507). All potential participants will receive detailed study information and will sign written informed consent after fully understanding it. Study findings will be published in peer-reviewed international journals and presented at relevant academic conferences; a lay summary will also be prepared for complete and transparent dissemination.

Discussion

cLBP is one of the leading causes of disability and socioeconomic burden worldwide.^1,2 Impaired postural control, including altered anticipatory postural adjustments (APAs) and compensatory postural adjustments (CPAs), has been observed in patients with cLBP and may contribute to persistent pain and functional limitations.¹¹ Because current first-line rehabilitation strategies, such as CSE, primarily enhance feedback-mediated CPA regulation, their effects on feedforward APA control may be limited. From the perspective of neurorehabilitation and integrative medicine, acupuncture combined with CSE may therefore represent a promising strategy to modulate feedforward and feedback postural control pathways and facilitate the restoration of overall postural control.

This study will employ a randomised controlled trial design to evaluate the feasibility and preliminary efficacy of a combined intervention comprising acupuncture, CSE, and SCT in patients with cLBP. Within this regimen, acupuncture and CSE are intended primarily to improve postural control, pain, and function, whereas SCT is intended to address pain-related psychological factors, including pain catastrophizing, fear-avoidance beliefs, and pain self-efficacy. In addition, this study will explore potential peripheral neuromuscular mechanisms underlying changes in APA function.

As the psychological component of the intervention, SCT may influence pain-related outcomes, including pain intensity, fear-avoidance beliefs, pain catastrophizing, self-efficacy, and disability. Therefore, future findings should be interpreted with caution as effects observed within a multimodal intervention framework, rather than being attributed solely to acupuncture or CSE. To minimise potential confounding, randomisation, standardised intervention delivery, assessor blinding where feasible, consistent outcome assessment procedures, and predefined ITT and sensitivity analyses will be applied.

This study has several limitations. First, owing to the nature of the interventions, participant and therapist blinding is not feasible, which may introduce performance and expectation bias. Second, as no formal a priori sample size calculation was performed, the sample size may be insufficient to support adequately powered subgroup analyses, thereby limiting investigation of heterogeneity in treatment effects across clinical phenotypes. Third, although the 3-month follow-up period can evaluate short- to mid-term outcomes, it is insufficient to determine the intervention’s impact on the long-term risk of LBP recurrence. Additionally, although the multimodal, synchronised assessment approach enhances objectivity, signal processing and result interpretation involve technical complexity and may be affected by intra-individual variability.

Funding Statement

This research was supported by the Hubei Provincial Administration of Traditional Chinese Medicine Joint Fund (Grant No. ZY2025L134), the China Three Gorges University “Bo Yi” Special Project (Grant No. 2025BYZ06), and the China Three Gorges University Talent Research Start-up Fund (Grant No. 2024RCKJ053).

Disclosure

The authors report no conflicts of interest in this work.