Effectiveness, Safety, and Economic Evaluation of Ultrasound-Guided Acupotomy for Degenerative Lumbar Spinal Stenosis Compared with Conventional Physical Therapy: A Randomized Controlled Trial Protocol

Abstract

Purpose

Acupotomy is an emerging minimally invasive option for patients with degenerative lumbar spinal stenosis (LSS) whose symptoms persist despite conventional conservative care; however, conventional blind procedures raise concerns about procedural safety and target localization. Ultrasound-guided acupotomy (UA) provides real-time visualization of lumbar anatomical structures and may improve procedural precision. This randomized controlled trial (RCT) protocol describes a study designed to compare UA with conventional physical therapy (CPT) in patients with LSS by assessing clinical effectiveness, safety, shear wave elastography (SWE)-assessed biomechanical changes, and cost-effectiveness.

Patients and Methods

This prospective, single-center, assessor-blinded, superiority RCT will enroll 64 patients with magnetic resonance imaging (MRI)-confirmed degenerative LSS. Participants will be randomized 1:1 to receive either UA or CPT, consisting of transcutaneous electrical nerve stimulation and microwave diathermy, twice weekly for 4 weeks. The primary outcome, upon which the sample size estimation is based, is the between-group difference in change from baseline in the Visual Analog Scale (VAS) pain score at Week 5, after 1 week of treatment. Secondary outcomes include VAS scores at Weeks 8 and 12; functional and quality-of-life outcomes assessed using the Zurich Claudication Questionnaire, Oswestry Disability Index, EuroQol-5 Dimensions, and maximum walking distance; exploratory quantitative SWE-assessed multifidus muscle stiffness; and adverse events. Clinical effectiveness will be analyzed using analysis of covariance in the full analysis set. A within-trial cost-utility analysis will also be performed from the healthcare system perspective to estimate the incremental cost-effectiveness ratio.

Conclusion

This protocol builds on a previous pilot study by incorporating assessor blinding, MRI-based diagnostic criteria, SWE-based biomechanical assessment, and cost-utility analysis. By integrating confirmatory clinical outcomes with safety, biomechanical, and economic outcomes, this trial is expected to generate evidence that may support clinical decision-making and inform future economic considerations for UA in the management of LSS.

Introduction

Degenerative lumbar spinal stenosis (LSS) is characterized by narrowing of the spinal canal or intervertebral foramina, causing low back pain, radiating leg pain, and neurogenic claudication.^1,2 Because many patients experience persistent symptoms despite guideline-recommended conservative care,^2,3 acupotomy has gained attention as a minimally invasive alternative.^4–6 Mechanistically, acupotomy is thought to alleviate refractory symptoms by releasing adhered tissues and reducing local fibrosis; recent systematic reviews have also reported favorable clinical effects of acupotomy.^4,7

However, the strength of existing evidence is limited by small sample sizes, high risk of bias, and restricted accessibility of primary studies.^4,7 Furthermore, conventional blind acupotomy raises concerns regarding targeting accuracy and neurovascular safety.^8–10 To address these limitations, ultrasound-guided acupotomy (UA) has been introduced to improve safety and precision via real-time visualization.^11–13 Building on our previous pilot study,¹⁴ which demonstrated the potential clinical utility of UA but was limited by its preliminary design, the current trial incorporates several methodological upgrades. These include magnetic resonance imaging (MRI)-based diagnostic criteria, rigorous assessor blinding, and shear wave elastography (SWE) to quantitatively evaluate biomechanical changes. Additionally, a within-trial cost-utility analysis will be conducted to assess the economic value of UA.

In this randomized controlled trial, we aim to confirm the superiority of UA over conventional physical therapy (CPT) in reducing pain intensity (primary objective). Our secondary objectives include assessing improvements in functional disability and quality of life, monitoring safety, and evaluating within-trial cost-effectiveness. Additionally, quantifying biomechanical changes in multifidus muscle stiffness using SWE will be included as an exploratory objective. The resulting evidence is expected to support informed clinical and economic decision-making for UA in the management of LSS.

Material and Methods

Study Design and Setting

The protocol was developed in accordance with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 statement¹⁵ and the SPIRIT-TCM extension.¹⁶ This study is a prospective, single-center, two-arm, randomized controlled trial (RCT) that will be conducted at Pusan National University Korean Medicine Hospital (PNUKH). The study will include a 12-week study period comprising screening, a 4-week treatment phase, and follow-up assessments at Weeks 5, 8, and 12. The schedule of enrollment, interventions, and assessments is presented in Table 1.

Table 1.

Schedule of Enrollment, Interventions, and Assessments

		Trial Period
		Enrollment	Post-Randomization								Close-Out
			Treatment Phasec								Follow-Up Phased
Timepoint	Visit	Screening	V1b	V2	V3	V4	V5	V6	V7	V8	V9		V10	V11
	Week	−2W	1W		2W		3W		4W		5W		8W	12W
Enrollment
Eligibility screen		O
Informed consent		O
Demographic characteristic		O
Medical history		O
Laboratory test†		O									O
Lumbar spine MRI‡		O
Physical examination		O
ABI, baPWV		O
Randomizationª			O
Interventions
UA			O	O	O	O	O	O	O	O
CPT			O	O	O	O	O	O	O	O
Assessments
VAS		O	O								O		O	O
ODI			O								O		O	O
ZCQ			O								O		O	O
MWD			O								O		O	O
EQ-5D			O								O		O	O
PGIC											O			O
SWE			O								O			O
WPAI			O								O		O	O
Rescue medicine
Dispensing											O		O
Verification													O	O
Adverse events			O	O	O	O	O	O	O	O	O		O	O

Notes: ^† Clinical laboratory tests include complete blood count, prothrombin time, liver function test, renal function test, high-sensitivity C-reactive protein, and glycated hemoglobin (HbA1c). A urine human chorionic gonadotropin test is performed for premenopausal women of childbearing potential (aged ≤50 years). ^‡ Existing magnetic resonance imaging (MRI) data can be submitted if available. If existing MRI data are unavailable or re-imaging is required, a new lumbar spine (L-spine) MRI will be performed. ª Performed prior to the first treatment session. ^b Visit 1 indicates the enrollment visit. Participants are eligible for enrollment at least 2 weeks after the discontinuation of analgesics. ^c During the intervention period, two visits per week are scheduled, with a visit window of +3 days permitted for each week. ^d A visit window of ±3 days is allowed for V9 (5W), and ±7 days is allowed for V10 (8W) and V11 (12W).

Abbreviations: ABI, ankle-brachial index; baPWV, brachial-ankle pulse wave velocity; CBC, complete blood count; CPT, conventional physical therapy; EQ-5D, EuroQol 5-dimension; HbA1c, glycated hemoglobin; hCG, human chorionic gonadotropin; hsCRP, high-sensitivity C-reactive protein; LFT, liver function test; MRI, magnetic resonance imaging; MWD, maximal walking distance; ODI, Oswestry Disability Index; PGIC, Patient Global Impression of Change; PT, prothrombin time; RFT, renal function test; SWE, shear wave elastography; UA, ultrasound-guided acupotomy; V, visit; VAS, visual analog scale; W, week; WPAI, work productivity and activity impairment; ZCQ, Zurich Claudication Questionnaire.

Participants

This trial will enroll 64 participants aged 40–74 years with degenerative LSS. Diagnosis will be confirmed by Korean Medicine doctors (KMDs) based on medical history, physical examination, and MRI findings.¹⁷ Key exclusion criteria include a history of spinal surgery, serious spinal pathologies, such as cauda equina syndrome or tumors, and hemostatic abnormalities. The detailed inclusion and exclusion criteria are presented in Table 2.

Table 2.

Eligibility Criteria for Participant Selection

Inclusion Criteria	Exclusion Criteria
1. Aged 40 years and older but younger than 75 years	1. Individuals diagnosed with congenital or developmental lumbar spinal stenosis
2. Confirmed to have lumbar spinal stenosis on imaging examination (L-spine MRI)	2. Individuals diagnosed with spinal disorders other than lumbar spinal stenosis, which may cause low back, buttock, or leg pain (eg, vertebral tumors, cauda equina syndrome, spinal infections, ankylosing spondylitis)
3. Pain or discomfort in the low back, buttock, or leg ≥ VAS 40mm	3. Individuals with lower extremity arterial stenosis or occlusion that can induce claudication (in cases where ABI < 0.9 or ABI > 1.4)
4. Individuals who have clinical manifestations of lumbar spinal stenosis, such as neurogenic claudication or symptoms that change with posture	4. Individuals with major structural issues of the spine (acute thoracolumbar vertebral fracture, spinal dislocation, spondylolysis)
5. Individuals who have received a final diagnosis of degenerative lumbar spinal stenosis based on comprehensive physical examination, medical history, and imaging examination results	5. Individuals with a history of spinal surgery, such as spinal fusion or laminectomy
6. Individuals who voluntarily decide to participate after being informed of the study purpose and procedure and provide written informed consent	6. Individuals deemed unsuitable for the acupotomy procedure due to abnormal hemostasis, such as those taking anticoagulants or antiplatelet drugs, or those with bleeding disorders
	7. Individuals with acupuncture hypersensitivity, metal allergies, severe atopic dermatitis, keloid skin, or other skin hypersensitivity
	8. Individuals with chronic diseases (eg, chronic kidney disease, diabetic neuropathy, stroke and myocardial infarction, dementia, epilepsy) that may affect treatment outcomes or research results
	9. Individuals taking medications (eg, steroids, immunosuppressants) that may affect treatment outcomes or research results
	10. Individuals with significant neuropsychiatric medical history or current illness
	11. Individuals with a history of alcohol or drug abuse
	12. Pregnant or lactating women or those planning to become pregnant 13. Individuals who are unable to complete patient-reported outcome measures, even with assistance from a caregiver or researcher
	14. Other individuals deemed inappropriate by the clinical researchers

Notes: VAS scores range from 0 mm (no pain) to 100 mm (worst imaginable pain). An ABI value between 0.9 and 1.4 is generally considered normal, while values outside this range indicate peripheral arterial disease or non-compressible arteries.

Abbreviations: ABI, ankle-brachial index; L-spine, lumbar spine; MRI, magnetic resonance imaging; VAS, visual analog scale.

Recruitment

Recruitment is scheduled to begin in November 2025 at PNUKH, with a target enrollment of 64 participants through multiple on-site and off-site recruitment channels. Interested candidates will undergo preliminary telephone screening by a research coordinator to assess basic eligibility. Candidates who appear eligible will then be scheduled for a screening visit. After receiving detailed information about the study and providing written informed consent, each participant will be assigned a unique screening number. Subsequently, they will undergo final eligibility assessments, including confirmation that any ongoing analgesic use has been discontinued for at least 2 weeks before enrollment.

Interventions

Participants in both groups will receive their assigned interventions twice weekly for 4 weeks. Procedural details of UA, including treatment sites, will be documented, and all adverse events will be recorded.

Experimental Group: Ultrasound-Guided Acupotomy

Procedures in the experimental group will be performed as described in Table 3,¹⁸ by a single licensed KMD with extensive clinical experience, with participants in the prone position. A Logiq P9 XDclear ultrasound system (GE Healthcare, Chicago, IL, USA) will be used. To maintain asepsis and prevent cross-infection, the ultrasound probe will be covered with a sterile sheath, and the insertion site will be disinfected with povidone-iodine. Sterile, disposable acupotomy needles (Lejiu acupuncture needle; Maanshan Bond Medical Instrument Co., Ltd., Ma’anshan, Anhui, China) with different specifications (0.60×75 mm, 0.80×80 mm, or 0.50×60 mm) will be selected according to each participant’s physique and the required depth of the target tissue.

Table 3.

Details of the Acupuncture Interventions Based on the STRICTA Guidelines

Item	Details in the Protocol
1. Acupuncture rationale	1a) Style of acupuncture	Ultrasound-Guided Acupotomy based on Korean Medicine and clinical findings
	1b) Reasoning for treatment provided	Based on traditional Korean Medicine and clinical anatomy, this ultrasound-guided treatment targets specific acupoints. It is hypothesized to address spinal stenosis through mechanical adhesiolysis of fibrotic tissues and controlled microtrauma to stimulate physiological wound healing.
	1c) Extent to which treatment was varied	Fixed plus individualized: • Fixed points: Selected corresponding to the affected stenotic levels (deep Hyeopcheok [EX-B2] and/or the Bladder Meridian [BL22-BL26])
		• Individualized points: Selected based on clinical findings (painful sites, symptomatic areas, palpable abnormalities, tender points, etc.) (eg, BL52, BL27, GV6, GB30, Ashi points)
2. Details of needling	2a) Number of needle insertions per subject per session	Clinically determined based on the affected stenotic levels
	2b) Names of points used	Fixed: Deep Hyeopcheok (EX-B2) and/or the Bladder Meridian (BL22-BL26), targeting the facet joints, the junction of the superior articular and transverse processes, and the para-nerve root area corresponding to the affected stenotic levels.
		Individualized: Selected based on clinical findings, including BL52 (quadratus lumborum/L3 transverse process), BL27 (sacroiliac joint), GV6 (thoracolumbar junction), GB30 (gluteal/piriformis muscles), and Ashi points.
	2c) Depth of insertion	Determined based on the target depth and patient’s physique (identified via ultrasound)
	2d) Response sought	Sensation of tissue release or dissection of adhesions
	2e) Needle stimulation	Manual multidirectional adhesiolysis, including vertical (pecking motion) and fan-shaped techniques. The cutting edge of the needle is oriented both longitudinally and transversely to release fibrotic adhesions.
	2f) Needle retention time	Not retained (immediate withdrawal following manipulation)
	2g) Needle type	Sterile, disposable acupotomy needles (Brand: Lejiu acupuncture needle; Maanshan Bond Medical Instrument Co., Ltd., Ma’anshan, Anhui, China)
		Sizes: 0.60×75 mm, 0.80×80 mm, or 0.50×60 mm
3. Treatment regimen	3a) Number of treatment sessions	8 sessions (twice a week for 4 weeks)
	3b) Frequency and duration of treatment sessions	Frequency: Twice a week
		Duration: Approximately 5–12 min per session, depending on the extent and severity of the lesion
4. Other components of treatment	4a) Details of other interventions	• Ultrasound guidance: Logiq P9 XDclear (GE Healthcare, Chicago, IL, USA)
		• Ultrasound approach: Primarily in-plane approach under transverse scanning; complementarily out-of-plane under longitudinal scanning.
		• Safety/Hygiene: Povidone-iodine sterilization, use of sterile probe covers, and strict adherence to site-specific safety protocols to prevent adverse events.
		• Post-procedure: Manual compression, sterile bandages, and ice packs to minimize pain and inflammation, education on post-procedural care
	4b) Setting and context of treatment	Pusan National University Korean Medicine Hospital
5. Practitioner background	Description of participating acupuncturists	A single licensed KMD with over 10 years of clinical experience in acupotomy
6. Control or comparator interventions	6a) Rationale for the control or comparator	Active comparator representing conservative care in real-world clinical practice for LSS to ensure practical clinical relevance and external validity
	6b) Description of the control or comparator	1. TENS: TM-6002 (SHINWOO MEDILAND Co., Ltd., Wonju, Gangwon-do, Republic of Korea); 20 Hz, 15 mins; Intensity above sensory/below motor threshold
		2. Microwave Diathermy (MW): ME-8250 (OG GIKEN Co., Ltd., Okayama, Okayama Prefecture, Japan); 120 W, 10 mins; 10 cm distance

Notes: Acupoint nomenclature is based on the World Health Organization (WHO) Standard Acupuncture Nomenclature.

Abbreviations: BL, Bladder Meridian; EX-B2, Extra Back Point 2 (Hyeopcheok); GB, Gallbladder Meridian; GV, Governor Vessel; KMD, Korean Medicine doctor; MW, Microwave Diathermy; STRICTA, Standards for Reporting Interventions in Clinical Trials of Acupuncture; TENS, Transcutaneous Electrical Nerve Stimulation.

Vertebral levels will be identified using longitudinal and transverse sonographic scans. The intervention will primarily be performed using an in-plane approach under transverse guidance. For procedures targeting the transverse process and the para-nerve root area, an out-of-plane approach during longitudinal scanning may be used as a complementary technique. Treatment targets will be classified into primary fixed points and adjunctive individualized points based on radiological and clinical assessments. Fixed points will correspond to the affected stenotic levels and will include the facet joints (including deep Hyeopcheok, EX-B2), the junction of the superior articular and transverse processes, the para-nerve root area, and the surrounding fibrotic tissues (Figure 1A). Adjunctive points will be selected based on symptomatic regions and tender points and will include the lumbar multifidus, BL52 (quadratus lumborum or L3 transverse process), BL27 (sacroiliac joint), GV6 (thoracolumbar junction), GB30 (gluteal/piriformis muscles), and relevant Ashi points.

Figure 1.

Ultrasound-guided acupotomy approach and multidirectional release technique. (A) In-plane approach under transverse sonographic guidance. The circle indicates the targeted facet joint, the triangle indicates the junction of the superior articular and transverse processes, and the square indicates fibrotic tissue. (B) Multidirectional release techniques can be applied in both longitudinal and transverse blade orientations using the in-plane approach. The numbers on the right margin of the ultrasound images indicate depth in centimeters. The images illustrate the procedure performed at the L4–L5 level on the right side.

Abbreviations: FJ, facet joint; SAP, superior articular process; SP, spinous process; TP, transverse process.

During the procedure, the practitioner will apply manual techniques, including vertical adhesiolysis, performed with a pecking motion, and fan-shaped adhesiolysis, to release fibrotic adhesions. To achieve multidirectional tissue release, the cutting edge of the needle will be oriented both longitudinally and transversely (Figure 1B). Site-specific safety protocols will be strictly followed to prevent adverse events. For instance, when targeting the junction of the superior articular and transverse processes, the blade will be aligned longitudinally along the anatomical course of the medial branch. Intentional bone contact will be made gently, and excessive subsequent manual stimulation will be avoided to minimize neural or periosteal irritation. When approaching the para-nerve root area, the needle will be advanced cautiously. If the participant reports radicular pain or paresthesia, mechanical stimulation will cease immediately, and the trajectory will be slightly redirected under continuous ultrasound guidance to ensure that adhesiolysis is performed in a safe zone.

After needle withdrawal, the treatment site will be disinfected, and manual compression with sterile bandages will be applied. Ice packs will be provided to reduce post-procedural discomfort and localized inflammation, and participants will receive relevant post-care instructions.

Control Group: Conventional Physical Therapy

Participants in the control group will receive conventional physical therapy (CPT), consisting of transcutaneous electrical nerve stimulation (TENS) and microwave diathermy (MW), administered by KMDs. Treatment sites will be selected based on radiological and clinical findings. For TENS, a TM-6002 device (SHINWOO MEDILAND Co., Ltd., Wonju, Gangwon-do, Korea) will be used. With participants in the prone position, moistened pads will be placed bilaterally over the lumbar treatment levels. Low-frequency stimulation (20 Hz) will be delivered for 15 min, with the intensity adjusted to a level above the sensory threshold but below the motor threshold to maintain participant comfort. MW will then be administered using an ME-8250 device (OG GIKEN Co., Ltd., Okayama, Okayama Prefecture, Japan). After all metal objects are removed from the treatment area, the heat source will be positioned approximately 10 cm from the skin. The therapy will be applied for 10 min at an output intensity of 120 W, with the distance adjusted based on participant feedback to prevent burns. TENS and MW will not be applied concurrently.

Concomitant Treatments

To preserve the internal validity of the study, concomitant treatments for LSS, including invasive conventional medical procedures, such as injections, endoscopic procedures, and surgery, as well as other Korean medicine treatments, will not be permitted during the study period. However, treatment for unrelated conditions may be allowed at the discretion of the investigators. Medications for preexisting conditions other than LSS that are judged unlikely to affect interpretation of the study results will be permitted and recorded. During the intervention period, participants must receive only their assigned study intervention. To minimize potential confounding of the primary endpoint at Week 5, rescue medication (acetaminophen, up to 3000 mg/day) will be permitted during the follow-up period only for participants experiencing intolerable pain.

Rescue medication will be provided at the first post-treatment assessment visit (V9, 1 week after the intervention) and the second post-treatment assessment visit (V10, 5 weeks after the intervention). Intake will be verified at each subsequent visit by reviewing participants’ medication diaries. Participants will be required to record all rescue medication use in a medication log, and investigators will document total consumption.

Outcome Measurements

Outcomes will be assessed at baseline and at 1, 4, and 8 weeks after the intervention (Weeks 5, 8, and 12, respectively).

Primary Clinical Outcome

The primary clinical outcome will be the mean change from baseline to Week 5 in the Visual Analog Scale (VAS) score, measured on a 0–100-mm scale, for pain or discomfort in the low back, buttock, or leg. Week 5 corresponds to 1 week after the intervention. This specific timepoint was selected based on clinical judgment to ensure the resolution of transient post-procedural soreness while minimizing the influence of unmeasured confounding variables associated with an extended post-intervention period, thereby providing a reliable assessment of the immediate therapeutic effect. The VAS is a 100-mm horizontal line anchored by “no pain” (0) and “worst pain imaginable” (100). Participants will be instructed to rate their current symptom intensity on this scale.

Secondary Clinical Outcome

Secondary outcomes will include mean changes in VAS scores at Weeks 8 and 12. Functional status, quality of life, and quantitative biomechanical changes will also be evaluated using the following measures:

Patient-Reported Outcome Measures

The Zurich Claudication Questionnaire (ZCQ), including the Symptom Severity and Physical Function domains,¹⁹ Oswestry Disability Index (ODI),²⁰ EuroQol-5 Dimensions (EQ-5D),²¹ Patient Global Impression of Change (PGIC),²² and Work Productivity and Activity Impairment (WPAI) questionnaire²³ will be administered. Participants will complete these forms independently, with minimal assistance from the investigator provided only when needed to clarify individual items.

Maximum Walking Distance (MWD)

Self-reported MWD is defined as the estimated maximum distance, in meters, that a participant can walk continuously without resting. It will be estimated by multiplying the reported walking time, in minutes, by the participant’s average walking speed, using established age-specific reference values.²⁴ This constant age-specific reference speed will be applied to assess relative changes in walking capacity over time.

Multifidus Muscle Stiffness (SWE)

As an exploratory endpoint, lumbar multifidus stiffness will be quantified using SWE (Logiq P9 XDclear). Measurements will be performed at the stenotic level identified during screening. If multiple levels are involved, the investigators will select the primary stenotic level based on imaging and clinical findings. To standardize baseline muscle tension, participants will first rest in a seated position with both feet flat on the floor and legs uncrossed for 5 min. Before measurement, participants will be placed in the prone position. To standardize lumbar positioning and minimize variation in lumbar lordosis and multifidus relaxation, the same abdominal pillow will be centered under the umbilicus, and the bed’s headrest will be tilted downward by approximately 10 degrees.^25,26 Bilateral measurements will then be obtained at the designated stenosis level. To minimize anisotropy artifacts and optimize image quality, the transducer will be slightly tilted and aligned along the muscle.^25–27 Furthermore, to standardize probe compression, a generous amount of ultrasound coupling gel will be applied. The examiner will lightly rest the transducer on the gel layer, ensuring no visible skin deformation on the B-mode image, thereby minimizing transducer-induced tissue compression to prevent artificial stiffening of the tissue.^26–28 Following a predefined protocol to enhance intra-rater reliability and exclude outliers, five independent measurements will be performed per side,^25,29 and the mean of the middle three values, excluding the maximum and minimum, will be used for final analysis.

Clinically Important Difference (CID)

CID will be determined by calculating the proportion of participants who achieve an absolute reduction in VAS score of ≥ 15 mm or relative reductions of ≥ 30% and ≥ 50% from baseline.^22,30

Rescue Medication

Usage details, including use status, start date, duration, and total dose, will be recorded in participants’ medication diaries and verified by investigators.

Safety Outcomes

Adverse events (AEs) are defined as any untoward medical occurrence, including unfavorable signs, symptoms, or disease, regardless of causality. Intervention-related discomfort will be recorded as an AE if it persists for >24 h. At each visit, investigators will assess, document, and verify AEs, including severity based on Spilker’s three-point scale³¹ and causal relationship based on a six-point scale.³² All AEs will be monitored until they resolve or stabilize, and their duration and outcome will be recorded. Any serious adverse events will be reported immediately to the Institutional Review Board (IRB), and appropriate medical management will be provided without delay.

Economic Evaluation

The within-trial economic evaluation will be conducted in accordance with the Korean Guidelines for Pharmacoeconomic Evaluations.³³ To assess cost-utility, direct medical costs, including procedure-related expenses and rescue medication use, and health utility values derived from the EQ-5D using a Korean-specific tariff²¹ will be collected. Productivity loss will also be measured using the WPAI questionnaire, which assesses absenteeism, presenteeism, overall work impairment, and activity impairment, and will be reported separately.

Sample Size Calculation

The sample size estimate was based on a previous pilot study¹⁴ (n = 50). The pilot data showed a mean (standard deviation [SD]) change in VAS scores 1 week after treatment of 25.0 (17.9) in the experimental group and 6.6 (12.4) in the control group, corresponding to a between-group mean difference of 19.4 (15.4). To derive a conservative estimate, the sample size calculation considered both the pilot study findings and the minimal clinically important difference (MCID),^22,30 while accounting for potential variability in a larger sample. Consequently, a mean difference of 15 and an SD of 20 were used for the calculation. Assuming a two-sided significance level (α) of 0.05, 80% power, and a 1:1 allocation ratio, 28 participants per group were required to detect a significant between-group difference. After allowing for an anticipated attrition rate of 10%, the study will recruit 64 participants, with 32 participants per arm.

Randomization and Blinding

Randomization and Allocation Concealment

An independent statistician not involved in trial conduct will use SAS® Version 9.4 (SAS Institute Inc., Cary, NC, USA) to generate a simple randomization schedule. The schedule will allocate participants to the experimental or control group in a 1:1 ratio with equal allocation probability. The randomization file will remain in the exclusive possession of the statistician to maintain confidentiality. Allocation codes will be placed in opaque, sealed envelopes and stored in a secure, locked cabinet. After eligibility is confirmed, the investigator will open the assigned envelope at Visit 1 only in the presence of the participant. Once the envelope is opened, a randomization number will be assigned to the participant. The allocation date and the investigator’s signature will be recorded on the envelope, which will then be stored securely.

Blinding

Because of the inherent differences between the interventions, practitioners and participants cannot be blinded to group allocation. To minimize detection bias, clinical outcomes will be evaluated by independent assessors who are blinded to group allocation and uninvolved in treatment procedures. These blinded assessors, clinical research coordinators, will provide minimal assistance only when needed to clarify questionnaire items, without influencing participants’ responses. SWE measurements will be performed by independent KMDs who are not involved in treatment procedures or questionnaire assessments.

Withdrawal and Dropout

Participants may withdraw from the study at any time for any reason, including symptom resolution, symptom worsening, or dissatisfaction with treatment. The investigator may discontinue a participant because of safety concerns, such as serious adverse events, pregnancy, or symptom worsening requiring surgery, or because of protocol violations or loss to follow-up. All reasons for withdrawal will be documented, and any adverse events leading to withdrawal will be monitored until resolution.

Data Collection and Monitoring

Baseline data collected at screening and Visit 1 will include demographic information, medical history, concomitant medications, clinical symptoms, and vascular status. Lumbar spine MRI will be performed to confirm the diagnosis and grade stenosis severity. In addition, the degree of fatty degeneration in the multifidus muscles will be qualitatively evaluated using MRI to assess baseline structural characteristics.³⁴ All data, including detailed procedural records to monitor treatment fidelity across sessions, will be recorded in a secure electronic case report form (eCRF) system. Regular monitoring will be conducted by an independent clinical research associate (CRA) to ensure data accuracy and treatment fidelity through source data verification. Monitoring activities will include site initiation, interim, and close-out visits, with the schedule adjusted according to participant recruitment progress.

Statistical Methods

General Statistical Analysis

All analyses will be performed by an independent statistician using SAS Version 9.4, with a two-sided significance level of p < 0.05. Missing data will be handled using multiple imputation (MI) under the missing-at-random assumption.

The Full Analysis Set (FAS) will include all randomized participants who receive at least one intervention. This analysis set will follow the intention-to-treat (ITT) principle with minimal exclusions and will be used to evaluate the effectiveness of assigned interventions regardless of participant adherence. The FAS will serve as the primary analysis set for both clinical and economic evaluations. The Per-Protocol (PP) set will include participants who attend at least 80% of the planned treatment sessions (at least seven sessions of UA or CPT), complete the outcome assessments, and have no major protocol violations. This set will be used for supportive clinical analyses. Safety analyses will be conducted in the Safety Set, defined as participants from the FAS who receive at least one intervention and have safety data verified at least once during study visits.

Demographic and Baseline Characteristics

Baseline characteristics will be summarized by group. Continuous variables will be compared using independent t-tests or Wilcoxon rank-sum tests, and categorical variables will be compared using chi-squared or Fisher’s exact tests.

Effectiveness and Safety Analysis

Effectiveness analyses will be conducted primarily in the FAS, with supportive analyses performed in the PP set. Safety data will be analyzed using the Safety Set. The primary outcome, the mean change in VAS score from baseline, will be analyzed using analysis of covariance (ANCOVA), with the baseline VAS score as a covariate and the treatment group as a fixed factor. For the FAS analysis, missing values will be handled using MI. If the assumptions of normality and homogeneity of variance are violated, rank-transform ANCOVA will be applied as a nonparametric alternative. The same statistical approach will be applied to secondary continuous outcomes, including ZCQ, ODI, MWD, EQ-5D, and SWE. Categorical outcomes, including PGIC and CID, will be compared using chi-squared tests. Correlation analyses will be conducted to explore associations between SWE-derived biomechanical changes and clinical improvements. Subgroup analyses will be performed based on sex, age, brachial-ankle pulse wave velocity, and stenosis severity, where appropriate. Safety data will be summarized descriptively.

Economic Analysis

As a prespecified secondary analysis, the base-case within-trial economic analysis will be performed in the FAS from a healthcare system perspective. The incremental cost-effectiveness ratio (ICER) and net health benefit (NHB) will be calculated using direct medical costs and quality-adjusted life years estimated using the area-under-the-curve method. No discounting will be applied because of the 12-week time horizon. Consistent with the healthcare system perspective, WPAI-derived productivity loss will be excluded from the base-case ICER and reported separately. Ultrasound guidance costs will be estimated using fees for comparable conventional procedures, with a scenario analysis that considers material costs only. Finally, a one-way sensitivity analysis will be conducted to explore uncertainty around key assumptions. Nonparametric bootstrapping will be performed to derive 95% confidence intervals for the ICER and construct a cost-effectiveness acceptability curve.

Dissemination

The final trial results will be disseminated through publication in peer-reviewed journals and conference presentations.

Discussion

This confirmatory RCT is designed to evaluate the clinical effectiveness, safety, and cost-effectiveness of UA in patients with degenerative LSS. To enhance external validity and better reflect routine clinical practice, CPT was selected as the active control. This protocol builds on the previous pilot study¹⁴ by incorporating MRI-based LSS grading, quantitative assessment of tissue elasticity, and a within-trial economic evaluation.

Degenerative changes in the lumbar spine, including facet joint hypertrophy and thickening of the ligamentum flavum, cause progressive narrowing of the spinal canal and lateral recesses.^35–37 These structural alterations place persistent mechanical stress on the facet joint capsule and periarticular tissues, leading to chronic local inflammation and subsequent fibrosis and adhesion of surrounding soft tissues.^38–40 This pathological microenvironment contributes to sensitization of the medial branch, which innervates the multifidus muscle,³⁸ and is associated with reflexive multifidus atrophy and fatty degeneration.^41–45 These changes may ultimately contribute to altered segmental mechanics and potential biomechanical instability of the spine.^46–48 A similar self-perpetuating cycle of structural degeneration, neurogenic inflammation, and neuromuscular dysfunction is observed in conditions involving altered segmental mechanics, such as adjacent segment disease after spinal fusion.^41–43,49 In such cases, compensatory hypermobility and increased shear forces at adjacent levels accelerate degenerative changes.⁴⁹ Although the specific biomechanical manifestations may differ across spinal disorders, similar alterations in segmental kinematics and load distribution may contribute to the pathological cycle in LSS, potentially increasing mechanical stress on the facet joints and surrounding soft tissues.^{36,42,46–48} Therefore, we hypothesize that interrupting this self-perpetuating cycle of neurogenic inflammation and muscle degeneration requires an intervention that effectively releases fibrotic adhesions and reduces mechanical stress on hypertrophied facet joints and periarticular soft tissues. While this mechanistic rationale forms the basis for this trial, these pathways largely remain theoretical in the context of LSS.

In current clinical practice, conventional non-surgical interventions for LSS, including epidural steroid injections, selective nerve root blocks, medial branch blocks, and hydrodissection, primarily rely on pharmacological modulation or fluid-mediated tissue separation. In contrast, acupotomy provides direct mechanical adhesiolysis of pathological soft tissues. Previous biomechanical studies have shown that compensatory flexion in LSS induces chronic tension and fatty degeneration of the multifidus muscle.^41–45 Chronic mechanical loading also contributes to thickening and adhesion of the deep fascia, which may alter normal segmental motion and load distribution, thereby increasing facet joint stress and medial branch irritation.^38–40 By mechanically targeting these pathological changes, acupotomy is thought to directly address the underlying biomechanical dysfunction. In addition, the controlled microtrauma intentionally induced by acupotomy is hypothesized to initiate a physiological wound-healing cascade, potentially promoting tissue remodeling and the restoration of microcirculation.^50–54 This potential dual mechanism, mechanical release and biological regeneration, is hypothesized to distinguish acupotomy from conventional interventions that often provide short-term symptom relief.

The primary justification for adopting ultrasound guidance is to address the inherent limitations of conventional blind procedures. Because deep lumbar structures are difficult to access accurately using surface landmarks alone, ultrasound enables real-time visualization of key extraspinal targets, including the facet joints, the junction of the superior articular and transverse processes, the para-nerve root area, and the multifidus muscle with adjacent fibrotic tissues.^9,10,13 However, acoustic shadowing from bony structures precludes real-time visualization of the spinal canal, making direct intraspinal decompression via UA technically unfeasible. To address this anatomical limitation, our protocol adopts a comprehensive extraspinal targeting framework designed to address facet-related mechanical dysfunction and potential segmental instability, relieve extraspinal fascial tension, and resolve myofascial adhesions. While this strategy provides a unified therapeutic repertoire across different LSS subtypes, specific release targets are tailored to each patient’s stenotic pattern identified on baseline MRI and subsequently localized under real-time ultrasound guidance. Ultimately, this individualized, multi-target approach may facilitate precise mechanical stimulation, reduce operator-dependent variability and minimize the risk of iatrogenic injury near neural structures.^9–13

Building on these design features, the trial further reinforces internal validity through multiple methodological strategies. Diagnostic rigor is strengthened by MRI-based inclusion criteria¹⁷ and by the quantitative assessment of stenosis severity and multifidus fatty degeneration.³⁴ Preplanned subgroup analyses will allow treatment effects to be evaluated according to structural pathology. To minimize detection bias in the context of an intervention that cannot be blinded to practitioners or participants, outcome assessors will remain blinded to group allocation. A rescue medication protocol is also incorporated to address ethical considerations while providing an additional outcome measure. The maximum daily dose of 3,000 mg is set below the standard 4,000 mg threshold to reduce the risk of hepatotoxicity, in accordance with safety guidelines.^55,56

Outcome measures were broadened beyond subjective symptom assessment to include an exploratory quantitative biomechanical evaluation. SWE will be used to assess multifidus muscle stiffness, thereby providing a quantitative measure of intervention-related biomechanical change.^25–29 Because paraspinal muscle properties in lumbar disorders vary according to underlying structural and clinical characteristics, biomechanical responses to treatment may differ across patient subgroups. Therefore, rather than prespecifying a single expected pattern of SWE change, this study will explore SWE outcomes through preplanned subgroup analyses incorporating multiple baseline factors, including MRI-based fatty degeneration grade, stenosis severity, age, symptom duration, and sex. This approach is intended to provide a more nuanced interpretation of biomechanical outcomes in a clinically and structurally heterogeneous LSS population. Active control interventions may induce short-term changes in muscle stiffness primarily through thermal and neuromodulatory mechanisms. In contrast, UA, which may exert bidirectional modulatory effects, is hypothesized to provide direct mechanical adhesiolysis of chronic deep fascial adhesions and promote structural tissue remodeling through controlled microtrauma. This mechanism may offer a distinct pathway for inducing biomechanical changes. Therefore, the patterns of elasticity change measured by SWE, including the direction and magnitude of change, may differ between the two groups and reflect these mechanistic differences. Furthermore, exploratory longitudinal and between-group analyses may help clarify differential biomechanical responses among subgroups defined by baseline structural pathology.

In addition to clinical and biomechanical outcomes, a within-trial cost-utility analysis will be conducted to evaluate the economic value of UA within the context of this study. From a healthcare system perspective, this trial is designed to generate evidence on the cost-effectiveness of UA compared with conventional physical therapy. By linking short-term changes in health-related quality of life with accumulated costs during the study period, the analysis will assess whether the anticipated clinical benefits of UA are accompanied by meaningful economic value. Alongside conventional cost-effectiveness metrics, including the ICER, the inclusion of NHB and uncertainty analyses will support policy-relevant interpretation across different willingness-to-pay thresholds. Although productivity losses will not be included in the base-case analysis, their separate reporting will provide complementary insight into the broader societal impact of the intervention. However, the assessment and interpretation of this economic evaluation are confined to the within-trial setting. While these findings may inform economic decision-making, they will not be generalized or serve as definitive evidence.

Despite these methodological strengths, several limitations exist. First, the assessor-blinded design without a sham control makes subjective outcomes susceptible to expectation bias and precludes differentiating specific therapeutic from non-specific effects. Second, estimating walking capacity via derived metrics rather than direct tests lacks precision and may overestimate distances. Third, the 12-week follow-up is insufficient to evaluate the long-term durability of the intervention. Fourth, as the trial is powered solely for the primary outcome, the economic evaluation is confined to a within-trial analysis, and SWE assessments (including their correlation with clinical improvement) remain exploratory; thus, caution is warranted when generalizing these findings. Finally, the single-center design and operator dependence may further limit generalizability.

Conclusion

In conclusion, by incorporating MRI-based diagnostics, quantitative biomechanical evaluation using SWE, and a within-trial cost-utility analysis, this study is designed to evaluate the clinical effectiveness and safety of UA compared with CPT for degenerative LSS. Furthermore, this trial will assess biomechanical changes using SWE on an exploratory basis and evaluate the economic value of the intervention through a cost-utility analysis. The findings are expected to inform clinical decision-making and support future economic considerations.

Funding Statement

This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: RS-2023-KH139597).

Abbreviations

AE, adverse event; ANCOVA, analysis of covariance; CID, clinically important difference; CPT, conventional physical therapy; CRA, clinical research associate; eCRF, electronic case report form; EQ-5D, EuroQol-5 Dimensions; FAS, full analysis set; ICER, incremental cost-effectiveness ratio; IRB, Institutional Review Board; ITT, intention-to-treat; KHIDI, Korea Health Industry Development Institute; KMD, Korean Medicine doctor; LSS, lumbar spinal stenosis; MCID, minimal clinically important difference; MI, multiple imputation; MRI, magnetic resonance imaging; MW, microwave diathermy; MWD, maximum walking distance; NHB, net health benefit; ODI, Oswestry Disability Index; PGIC, Patient Global Impression of Change; PNUKH, Pusan National University Korean Medicine Hospital; PP, per-protocol; RCT, randomized controlled trial; SD, standard deviation; SPIRIT, Standard Protocol Items: Recommendations for Interventional Trials; STRICTA, Standards for Reporting Interventions in Clinical Trials of Acupuncture; SWE, shear wave elastography; TCM, traditional Chinese medicine; TENS, transcutaneous electrical nerve stimulation; UA, ultrasound-guided acupotomy; VAS, Visual Analog Scale; WPAI, Work Productivity and Activity Impairment; ZCQ, Zurich Claudication Questionnaire.

Data Sharing Statement

No datasets were generated or analyzed for the current study protocol.

Ethics Approval and Informed Consent

The study protocol was approved by the PNUKH Institutional Review Board (PNUKHIRB2025-10-004) and was prospectively registered with the Clinical Research Information Service (KCT0011163) before participant recruitment. All procedures will be conducted in accordance with the ethical principles outlined in the Declaration of Helsinki.⁵⁷ Written informed consent will be obtained after participants receive detailed study information through an information sheet covering the purpose, duration, methods, risks, benefits, and right to withdraw, with sufficient time to consider participation. For potentially vulnerable participants, including hospital staff and students, a handwritten statement confirming voluntary participation will also be required to ensure that participation is voluntary. To maintain confidentiality, participants will be identified using unique identification codes, and all personal information will be de-identified throughout the study and dissemination process.

Consent for Publication

Not applicable. This manuscript does not contain any identifiable individual data, images, videos, or recordings.

Disclosure

The authors report no conflicts of interest in this work.