Low-Frequency Electroacupuncture versus Sham Electroacupuncture for Diabetic Peripheral Neuropathy: Study Protocol for a Randomized Controlled Trial

Abstract

Background

Diabetic peripheral neuropathy (DPN) is a prevalent and disabling complication, with over 50% of diabetic patients affected. Current pharmacologic therapies are limited by adverse effects, highlighting the need for safer alternatives. Preliminary evidence suggests that low-frequency (2 Hz) electroacupuncture (EA) may be beneficial, but high-quality clinical trials investigating its mechanisms are lacking.

Methods and Analysis

This prospective, randomized, sham-controlled trial will enroll 92 patients with DPN, who will be randomly allocated in a 1:1 ratio to receive 6 weeks of treatment with either active 2 Hz EA or sham EA. The primary outcomes are overall response rate and lower-limb nerve conduction velocity (NCV). Secondary outcomes include pain intensity assessed using the Visual Analog Scale (VAS), quality of life scores, and serum biomarkers measured by enzyme-linked immunosorbent assay (ELISA), namely nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), tumor necrosis factor-alpha (TNF-α), and interleukin-6 (IL-6). All statistical analyses will be conducted according to the intention-to-treat (ITT) principle.

Conclusion

This study is expected to provide clinical and mechanistic evidence for the potential role of standardized low-frequency EA as a non-pharmacological treatment option for DPN.

Trial Registration

ClinicalTrials.gov Identifier: NCT07551726.

Introduction

Diabetic peripheral neuropathy (DPN) is one of the most common chronic neurological complications of diabetes mellitus (DM), affecting more than 50% of patients with DM during the disease course.^1,2 Given the large population of patients with DM in China,³ DPN represents an important clinical and public health concern. Clinically, DPN often presents with symmetrical sensory symptoms, including pain, numbness, burning sensation, hyperalgesia, or sensory loss, usually in a “glove-and-stocking” distribution.⁴ With disease progression, patients may develop motor dysfunction, gait disturbance, foot ulcers, infection, and even amputation.^5,6 These complications can substantially impair quality of life. Therefore, effective and safe interventions for improving neuropathic symptoms and nerve function in patients with DPN are still needed.

The pathogenesis of DPN is complex and is closely related to chronic hyperglycemia. Sustained hyperglycemia may disturb metabolic homeostasis, increase oxidative stress, and promote the accumulation of advanced glycation end products.^7,8 These pathological changes can activate inflammatory responses and contribute to demyelination and axonal injury. In addition, impaired neurotrophic support may reduce the capacity for peripheral nerve repair. For example, reduced expression of neurotrophic factors, including nerve growth factor (NGF), has been associated with neural injury under diabetic conditions.⁹ These findings suggest that inflammation and neurotrophic regulation may be relevant to the development and recovery of DPN.

Current management of DPN mainly includes glycemic control, neurotrophic agents, antioxidant therapy, and medications for neuropathic pain. Tricyclic antidepressants, serotonin-norepinephrine reuptake inhibitors, and antiepileptic drugs such as gabapentin are commonly used for painful DPN.^10–12 Although these treatments may relieve symptoms in some patients, their long-term use can be limited by adverse effects such as dizziness, drowsiness, blurred vision, gastrointestinal discomfort, and cardiovascular concerns.^13–16 Therefore, safer and more acceptable non-pharmacological treatment options remain clinically relevant.

Acupuncture has been used in the management of chronic pain, nerve injury, neuropathy, and neuropathic pain.^17,18 Electroacupuncture (EA), which combines acupuncture with electrical stimulation, allows relatively standardized stimulation parameters and may be useful for conditions involving neuropathic symptoms. Previous studies have suggested that EA may relieve DPN-related pain, improve nerve conduction velocity (NCV), and regulate biological processes related to microcirculation, inflammation, and nerve repair.^19–22 However, existing studies vary in acupoint selection, stimulation parameters, and outcome measures, which limits the interpretation and application of current evidence. Frequency is a key parameter, as low- and high-frequency EA act through different mechanisms: 2 Hz preferentially promotes release of enkephalin, β-endorphin, and endomorphin, whereas 100 Hz mainly triggers dynorphin, and low-frequency EA appears more effective against neuropathic hyperalgesia.^23,24 Consistent with this, our previous work showed that although both 2 Hz and 100 Hz EA reduced P2X3 receptor expression in dorsal root ganglia and alleviated DPN-related symptoms,²⁵ 2 Hz showed potential advantages in relieving diabetic neuropathic pain and regulating hyperalgesia-related proteins.²⁶ However, given the physiological differences between animal models and humans, these preclinical findings require confirmation in a clinical setting before they can guide practice. This provides the rationale for the present trial, in which 2 Hz EA will be prospectively evaluated in patients with DPN. Based on these findings, this trial will use low-frequency EA at 2 Hz.

The potential mechanisms of EA in DPN also deserve further investigation. Previous studies have reported that EA may regulate neurotrophic factors, including NGF and brain-derived neurotrophic factor (BDNF), and may reduce inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6).^27–30 NGF and BDNF are involved in neuronal survival, axonal regeneration, and peripheral nerve repair, whereas TNF-α and IL-6 are closely related to inflammatory and neuroinflammatory processes that contribute to neuropathic pain, hyperalgesia, and impaired nerve conduction in DPN.^31,32 These findings provide a rationale for assessing both neurotrophic and inflammatory markers in clinical studies of EA for DPN. However, clinical trials that simultaneously evaluate symptoms, electrophysiological outcomes, and serum biomarkers before and after EA treatment remain limited. Therefore, measuring NGF, BDNF, TNF-α, and IL-6 may help explore whether changes in peripheral inflammatory and neurotrophic responses are associated with clinical and electrophysiological improvement.

This prospective randomized sham-controlled trial aims to evaluate the clinical effect and safety of low-frequency EA in patients with DPN. The study will assess clinical symptoms, NCV, quality of life, and serum levels of NGF, BDNF, TNF-α, and IL-6. We hypothesize that, compared with sham EA, low-frequency EA will yield a higher overall response rate and greater improvement in lower-limb nerve conduction velocity (primary hypotheses), and will also improve quality of life while favorably modulating serum NGF, BDNF, TNF-α, and IL-6 (secondary hypotheses). By combining clinical, electrophysiological, and biomarker outcomes, this trial may provide preliminary evidence for the potential role of standardized low-frequency EA as a non-pharmacological treatment option for DPN.

Methods

Study Design

This study employs a single-blind, randomized, placebo-controlled, parallel-group design. Participants who meet the eligibility criteria will be randomly allocated in a 1:1 ratio to either the active low-frequency EA group or the sham EA control group. The overall trial workflow, from screening to final follow-up, is depicted in the CONSORT-style flowchart (Figure 1 and Table 1). The complete schedule of enrollment, interventions, and assessments is presented in accordance with the SPIRIT guidelines.³³

Figure 1.

Flow chart of the study process.

Table 1.

Schedule of Enrolment, Interventions and Outcome Assessments

	Baseline	Treatment Phase (6 Weeks)		Follow-Up (4 Weeks)
Time point	Week 0	Week 3 ± 3days	Week 6 ± 3days	Week 10 ± 3days
Enrollment
Demographic information	×
Medical history and treatment history	×
Determine diagnosis criteria	×
Determine inclusion criteria	×
Determine exclusion criteria	×
Sign informed consents	×
Interventions
Sham EA
EA treatment
Outcome assessments
Primary Outcome assessments
1. Overall response rate			×
2. NCV	×		×
Secondary Outcome assessments
1. VAS	×	×	×	×
2. TCSS scale	×	×	×	×
3. DQoL	×	×	×	×
4. Physico chemical examination
HbA1c	×		×
FPG, 2hPG	×	×	×
5. NGF, BDNF	×	×	×
6. TNF-α, IL-6	×	×	×
Other work
Safety evaluation		×	×
Adherence evaluation			×
Combined medication	×
Adverse event	×
Processing data	×
Researcher review	×

Sample Size Estimation

The sample size was determined based on the overall response rate, which served as the principal endpoint driving the calculation. Notably, this composite endpoint already incorporates NCV within its efficacy criteria; lower-limb NCV is additionally reported as a continuous measure to provide objective, quantitative support for the response-rate outcome. Accordingly, the study was powered for the overall response rate, and the NCV analysis is regarded as supportive rather than independently powered.

Based on previous studies of EA for DPN and clinical experience, a between-group difference of 20% in the overall response rate was considered clinically meaningful and was used as the target effect size. Power analysis was conducted using PASS 2021 software (NCSS, LLC). Assuming a 1:1 parallel-group design, a two-sided α of 0.05, and 80% power, 41 participants per arm were required to establish statistical significance. Factoring in an estimated 10% attrition rate over the study duration, the final sample size was expanded to 46 patients per group, yielding a total cohort of 92 participants.

Setting

This study was approved by the Medical Ethics Committee of the First Affiliated Hospital of Zhejiang Chinese Medical University (Ethical Number: 2026-KLS-189-01) and registered with the ClinicalTrials.gov Identifier: NCT07551726.

Participant Recruitment

Participants will be recruited from the endocrinology outpatient clinics and the endocrinology inpatient ward of the First Affiliated Hospital of Zhejiang Chinese Medical University. Recruitment will be facilitated through advertisements on WeChat and posters displayed within the hospital and local communities. Eligible individuals who provide informed consent will subsequently undergo baseline assessments.

Eligibility Criteria

Inclusion Criteria

• 1.Aged 18–70 years, with no restriction on disease duration or sex.
• 2.
  Diagnosed with DPN, meeting all of the following criteria:

  1. Objective evidence of neuropathy, ie, decreased motor and/or sensory NCV on lower-limb electroneurography;
  2. Clinical symptoms of peripheral neuropathy, ie, persistent pain and/or sensory abnormalities in the limbs (at least in both lower limbs);
  3. Neurological signs, ie, diminished ankle reflex (unilateral or bilateral) and reduced vibration sense; and
  4. A Toronto Clinical Scoring System (TCSS) score ≥ 6.
• 3.Able to communicate effectively.
• 4.No severe systemic medical conditions, severe psychiatric illnesses, or cognitive impairment.
• 5.Mentally competent, voluntarily agree to participate in the study, and provide written informed consent.

Exclusion Criteria

1. Peripheral neuropathy due to other etiologies, presence of limb ulcers or gangrene, or a history of skin ulceration or poorly healing lesions.

2. Severe comorbid conditions, including renal, cardiovascular, cerebrovascular, pulmonary, or hepatic diseases, infectious diseases, malignancies, or severe psychiatric disorders.

3. History of knee/hip replacement surgery or lower limb fracture within the past 3 months, or any other condition that could interfere with the assessment of neuropathy.

4. Received acupuncture or moxibustion treatment specifically for DPN within the past 3 months.

5. Concurrent participation in another interventional clinical trial.

6. Women who are planning pregnancy, are pregnant, or are lactating.

7. Unwillingness to be randomized to either the EA group or the sham EA group.

8. Chronic abuse of opioids, analgesics, illicit drugs, or alcohol.

Randomization and Allocation Concealment

Using SPSS 26.0 software, random numbers and corresponding group assignments were generated in a 1:1 ratio for 92 participants. Opaque envelopes were numbered from 1 to 92 by a third-party personnel, who placed the random numbers and group allocations inside and sealed them. According to the enrollment order, each envelope was opened sequentially to record the patient’s random number and group assignment. Eligible DPN subjects were thus randomly allocated to either the EA group or the sham EA group.

Blinding

Given the unique nature of the intervention in this study, acupuncturists must engage in direct physical contact with participants, making blinding of the interventionist impossible; however, outcome assessors and statisticians remain blinded. Participants will be blinded to group allocation. Each participant will receive treatment in an independent private room to minimize communication regarding treatment details, needling sensations, or perceived therapeutic effects. To evaluate the effectiveness of participant blinding, at the end of the treatment period participants will be asked to guess their group assignment (active EA, sham EA, or “do not know”), and the success of blinding will be quantified using Bang’s Blinding Index.

Intervention

All participants will receive standardized diabetes education and glycemic control via diet, exercise, oral hypoglycemics, and insulin as needed. Concomitant hypertension or dyslipidemia will be managed with pre-existing medications. Participants will be randomized to EA group or sham EA group, receiving 12 sessions over 6 weeks (twice weekly).

EA Group

Patients will receive EA in addition to their existing treatment regimen. The primary acupoints, localized according to World Health Organization standard anatomical landmarks, will include Yanglingquan (GB34), Zusanli (ST36), Xiajuxu (ST39), Xuanzhong (GB39), Yinlingquan (SP9), Sanyinjiao (SP6), Jiexi (ST41), Taichong (LR3), and Zulinqi (GB41). Supplementary acupoints will be added based on clinical presentation: Kunlun (BL60), Neiting (ST44), and Bafeng (EX-LE10) for significant lower limb pain; and Quchi (LI11), Shousanli (LI10), Waiguan (SJ5), Hegu (LI4), and Baxie (EX-UE9) for concomitant upper limb pain. These supplementary acupoints will be selected strictly according to the above predefined, symptom-based criteria (a lower-limb group for significant lower-limb pain and/or an upper-limb group for concomitant upper-limb pain), and no other points will be added. For each participant, the acupoint prescription will be determined at baseline and kept unchanged throughout the treatment course, and will be recorded on the case report form for every session to ensure treatment fidelity and reproducibility.

Disposable sterile acupuncture needles (Hwato brand) in sizes of 0.18 mm × 25 mm and 0.25 mm × 40 mm will be used. After disinfecting the skin with 75% alcohol, needles will be inserted into the selected acupoints and manipulated using standard needle stimulation techniques to elicit Deqi sensation, characterized by sensations of pain, numbness, or local swelling.

Bilaterally, electroacupuncture electrodes will be connected to acupoints GB34 and SP6 using the Hwato SDZ-IIB electronic acupuncture stimulator. Electrical stimulation will be delivered in continuous wave mode at a frequency of 2 Hz for a duration of 30 minutes per session, with the intensity adjusted to a level perceived as comfortable by the participant. To standardize and document the stimulation intensity, it will be increased gradually to the maximum level the participant can tolerate comfortably without pain, using the same operator-defined endpoint for all participants (visible mild local muscle twitching without discomfort); the corresponding output current (mA) will be read from the stimulator and recorded on the case report form for every session, and any mid-session adjustment due to adaptation will also be recorded.

Sham EA Group

In the sham acupuncture group, needles (0.18 mm × 25 mm, Hwato brand disposable sterile acupuncture needles) will be inserted superficially at points 1 cun lateral to the EA group acupoints to a depth of 1–2 mm, and a modified HWA-SDZ-IIB electroacupuncture stimulator (with intentionally disconnected internal circuitry) will be connected to these needles, actively displaying preset frequency and intensity parameters while delivering no electrical current, with each session lasting 30 minutes. This design was chosen because the type of sham control substantially influences the magnitude of the placebo response.^34,35 By combining shallow, non-acupoint superficial needling with a non-functional stimulator that mimics the operational appearance of active EA, this approach maintains participant blinding and controls for expectation-related placebo effects, while minimizing the specific physiological stimulation associated with deep acupoint needling and true electrical current. The acupuncture and sham procedures are reported in accordance with the STRICTA guidelines (an extension of CONSORT for reporting acupuncture interventions), and the sham design additionally follows recommendations for reporting sham acupuncture (ACURATE).^36,37 The design and reporting of the control intervention were further informed by the CoPPS statement for control interventions in non-pharmacological trials.³⁸ We acknowledge that superficial non-acupoint needling may not be entirely physiologically inert; it was chosen to preserve the credibility and blinding of the control while minimizing specific therapeutic stimulation. Accordingly, the success of participant blinding will be formally assessed using Bang’s Blinding Index and reported by group.

The Position and Function of Acupoints

Figure 2 provides a comprehensive reference to the locations of acupoints in accordance with the 2021 People’s Republic of China National Standard (GB/T 12346-2021) “Acupoints names and positioning”.

Figure 2.

Acupoint locations used in the EA group. Dashed reference lines and tick marks indicate proportional anatomical measurements used for acupoint localization. Numbers beside the tick marks indicate proportional units in cun. Dashed leader lines indicate grouped acupoints, including Bafeng and Baxie. Black dots indicate main acupoints; red dots indicate additional acupoints for lower-extremity pain; blue dots indicate additional acupoints for upper-extremity pain.

Abbreviations: EA, Electroacupuncture; SP, Spleen Meridian; ST, Stomach Meridian; GB, Gallbladder Meridian; BL, Bladder Meridian; LR, Liver Meridian; LI, Large Intestine Meridian; SJ, Sanjiao Meridian.

Rescue Medication and Management

All participants were instructed to refrain from taking any medications intended to alleviate DPN symptoms on their own initiative during the trial. However, rescue analgesic medication (prescribed by a study physician) was permitted if a participant experienced uncontrolled pain that substantially interfered with daily activities and quality of life. The treating clinician was required to promptly document the timing, dosage, and frequency of any rescue medication used by each participant.

Outcomes

The total observation period for this phase of the trial was 10 weeks, comprising a baseline phase, a 6-week treatment phase, and a 4-week follow-up phase. VAS, TCSS, and DQoL will be assessed at all four time points (weeks 0, 3, 6, and 10); the overall response rate at week 6; NCV and HbA1c at weeks 0 and 6; FPG/2hPG and serum biomarkers (NGF, BDNF, TNF-α, IL-6) at weeks 0, 3, and 6.

Primary Outcome

1. The overall response rate to EA for DPN, assessed at the end of the 6-week treatment period. To avoid subjective ambiguity, symptom improvement will be operationally defined using the change in TCSS score from baseline. Treatment efficacy will be categorized into three levels:

• i)Markedly effective: a reduction in TCSS score of ≥50% from baseline, accompanied by an increase in NCV of ≥5 m/s on nerve conduction studies (or restoration to near-normal levels);
• ii)Effective: a reduction in TCSS score of ≥20% but <50% from baseline, with an NCV increase of <5 m/s;
• iii)Ineffective: a reduction in TCSS score of <20% from baseline and no demonstrable improvement in NCV.

The overall response rate = (markedly effective + effective cases)/total participants × 100%.

2. Lower limb NCV, measured at baseline and after the 6-week intervention. This included the motor NCV and sensory NCV of the bilateral common peroneal nerves and tibial nerves.³⁹

Secondary Outcome

1. Visual Analog Scale (VAS) Scores: Used to assess pain, numbness, and burning sensations, where 0 represents “no pain” and 10 represents “severe pain”. Higher scores indicate more severe symptoms.⁴⁰

2. Toronto Clinical Scoring System (TCSS): A 19-point composite score for assessing DPN severity, comprising three subscales: Symptom Score (6 points, rated as 0=absent, 1=present), Reflex Score (8 points, rated as 0=normal, 1=diminished, 2=absent), and Sensory Test Score (5 points, rated as 0=normal, 1=abnormal). The total score is the sum of all subscale scores.⁴¹

3. Diabetes Quality of Life (DQoL) Scale Scores: Items were rated using Likert scales. Most items used a 5-point scale. Psychological distress items used a 6-point scale (1=no distress to 6=very severe distress), with a mean score ≥3.0 indicating serious distress. Scoring: The physical functioning dimension comprises 12 items (total score range 12–60). In the version used, the overall quality of life score ranges from 27 to 135, with lower scores indicating better quality of life.⁴²

4. Serum NGF and BDNF: Fasting morning peripheral venous blood will be collected at baseline (week 0), at the end of week 3, and at the end of week 6. Serum will be separated by centrifugation (3000 rpm, 10 min, radius = 12.7 cm) and NGF and BDNF measured by ELISA.³¹

5. Serum TNF-α and IL-6: collected and processed at the same three time points (weeks 0, 3, and 6) using the same procedure and measured by ELISA.⁴³

6. Glycemic Parameters: Hemoglobin A1c (HbA1c), Fasting Plasma Glucose (FPG), and 2-hour Postprandial Glucose (2hPG) levels.

Safety and Adverse Events

All adverse events (AEs) during the trial—including bleeding, hematoma, dizziness, or severe needling pain—will be systematically monitored and managed until clinical resolution. Investigators will detail the onset, duration, severity, interventions taken, and causality for every AE. Overall safety will be assessed using the AE incidence rate: (number of participants experiencing AEs/total participants) × 100%.

Any serious adverse event (SAE) must be immediately reported to the Principal Investigator and the Medical Ethics Committee to determine if trial suspension is required. An SAE report form will be filed, and the participant will be continuously followed until the event is fully resolved.

Data Collection and Management

1. Data Entry: Data will be entered and verified using Microsoft Excel spreadsheets to ensure the accuracy and reliability of the original research data. Dedicated personnel will perform the statistical entry, and the principal investigator will review the statistical methodology prior to final analysis and discussion.

2. Data Storage: All study documents will be categorized and managed by designated personnel, stored in dedicated folders within secure archival facilities. Access will be restricted to authorized trial personnel, with all viewings and retrievals logged. Study documents will be protected in strict accordance with confidentiality protocols and will be accessible only to relevant research staff. Storage facilities will be equipped with appropriate safety measures, including protection against pests, fire, moisture, and unauthorized access.

Statistical Analyses

Statistical analyses will be performed using SPSS version 25.0. All analyses will be conducted according to the intention-to-treat principle, including all randomized participants. The primary approach for handling missing data will be multiple imputation using chained equations, performed under the missing-at-random assumption and generating 20 imputed datasets, with results combined using Rubin’s rules. To assess the robustness of the findings to the missing-data assumptions, sensitivity analyses will be conducted, including a complete-case analysis and the last observation carried forward method, and consistency across these approaches will be examined. Baseline characteristics will be summarized for all randomized participants. Continuous variables will be presented as mean (standard deviation) if normally or approximately normally distributed, or as median (25^th percentile, 75^th percentile) otherwise. Normality will be assessed using appropriate tests. For comparisons, normally distributed data will be analyzed using t-tests, paired t-tests, analysis of variance (ANOVA), or analysis of covariance (ANCOVA), as appropriate. Non-normally distributed data will be analyzed using nonparametric rank-sum tests. Categorical data will be analyzed using the chi-square test or Fisher’s exact test. A two-sided P value < 0.05 will be considered statistically significant. Analyses of the primary and secondary outcomes will follow a predefined hierarchy. Confirmatory hypothesis testing is reserved for the primary outcome (the overall response rate), for which the two-sided α of 0.05 applies. The secondary outcomes are regarded as exploratory and supportive; to limit the risk of false-positive findings arising from multiple comparisons, their analyses will be interpreted descriptively and, where appropriate, adjusted for multiplicity using the Benjamini–Hochberg procedure to control the false discovery rate. Effect estimates for secondary outcomes will be reported with 95% confidence intervals rather than as confirmatory p-values. For outcomes measured repeatedly at weeks 0, 3, 6, and 10 (eg, VAS, TCSS, and DQoL), between-group differences will be assessed using linear mixed-effects models for repeated measures, with treatment group, time, and their group-by-time interaction as fixed effects, the baseline value as a covariate, and a random intercept for each participant. An unstructured covariance structure will be used to account for within-participant correlation, and under the missing-at-random assumption these models provide valid estimates using all available data. Between-group differences at each time point will be estimated with 95% confidence intervals.

Discussion

DPN remains a formidable clinical challenge due to the limited efficacy and prominent adverse effects of conventional pharmacological treatments. This randomized, sham-controlled trial is meticulously designed to evaluate the clinical efficacy of low-frequency (2Hz) EA in alleviating DPN symptoms, and to elucidate its underlying mechanisms targeting the peripheral neuro-inflammatory axis.

The selection of 2Hz EA as the core intervention is predicated on solid preclinical and clinical evidence. While high-frequency EA often yields rapid but transient analgesia, low-frequency EA has demonstrated a superior capacity for sustained neuromodulation and structural nerve repair. Our previous preliminary studies revealed that 2 Hz stimulation effectively suppresses P2X3 receptor overexpression in dorsal root ganglia, thereby ameliorating neuropathic hyperalgesia.²⁵ By translating these findings into a rigorous clinical setting, we aim to establish a standardized and highly reproducible EA protocol for DPN management.

A distinctive strength of this study is its dual focus on clinical outcomes and molecular mechanisms, specifically the crosstalk between peripheral inflammation and neurotrophic signaling. Chronic hyperglycemia initiates a cascade of pro-inflammatory cytokine release (TNF-α and IL-6), which actively degrades myelin sheaths and exacerbates axonal injury. Concurrently, the deficient synthesis of neurotrophic factors (NGF and BDNF) deprives the peripheral nerves of essential regenerative support. We hypothesize that EA exerts a dual-action therapeutic effect: dampening the excessive inflammatory response while simultaneously upregulating the NGF/BDNF neurotrophic axis. By correlating dynamic changes in these serum biomarkers with objective neurophysiological improvements (nerve conduction velocity), this trial will provide robust biological plausibility for the clinical application of EA.

Methodologically, this study addresses several limitations prevalent in prior acupuncture trials. The implementation of a rigorous sham EA control—utilizing superficial insertion at non-acupoints with visibly active but electrically disconnected stimulators—effectively minimizes the placebo effect and isolates the specific physiological impact of electrical acupoint stimulation. Furthermore, the integration of subjective patient-reported outcomes (TCSS, VAS, and DQoL) with objective biochemical (ELISA) and electrophysiological metrics ensures a highly comprehensive evaluation of the therapeutic efficacy.

Nevertheless, several limitations of this protocol must be acknowledged. First, due to the inherent physical nature of acupuncture interventions, double-blinding is unattainable, and the sham procedure may not fully conceal group allocation from participants, raising the possibility of performance bias related to differential expectations. To mitigate this, a strict single-blind design will be enforced, ensuring complete separation among the treating acupuncturists, outcome assessors, and statisticians. In addition, the success of participant blinding will be formally assessed at the end of treatment using a blinding questionnaire, by asking participants to guess their group assignment. Second, participants will be recruited from a single tertiary hospital, which may limit the broad generalizability of the findings. In addition, the superficial needling at non-acupoints used in the sham group may not constitute a completely inert control, as shallow needle insertion can still elicit mild non-specific physiological responses; this may attenuate the apparent between-group difference and lead to a conservative estimate of the true treatment effect. Finally, the 4-week follow-up period, while sufficient to observe immediate post-treatment effects, may not fully capture the long-term durability of EA-induced neuro-regeneration. Future multi-center, large-scale trials with extended longitudinal follow-ups will be warranted.

In conclusion, this protocol describes a randomized, sham-controlled trial designed to evaluate the clinical effect and safety of standardized low-frequency EA for patients with DPN. By integrating clinical, electrophysiological, and serum biomarker outcomes, this trial is intended to generate rigorous, standardized data on low-frequency EA in DPN and to inform the design of future confirmatory studies. The findings will help clarify whether EA warrants further investigation as a potential non-pharmacological approach for DPN, rather than to establish its clinical efficacy at this stage.

Funding Statement

This study was supported by the Research Special Project for Affiliated Hospitals of Zhejiang Chinese Medical University—Youth Talent Program (2025FSYYZQ08).

Generative Artificial Intelligence

During the preparation of thisarticle, the authors used ChatGPT based on GPT-5.5 Thinking (OpenAI, San Francisco, CA, USA) solely for English language polishing and grammatical refinement. The tool was not used to generate scientific ideas, design the study, or create references. All AI-assisted text was carefully reviewed and edited by the authors, who take full responsibility for the accuracy, originality, and integrity of thearticle.

Data Sharing Statement

No datasets were generated or analyzed during the preparation of this study protocol. After completion of the trial, de-identified data supporting the findings of the study will be available from the corresponding author upon reasonable request, subject to ethical and privacy considerations.

Ethics Approval and Consent to Participate

Ethical approval was obtained from the ethics committee at the First Affiliated Hospital of Zhejiang Chinese Medical University (Ethical Number: 2026-KLS-189-01). The trial was prospectively registered with ClinicalTrials.gov (NCT07551726). Furthermore, the trial will be conducted in accordance with the Declaration of Helsinki and applicable Chinese regulations, with written informed consent obtained from all participants.

Author Contributions

Nisang Chen: Conceptualization, Project administration, Writing – original draft, Writing – review & editing. Rongrong Li: Investigation, Data curation, Visualization, Writing – original draft. Ning Luo: Methodology, Supervision, Funding acquisition, Writing – review & editing. All authors made a significant contribution to the work reported, whether in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising, or critically reviewing the article; gave final approval of the version to be published; agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The author reports no conflicts of interest in this work.