Perineural injections of incobotulinumtoxin-A for diabetic neuropathic pain of the lower extremities: protocol for a phase II, single-centre, double-blind, randomised, placebo-controlled trial (the PINBOT study)



Diabetes affects an estimated 422 million people, equivalent to 5.3% of the world’s population and is the most common cause of neuropathic pain.1 Diabetic neuropathic pain (DNP) is a serious and debilitating complication affecting 15–20% of people with diabetes. DNP is thought to develop as part of the chronic sensorimotor distal polyneuropathy typical of diabetic polyneuropathy, usually arising in the toes and distal parts of the feet and progressing proximally over time in a stocking-shaped pattern.2 Although related to diabetic polyneuropathy, the severity of DNP is not necessarily associated with the severity of neuropathy and can occur without other sensorimotor symptoms.3 Pain is generally characterised as being sharp, burning, tingling or shocking in the affected area, of moderate to severe intensity, and worse at night.4 Cutaneous allodynia and hyperalgesia are common. The underlying mechanisms responsible for DNP are not fully understood, but ischaemia resulting from microvascular damage, changes in sodium channel expression and central excitation mechanisms are thought to contribute.5 6 DNP is a strong predictor of depression and poor sleep and has a significant negative impact on self-reported quality of life, health-related quality of life and physical activity.7 8

Current treatment options

The 2022 pharmacotherapy guidelines published by the American Association of Neurology (AAN)9 recommend tricyclic antidepressants, selective norepinephrine reuptake inhibitors, gabapentinoids and sodium channel blockers such as valproaic acid, lamotrigine and oxcarbazepine as equal first-line treatments. Topical treatments such as capsaicin and glyceryl trinitrate spray may be effective, but the available evidence is weak. The 5% lidocaine patch was included in previous guidelines10 11 (with level C evidence) but was not addressed in the latest edition. All recommended classes of oral medications list dizziness, somnolence and fatigue as common side-effects, in addition to other side-effects particular to the individual drug and all topical treatments can cause local reactions.

Botulinum toxin

Botulinum toxins are a group of potent neurotoxins, produced by the anaerobic gram-positive soil bacterium Clostridium botulinum. Currently, eight different serotypes of toxins are recognised (A–H), of which only A and B are used medically.12 Botulinum toxin is a well-established treatment for muscle overactivity, such as spasticity and dystonia, as well as hyperhidrosis and sialorrhea, chronic migraine, for cosmetic purposes and, more recently, for neuropathic pain.13

The use of botulinumtoxin-A (BonT-A) was not covered in the AAN 2022 publication. However, the International Association for the Study of Pain (IASP), Neuropathic Pain Special Interest Group (NeuPSIG) 2015 guidelines14 recommend subcutaneous BonT-A as a third-line treatment for peripheral neuropathic pain, citing excellent safety and low placebo effect. There is mounting evidence for the safety and efficacy of subcutaneous BonT-A for treating peripheral neuropathic pain in general.15 For DNP especially, five randomised-controlled trials (RCTs) have been published to date, with varying techniques, but all with positive results.16–20

The use of subcutaneous injections can be limited by the need to distribute the drug directly in the affected neuropathic area. Typically, an affected area requires a dozen or more individual injections in a grid formation, a process that can be both time-consuming and limits the size of the area that can be treated. For patients suffering from allodynia or mechanical hyperalgesia, such injections can be especially painful. Perineural injections proximal to the site of pain could reduce the number of injections needed by treating the afferent sensory nerve of the painful area, instead of the area itself. However, evidence for perineural use of BonT-A is limited and no definite mechanism of action has been established. The largest systematic work to date by Meyer-Frießem and colleagues is a case series on ultrasound-guided perineural BonT-A injections in 60 patients with postherpetic or peripheral nerve injury pain. The authors describe the treatment as safe, well tolerated and effective, with responders reporting sufficient pain relief for 3 to 4 months after injection.21 For occipital neuralgia, several smaller case reports have reported significant pain relief, with no adverse events.22–24

This paper describes the protocol and presents the rationale for an adequately powered, double-blind, randomised, placebo-controlled study. The current phase II trial represents the largest RCT investigating perineural BonT-A, and the first in this patient group.

Our primary objective is to investigate if BonT-A injected perineurally around the distal sciatic nerve is effective in reducing average daily pain scores in participants with DNP of the lower extremities. Second, we wish to assess the safety of this route of administration, as well as the treatment’s effects on use of rescue medication, depression symptoms, activities of daily living, levels of physical activity, health-related quality of life and sensory profiles.

Methods and analysis


The trial is a single-centre, randomised, double-blind, placebo-controlled study. 80 participants will be randomly assigned in a 1:1 ratio to receive treatment with either 100 units of incobotulinumtoxin-A (Xeomin), injected perineurally around each distal sciatic nerve, or a saline placebo. Following a 1-week baseline period, participants will be treated twice, once every 12 weeks, for a total study period of 25 weeks.

Patient and public involvement

Patients were not directly involved in the design and drafting of this study, nor was any direct involvement planned for recruitment and active phases. All study participants will be provided access to any publications resulting from this trial via email.

Sample selection and recruitment

Participants will be recruited from endocrinological and pain clinics in the Capital and Zealand regions of Denmark. All examinations and treatments will take place at the Neurological Pain Clinic, Rigshospitalet Glostrup in Copenhagen, Denmark. Potential participants are referred by their practicing physician and contacted by phone for initial information and screening. A formal information session is held before informed consent can be provided. No incentives—economic or otherwise—are offered for participation.

Eligibility criteria

Persons fit for inclusion are those who (1) are aged >18 years; (2) have a diagnosis of diabetes type I or II; (3) score 3 or above on the interview section of the Douleur Neuropathique 4 questionnaire;25 (4) suffer from pain of the lower extremities, which is (a) considered the participant has their overall dominant pain; (b) is rated at least 4 out of 10 on the Numeric Rating Scale (NRS) pain scale in both legs (on average over the past 7 days); (c) is present in both feet, roughly symmetrically; and (d) has been present for at least 6 months; (5) show sensory deficits and/or allodynia or hyperalgesia in the painful area, consistent with the IASP definition on probable chronic neuropathic pain;26 (6) are in a stable analgesic treatment regime for at least 1 month prior to inclusion and for the duration of the study (note: participants with existing treatments are not altered prior to inclusion or during study); (7) are using an approved, safe contraceptive (for premenopausal women); and (8) speak, read and understand Danish.

Criteria for exclusion are those who (1) are allergic or hypersensitive to BonT-A; (2) have been treated with BonT-A for up to 6 months prior to inclusion; (3) are diagnosed with myasthenia, Eaton-Lambert syndrome or amyotrophic lateral sclerosis; (4) have a known malignant condition; (5) have an ongoing infection in the area of injection; (6) are expecting to change their pain medication during the study period; (7) have received treatment with topical agents such as capsaicin or lidocaine in the affected area for up to 3 months prior to inclusion; (8) have a known competing cause of central or peripheral neuropathic pain or other painful chronic pain conditions of the lower extremities such as spinal stenosis, claudication, previous trauma or nerve injury or cancer-related pain; (9) have a psychiatric disorder which prevents completion of the study (as assessed by the investigators); (10) are active on abuse of alcohol or illegal narcotics, including cannabis; (11) are using or receiving treatment with cannabis products of any kind; (12) are pregnant or planning pregnancy during the study period; and (13) score ≥12 on the Charlson Comorbidity Index.27

Randomisation and blinding

All eligible participants will, on giving their informed consent, be enrolled consecutively in the study and assigned a study ID. Software-based simple block randomisation will be performed in a 1:1 ratio by an external party at the Capital Region Pharmacy (Region Hovedstadens Central Apotek). All study personnel, participants and outcome assessors will be blinded to the participants’ allocation. Vials of active and placebo are visually identical and labelled identically with the participants’ study ID, trial information and expiration dates.

Sample size

A meta-analysis of subcutaneous BonT-A treatment for painful neuropathy showed an average reduction of 1.49 at 12 weeks (NRS or Visual Analogue Scale (VAS)).15 A study of 5514 Danish patients with diabetes reported an average pain score of 5.3 (NRS, SD=2.1) in those suffering from painful neuropathy.28 Assuming a power of 80% and alpha=0.05, using a two-tailed formula, a sample size of 31 participants in each arm is calculated. Compensating for dropout, we aim to include 40 participants in each arm, for a total of 80.

The calculation is conservative by design, given the lack of data on this mode of administration. A recent RCT of intradermal BonT-A for treating DNP reported a mean change in pain from 6.8±1.1 to 2.1±1 at 4 weeks follow-up in the group treated with BTX-A in both feet.20

Procedure description

Treatment consists of a perineural injection around the distal sciatic nerve of both legs. The procedure is identical for placebo and active groups and is performed as follows: 5 mL of sterile saline (0.9%) is added to the vial containing either BonT-A or a placebo. The participant is positioned supine with the lower leg elevated, allowing scanning from the posterior surface of the thigh. The operator stands at the side of the patient. The display of the ultrasound scanner is positioned across the table, ensuring optimal ergonomics. Using a high-frequency linear transducer (Sonosite PX, L15-4), footprint of 50 mm, with initial depth setting of 35–45 mm, scanning is initiated with the probe along the knee crease. The operator slides the transducer proximally from the knee crease until the common peroneal and tibial nerves are identified. The target of injection is the popliteal fossa immediately after the sciatic nerve has given rise to the tibial and the common peroneal nerves where the two nerves are still enveloped in a common epineural sheath. The skin is penetrated from the lateral side using a non-cutting needle (Pajunk, SonoBlock, 22G×80 mm, Facet S Tip). With the needle in plane in relation to the ultrasound probe, the nerves are visualised in short axis. The needle tip is placed inside the common sheath surrounding the tibial and common peroneal nerves, that is, in the shared subepineural space. The location of the needle tip is verified with small boluses of sterile saline solution (0.9%) in combination with ultrasound. Correct distribution of the injectant is confirmed with dynamic scanning. Fluid distention must be seen around both nerves. When optimal needle placement is confirmed, 100 units of BonT-A or placebo are injected. The procedure is repeated for both legs. The second treatment will be performed regardless of the perceived efficacy of the first treatment (assuming there are no adverse events or other safety concerns). This includes any residual analgesic effects.

Study timeline and data collection

The study period will consist of 1 week of run-up to establish baseline measurements of pain and rescue medication usage and two 12-week cycles of treatment and follow-up, for a total of 25 weeks or 168 days (figure 1). Data collection consists of a daily physical diary recording average pain scores and use of rescue medication, in-person visits, in which participants complete questionnaires on secondary outcomes and undergo physical examination, and regular follow-up by phone where side-effects and changes in lifestyle or medication are recorded. The scheduled visits and follow-ups are identical for the active and placebo groups and consist of the following (±4 days):

  • Day −7: Participants will be contacted by telephone and reminded to start their daily registration of pain intensity and use of rescue medication for the duration of the study.

  • Day −2: Baseline physical examination and interview.

  • Day 0: First injection of both lower extremities.

  • Day 7: Follow-up via telephone. Side-effects and changes in lifestyle and medication are recorded. Reminder to keep daily log of pain scores and use of rescue medication.

  • Day 28: Follow-up physical examination and interview.

  • Day 42: Follow-up via telephone.

  • Day 82: Start of second 12-week cycle: Follow-up physical examination and interview.

  • Day 84: Second injection of both lower extremities.

  • Day 91: Follow-up via telephone, questionnaire on side-effects and changes in lifestyle and/or medication. Reminder to keep daily log of pain scores and use of rescue medication.

  • Day 112: Follow-up physical examination and interview.

  • Day 126: Follow-up via telephone.

  • Day 168: Final interview and physical examination.

Figure 1
Figure 1

Timeline of visits, follow-ups, treatments and data collection.

Recruitment is currently ongoing, and the first treatments are scheduled for January 2024. The study is expected to conclude by late 2025.

Outcomes and instruments

The choice of instruments and techniques used are based on NeuPSIG best-practice guidelines29 for neuropathic pain research to enable direct comparison to similar studies.

Primary outcome

Participants will be asked to rate their average pain for the preceding 24 hours on the 11-point NRS, once a day, at a consistent time for 7 days prior to the first injection, and for the duration of the study. Daily pain scores are recorded by participants in a provided pain diary.

Secondary outcomes

Use of rescue medication

The provided pain diary also contains a section for recording daily use of rescue medication. Participants are asked to record if they have used rescue medication on a given day. Dosage, frequency and timing are not recorded here, but participants’ analgesic medication regimen is registered at baseline and expected to remain unchanged for the duration of the study. Any analgesic taken pro re nata for the symptoms of DNP is considered rescue medication for this purpose.

Neuropathic Pain Symptom Inventory (NPSI)

The NPSI is a 12-item questionnaire designed to characterise and quantify different aspects of neuropathic pain.30 It covers 10 descriptors of the different symptoms (burning, squeezing, pressure, electric, etc.) and two items for assessing the duration of spontaneous ongoing and paroxysmal pain. Questions dealing with severity of symptoms are rated on an 11-point Likert scale, while those dealing with frequency of symptoms use a 5-point interval. Scoring of symptoms can be done as the sum of all questions or as symptom groups, for example, evoked pain or paraesthesia/dysesthesia.

The NPSI has been validated in multiple languages including Danish and is able to differentiate between nociceptive and neuropathic pain with a high degree of certainty, while also being sensitive to change.31

Beck Depression Inventory II (BDI-II)

The BDI-II is a widely used and validated psychometric test for measuring the severity of depression.32 It consists of a 21-question multiple-choice self-report inventory covering different mental and physical symptoms related to depression, each rated on a 4-point scale (0–3) of severity. The final score is denoted as the sum of the total severity ratings for each item, ranging from minimal symptoms (0–13), mild (14–19), moderate (20–28) and severe (28–63).


The EQ-5D-5L (EuroQoL – 5 Dimensions – 5 Levels questionnaire) is a generic measure of health-related quality of life, consisting of two parts, a descriptive system assessing five different dimensions of health, and a VAS on which the participant rates his or her perceived health from 0 (the worst imaginable health) to 100 (the best imaginable health).33 It is widely used to assess the self-perceived disease burden and its impact on quality of life.

Canadian Occupational Performance Measure (COPM)

The COPM allows assessment of real-life functional impact of treatment by measuring changes in self-perceived ability and satisfaction with everyday activities.34 During their first visit on day −2, participants are asked to identify the five most important activities that he or she is not currently satisfied with his or her ability to perform. Degree of satisfaction with and ability to perform the activities is then rated on a 1–10 scale. The same activities are rated again by the participant on subsequent visits using the same scales and attempts to measure real-world impacts of treatment. The COPM is widely used in Danish clinical occupational therapy settings and has been validated in Danish.35

Physical activity

To investigate if the treatment promotes an increased level of everyday physical activity, the Grimby-Saltin Physical Activity Scale (PAS-2 DK) is used.36 The PAS-2 is a 7-point questionnaire measuring self-reported minutes and hours spent on a range of physical activities, on a daily and weekly basis.

Quantitative Sensory Testing (QST)

Previous RCTs, notably the 2016 BOTNEP study by Nadine Attal and colleagues,37 have shown that sensory profile may predict efficacy of BonT-A treatment for neuropathic pain. To assess if perineural BonT-A injections affect sensorium distal to the injection site and to establish sensory profiles, participants will have the sensory function of their lower extremities tested using QST, as specified by the German Research Network for Neuropathic Pain (DFNS) protocol.38 The DFNS protocol contains seven separate testing modalities for determining 13 parameters: cold detection threshold, warmth detection threshold, thermal sensory limen, paradoxical heat sensation, heat pain threshold, cold pain threshold, mechanical detection threshold, mechanical pain threshold, mechanical pain sensitivity, dynamic mechanical allodynia, wind-up ratio, vibration detection threshold and pressure pain threshold. Thermal testing will be performed using a thermal stimulator (TSA-2, Medoc), mechanical testing using graded Von Frey (OptiHair-2, MRC) filaments and calibrated weighted needles (Pin-Prick, MRC). Vibration testing will be performed using Rydel–Seiffer graded tuning fork (64 Hz, 8/8 scale), allodynia testing with a SenseLab Brush-05 (Somedic) and pressure pain testing with an algometer (Algomed, Medoc) fitted with a 1 cm2 probe. All tests will be performed on the dorsum of the foot, with the exception of vibration testing (medial malleus) and pressure pain threshold (sole of the foot). Both lower extremities will be tested on each occasion. Data will be analysed separately for each leg.

Motor function

As BonT-A is a potent paralytic and the distal sciatic nerve is both a motor nerve and is surrounded by skeletal muscle, transient paresis could potentially occur. To establish if repeated perineural BonT-A injections affect motor function in and around the innervated area, muscle power at the ankle and knee joint will be recorded using the Oxford MRC Muscle Power Assessment.39 Ankle dorsiflexion, plantarflexion, inversion, eversion, and knee flexion and extension are tested for each leg.

Data analysis and management

Data are recorded electronically and stored in a secure RedCap database secured by two-factor authentication, access to which is restricted to study personnel.

Intention to treat (ITT) analyses will be conducted for all participants who have been enrolled in the study and have received the first treatment, regardless of degree of completion.

Per protocol (PP) analyses will include all participants who complete the full 24 weeks of treatment and follow-up, according to protocol. All questionnaires must be completed within 4 days of the scheduled time. All analyses detailed below will be conducted for both ITT and PP populations. Categorical baseline characteristics will also be compared between the groups using a χ2 test, and continuous variables will be compared using an independent t-test (in the case of normality of data) or a Wilcoxon Rank-Sum test in the case of non-normality.

Handling of missing data

Missing NRS values are acceptable, but at least five daily NRS values per week must be entered on average. Missing values for NRS are imputed using the multiple imputations by chained equations method.

Primary outcome

The primary outcome will be tested using a generalised linear mixed model of repeated measures of daily (or weekly means) pain scores, comparing active and placebo groups for 24 weeks, with the mean baseline pain score of each group as a covariate.

Secondary outcomes

Average days with use of rescue medication per week, NPSI total and subscores, EQ-5D-5L VAS of self-perceived health scores, COPM scores, PAS-2 DK scores, BDI-II scores and motor function for active and placebo groups will be compared with baseline and between groups using a generalised linear mixed model of repeated measures, with baseline values as a covariate.

Sensory profiles will be generated for each QST sample as stipulated by the DFNS protocol for both legs separately. Accordingly, recorded values will be z-transformed by subtracting the mean value of an age-matched and sex-matched reference group. Z-scores of >1.96 or <−1.96, that is, falling outside the 95% CI for matched healthy controls, will be considered abnormal, either as loss of function or gain of function. Absolute scores will also be compared before and after treatment and between the two groups.

Ethics and dissemination

The study will be conducted in accordance with the Helsinki Declaration, ICH Guidelines and the Danish code of law and is subject to inspection by national Good Clinical Practice (GCP) authorities. All participants are insured by the Danish Patient Compensation Association.

Safety profile

Both the use of BonT-A across a wide range of conditions and perineural infiltration with local anaesthetic agents have excellent safety profiles.40 41 BonT-A is not directly neurotoxic42 and we do not foresee any additional risks in this route of administration, compared with regional blocks with local anaesthetics or BonT-A treatments for other conditions using similar dosages.

Adverse events will be monitored and documented throughout the study period in accordance with Danish GCP guidelines. At every visit and follow-up by telephone, participants will be asked a set of questions regarding side-effects and changes in sensation or motor function. Objective measurements of motor and sensory function will be performed on days −2, 28, 82, 112 and 168 during physical visits. Premenopausal women will have a urine pregnancy test performed prior to both treatments on days −2 and 82. A direct, all-hours phone number will be provided to participants, for use in the event of side-effects or adverse events for the duration of the study. All participants experiencing adverse events or side-effects will, if deemed necessary by the investigator, be asked to undergo an additional medical evaluation and if further action is needed, be referred to a neurologist or other relevant specialist for evaluation. Seriousness and causality of a suspected adverse event will be assessed by the investigators and the investigational medical product dossier for incobotulinumtoxin-A will be used to determine if an adverse event can be considered unexpected.

Publication and data statement

All results, positive, inconclusive and negative, will be published in peer-reviewed journals in open-access formats and will be provided to the funders of the study as well as participants. Data will be made available in anonymised form alongside publications.

This post was originally published on