Acupuncture for diabetic peripheral neuropathy: study protocol for a randomized, placebo-controlled, single-blinded trial

Haiping Deng Shanghai University of Traditional Chinese Medicine https://orcid.org/0000-0001-7124-5716 Yu Shu Shanghai University of Traditional Chinese Medicine Peiran Lv Shanghai University of Traditional Chinese Medicine Ling Zhao Shanhai University of Traditional Chinese Medicine Ke Cheng Shanghai University of Traditional Chinese medicine, Shanghai Research Center for Acupncture and meridian Tingting Zhang Shanghai University of Traditional Chinese Medicine Yi Song Long-Hua Hospital a liated to Shanghai University of Traditional Chinese Medicine Hua Yang Long-Hua Hospital A liated to Shanghai University of Traditional Chinese Medicine Hong Tang Long-Hua Hospital A liated to Shanghai University of Traditional Chinese Medicine Jian Pei Long-Hua Hospital A liated to Shanghai University of Traditional Chinese Medicine Xueyong Shen (  sxy1@shutcm.edu.cn ) https://orcid.org/0000-0003-4177-8970


Abstract
Background Diabetic peripheral neuropathy (DPN) is the most common chronic complication of diabetes mellitus that has a considerable impact on quality of life, but there are few effective therapeutic strategies. The aim of this trial is to determine the e cacy and safety of manual acupuncture (MA) versus sham acupuncture (SA) for DPN.

Methods/design:
This is a study protocol for a randomized, two-arm, single centre, clinical trial. A total of 118 patients with DPN will be recruited and randomly assigned to the MA group or the SA group in a 1:1 ratio. All patients will receive 24 sessions over 12weeks. Participants will complete the trial by visiting the research centre at month 6 for a follow-up assessment. The primary outcome is the peroneal motor nerve conduction velocity (peroneal MNCV) at week 12 compared with baseline. Secondary outcomes include the peroneal motor nerve action potential amplitude (peroneal MNAP) and latent period(peroneal MNLP), the sural sensory nerve conduction velocity (sural SNCV), action potential amplitude (sural SNAP) and latent period (sural SNLP), fasting plasma glucose(FPG), 2-hour postprandial blood glucose (2hPG), glycated hemoglobin (HbAlc) at week 12 compared with baseline, the Michigan Neuropathy Screening Instrument (MNSI) score and Diabetes Speci c Quality of Life scale(DSQL) at week 12 and month 6 compared with baseline. Safety is assessed during the whole trial. Masking effectiveness is assessed by patients.

Discussion
This trial may provide high-quality evidence for evaluating the e cacy and safety of MA treatment for DPN compared with SA treatment. The results of this study will be published in peer-reviewed journals. Background DPN is the commonest cause of neuropathy worldwide, and its prevalence increases with the duration of diabetes [1]. Approximately 50% of patients with diabetes will develop peripheral neuropathy [1,2,3]. DPN is symmetric and predominantly sensory, starting distally and gradually spreading proximally in a gloveand-stocking distribution [1]. DPN is associated with the development of foot problems and is one of the leading causes of amputations and the development of foot ulcers. DPN is frequently underreported and undertreated and can lead to an increased risk for morbidity and mortality and decreased quality of life [1,2]. Once symptoms appear, there are few effective therapeutic strategies [4]. Whilst neuropathic pain and paresthesia can be palliated by anti-convulsants, tricyclic antidepressant drugs or serotoninnoradrenalin re-uptake inhibitors [5], pharmacologic management of decreased sensation is generally ineffective. Therefore, more and more patients have turned to seek nonpharmacological treatments. Multiple complementary and alternative medicine therapies such as acupuncture and yoga show e cacy in the treatment of painful peripheral neuropathy [6]. A pilot RCT has demonstrated the practicality and feasibility of acupuncture as an additional treatment for people with DPN. The treatment was well tolerated with no appreciable side effects [7]. Acupuncture may effectively ameliorate selected DPN symptoms such as aching pain, burning pain, prickling sensation, numbness, allodynia in American Indian patients [8]. Reviews concerning acupuncture for peripheral neuropathy (PN) have found that, despite the majority of studies reporting positive results, a reliable statement of effectiveness is not possible due to methodological limitations [9][10][11][12]. A recent meta-analysis of acupuncture for DPN indicates acupuncture for DPN appears to improve symptoms [13]. However, the application of acupuncture varies greatly, and the quality of included studies was generally low. Available studies have varying methodologies and different outcome measures. Further, suitably powered studies using appropriate DPN outcome measures are required [13]. Therefore, we have designed a study with a 6-month observation period to explore its effectiveness and safety.

Study design
This randomized, participant-blind, sham controlled trial will be conducted at Longhua hospital a liated to Shanghai University of Traditional Chinese Medicine in China. The protocol for this trial is reported

Patient recruitment
Patients meeting the DPN diagnostic criteria will be recruited from the endocrinology departments of Longhua Hospital a liated to Shanghai University of Traditional Chinese Medicine in Shanghai, China. Recruitment posters will be posted in the hospital and also be uploaded to social Internet media (WeChat) from October 2016 to December 2021. All patients will be required to provide written informed consent (Additional le 2) before randomization.
Inclusion criteria 1. Aged 18-75years (either sex); 2. Clinically Con rmed diagnosis of diabetic peripheral neuropathy; symmetric and predominantly sensory, starting from the lower limbs distally and gradually spreading proximally in a glove-andstocking distribution; 3. Ability to understand study procedures and willingness to comply with them for the entire length of the study;

Written informed consent
Exclusion criteria 1. PN caused by conditions other than diabetes (e.g., alcohol abuse, chemotherapy, hereditary causes, chronic in ammatory or idiopathic PN, and others); 2. Psychiatric illnesses other than mild depression; 3. severe or unstable cardiovascular, liver, kidney, respiratory or hematological disorders; 4. Acupuncture treatment in the last 3 months; 5. Pregnant or breast-feeding women; 6. Research unit personnel directly related to the study and their immediate family members; 7. Incapacity in giving informed consent or in following the study instructions due to language disturbances, serious cognitive de cits, or lack of time; 8. Current participation in other clinical trials.

Randomization and allocation concealment
All eligible patients will be randomly assigned to the A group or SA group in a 1:1 ratio. The blocked randomization sequence will be computer-generated with the EXCEL software by an independent statistician, who is not involved in the implementation and statistical analysis of the trial. The randomization will be strati ed by the number of limbs affected by DPN (i.e., both only the lower limbs and four limbs) and with a random block size of 6 and 4. The allocation was placed inside sequentially ordered sealed opaque envelopes, opened only after enrolment. The treatment allocation was revealed to the acupuncturists out of sight of the patients to ensure blinding. To reduce the risk of observer bias, the acupuncture practitioners were discouraged from discussing the treatments or previous results with the patients. The random allocation sequence and the opaque sealed envelopes will be kept separately by two speci c researchers. The clinical research coordinator will be responsible for enrolling patients, obtaining informed consent and requesting randomization.

Masking
Due to the nature of acupuncture, masking of acupuncturists is quite di cult to achieve. Patients, outcome assessors, and statisticians who perform the statistical analyses will be blinded to group assignment. Participant's allocated intervention will be not revealed until the statistical analysis is completed.
Interventions Treatment will be performed by licensed acupuncturists who have at least ve years of experience in acupuncture. All the acupuncturists will be trained how to locate acupoints, puncture, and manipulate needles before the trials. Single-use sterile disposable acupuncture needles (length: 25-40mm, diameter: 0.25mm; Jiajian, Wuxi China) will be used. Both the MA and SA treatments will consist of 24 sessions of 40-min duration over 12 weeks (two sessions per week). Acupuncture will be discontinued if the patients suffer from any serious adverse events (AEs).

Manual acupuncture
Patients allocated to the MA group will take treatment with needles inserted at the prespeci ed  Table 1 and Fig. 2. After skin disinfection, 25mm or 40mm-length acupuncture needles were inserted approximately 10-20 mm into the skin according to acupoint, then manipulations of twirling, lifting, and thrusting will be performed on all needles for at least 10s to reach De qi (a compositional sensation including soreness, numbness, distention, and heaviness), which is believed to be an essential component for acupuncture e cacy. Firstly, RN12, ST36, GB34, SP6, KI3, EX-LE10, EX-UE9 will be punctured when patients lie supine, then, EX-B3, BL18, BL20 and BL23 will be punctured when patients lie prone, the needles will be retained in each acupoints for 20min.

Sham acupuncture
Patients in the SA group received sham acupuncture. After skin disinfection, sterile adhesive pads were placed on the same acupoints with the MA group, then the 25mm-length blunt tipped placebo needle were inserted through the adhesive pads to the surface of the skin, but no skin penetration, without needle manipulation for De qi.
Although it is di cult to set an eligible placebo control, using a blunt tipped needle which does not penetrate patient's skin is the most commonly used approach for administering sham treatments among acupuncture trials on the basis of a literature review [14,15]. And this study will exclude those received acupuncture treatment in the last 3 months. This population can most probably distinguish SA from MA. The same control, no skin penetration at acupoints, was adopted and is successful to mask patients with Page 7/17 DPN [7]. Furthermore, all patients will be asked to guess which treatment they have received to test the patient-blinding effects at week 12.

Concurrent Treatments
Participants will receive routine health care as provided to all other patients with DPN. These treatments include glucose control, antihypertensive therapies, dyslipidemia control, analgesic and neurotrophic treatments, if necessary. All information regarding the use of medications (including date of administration, and types and dosage) will be recorded. This contribute to improving adherence.

Primary endpoint
The primary endpoint is the difference of the 12-week measurement minus baseline of the peroneal motor nerve conduction velocity (peroneal MNCV).

Key secondary endpoints
Secondary endpoints are the differences of the 12-week measurement minus baseline of the peroneal motor nerve action potential amplitude (peroneal MNAP) and latent period(peroneal MNLP), the sural sensory nerve conduction velocity (sural SNCV), action potential amplitude (sural SNAP) and latent period(sural SNLP).

Quality of life (QoL)
Disease-speci c QoL will be assessed at baseline and at weeks 12, month 6 after randomization using Diabetes Speci c Quality of Life scale (DSQL) which consists of four domains: interference (12 items); psychology (8 items); social relations (4 items); and treatment (3 items). Each item is measured with a 5point Likert scale ranging from "not at all" to "extremely." Higher scores indicate a worse QoL.

Patients' global assessment of DPN
Patients' global assessment of DPN is evaluated on a 5-point Likert scale at week 12. Patients are asked to respond to the following question: 'Considering all the ways your DPN affects you, how are you doing today?' 1=very poor; 2= poor; 3=fair; 4= good; and 5=very good.

Blinding assessment
To test the patient-blinding effects, all patients will be asked to guess their group assignment allocation within 2min after the last treatment session in week 12 as following: "which group do you think you are in?: A. traditional acupuncture; B. modi ed acupuncture; or C. not sure".

Credibility and expectancy
Credibility and expectancy of patients will be assessed using the Credibility/Expectancy Questionnaire before the rst treatment.

Adverse events
AEs data will document the occurrence, duration, and severity of adverse reactions (symptoms and signs), and how the event was resolved (or not) during the treatment. Based on their potential association with the acupuncture needling procedure, AEs will be categorized by acupuncturists and related specialists as treatment-related or not within 24h of occurrence. Common treatment-related AEs include local subcutaneous hematoma, itching at the sites of needle insertion, continuous post-needling pain, dizziness, and so on. All participants will receive routine blood test, liver function (alanine transaminase and aspartate transaminase), and kidney function tests (serum creatinine and blood urea nitrogen). These tests will be performed twice after randomization and at the end of the 12-week treatment period. Serious adverse events are reported to Medical Ethics Committee and treat the participants with relevant conventional therapy or hospitalization if necessary (the participant's allocated intervention will be revealed).

Assessment of safety
The patients are asked to assess the safety of the treatment after 12 weeks' treatment, in terms of 4 grades: safe, relatively safe, unsafe, and very unsafe.

Sensation during the treatment
The patients are asked about their sensation during the treatment at every visit as following: "what sensation do you feel during the treatment? A. soreness; B. distention; C. pain; D. no feeling; E. else, please specify___."

Usage of medication
Medication Usage Log: The participants are given a printed log to record their daily intake of medications prescribed on a symptom-contingent (i.e., as per necessary, prn) basis. We measure the outcomes at week 12.
The schedule of enrolment, interventions, and assessments are presented in Fig. 3.

Data collection, management and monitoring
All researchers including acupuncturists, outcome assessors, and statisticians will receive training regarding data management. The Case Report Form (CRF) will be rst completed and then double entered into the Electronic Data Capture (EDC) system electronically by two independent investigators to ensure the accuracy of the data.
All research documents, including both the paper les and electronic documents, will be preserved for at least ve years after publication. If reviewers or readers have any questions regarding our published data, they can contact the corresponding author for access to the original data or visit ResMan (http://www.medresman.org/uc/project/projectadd.aspx). The private information of patients-including name, telephone number, and ID number will be anonymous to ensure participant con dentiality.
The safety of the study will be monitored by a Data and Safety Monitoring Board (DSMB) of the clinical evaluation center of Longhua hospital, a liated Shanghai university of TCM, which consists of independent clinical experts and statisticians with access to unblinded data. The DSMB is independent from the sponsor, the competing interests and the investigational site and will review the performance and safety of the trial per six months.
The criteria for unblinding and discontinuing allocated interventions for a given trial participant experiencing serious acupuncture-related AEs (if any), which have been described previously. The DSMB will reveal a participant's allocated intervention, and make the nal decision to terminate the trial.

Sample size
Based on the results of a previous study, a sample size of 49 participants per group was estimated to provide 80% power to detect a between-group difference of 2.4m/s of the conduction velocity improvement of the common peroneal nerve, assuming a 2-sided signi cance level of 5%. To compensate for a 17% loss to follow-up, the sample size was increased to 59 participants in each group.

Statistical analysis
The statistical analysis will be performed by an independent statistician who is not aware of the group allocation. SPSS 21.0 statistical software (IBM SPSS Statistics, New York, USA), will be used for data analysis. The level of signi cance will be established at α< 0.05 with a two-sided test. Continuous data will be represented as the mean ± standard deviation or median (range), whereas categorical data will be represented by percentage. All e cacy analyses will be performed using the intent-to-treat (ITT) approach. For the ITT analysis, the population will consist of all patients who have been randomized. Missing data will be imputed using the last observation carried forward (LOCF) method. Continuous variables will be compared using a Student's t-test or Wilcoxon rank-sum test as appropriate. Categorical variables will be compared using Fisher's exact test or Wilcoxon rank-sum test as appropriate.

Ethics and dissemination
The study protocol follows the principles of the Declaration of Helsinki and was approved by the Medical Ethics Committee of Long-Hua Hospital a liated to Shanghai University of Traditional Chinese Medicine on 7 July 2016 (Approval No. 2016LCSY028). The study was registered with Chinese Clinical Trial Registry, ChiCTR1800020444. The results will be disseminated through peer-reviewed publications, a master's thesis, or conference presentations. The data will be anonymized prior to publication to prevent identi cation of individual participants.

Discussion
DPN impacts patients' QoL and causes a nancial burden for society. This trial will evaluate the e cacy of MA versus SA in improving the symptoms of DPN.  Locations of acupoints