Use of botulinum neurotoxin in the treatment of piriformis syndrome: A systematic review

Abstract

Objective

To determine the pain, functional and adverse outcomes of patients with piriformis syndrome who received botulinum neurotoxin injection, and to determine the optimal dosing of botulinum neurotoxin and choices of modality used during this intervention.

Literature survey

Systematic review of relevant clinical studies published in English language using PubMed/Medline, Embase and CINAHL databases from October 1, 2002 to October 6, 2020.

Methodology

A comprehensive search was performed to identify all studies addressing the treatment of piriformis syndrome with botulinum toxin. Two reviewers independently screened the titles, abstracts, and full texts and extracted data based on a set of predefined inclusion and exclusion criteria. 23 full-text articles were identified of which consensus was achieved for seven articles for data extraction and quality assessment. The qualities and risk of potential bias of the seven studies were appraised using the National Heart, Lung and Blood Institute (NIH) Study Quality Assessment tools for case controls, cohort studies and randomized trials.

Synthesis

Seven studies (n = 152 patients) were included consisting of three randomized controlled studies (RCTs), two case control studies and two cohort studies. The qualities of these studies were: Two good and one fair for the RCTs, fair for both the case controls and one good and fair for the cohort studies. Most studies reported some reduction in pain using various modalities to guide injection (CT, EMG, US or fluoroscopy). However, the included studies were heterogeneous, making it difficult to quantify pain reduction. There was minimal description of other functional outcomes. Botulinum toxin A doses range from 100 to 300U. Mild adverse effects were reported with no medical intervention needed.

Conclusions

There is fair quality of evidence to suggest botulinum toxin is safe to reduce pain in piriformis syndrome. There is insufficient data to quantify pain reduction and to describe other functional outcomes. The optimal dose of botulinum toxin A remains unclear. Modalities to guide botulinum injection into the piriformis muscle remain heterogeneous.

1. Introduction

Piriformis syndrome is a neuromuscular disorder that is presumed to involve the compression of the sciatic nerve at the level of the piriformis muscle.^1,2 In piriformis syndrome, pain is produced by the prolonged or excessive contraction of the piriformis muscle and a sciatic-like pain is commonly described.^2,3 There are many possible etiologies of piriformis syndrome including gluteal trauma, myofascial trigger points and anatomical anomalies.⁴ Clinically piriformis syndrome poses both diagnosis and treatment challenges due to lack of clinical signs which are pathognomonic.^5,6 Use of botulinum toxin injection is not fully understood in piriformis syndrome but more evidence has emerged to indicate the former might reduce muscle hypertonia and inhibit pain substances such as substance P and inflammatory factors.⁷ Specifically, botulinum toxin A has been found to act at the neuromuscular junction to cause muscle paralysis by inhibiting the release of acetylcholine from presynaptic motor neurons.^8,9 In additional, this neurotoxin inhibits substance P release from embryonic dorsal root ganglion neurons and to reduce stimulated release of neuropeptide calcitonin gene-related peptide from trigeminal ganglia neurons.⁷ Besides being effective in reducing pain, botulinum toxin infiltration into the piriformis muscle had been reported to be safe.^10,11 For the last two decades, there were studies which specifically mentioned the diagnosis of piriformis syndrome and its relevant treatment strategies. Botulinum toxin was frequently used as one of the treatment strategies. Despite clinical reviews describing the therapeutic effectiveness of botulinum toxin, there is a lack of a systematic review to evaluate the pain and functional outcome of patients diagnosed with piriformis syndrome who had received botulinum toxin injections.^7,10 Second, even though the list of potential adverse effects from botulinum toxin had been summarized previously, the severity of these adverse effects and whether any interventions were needed to manage these adverse effects remained under-discussed.⁷ Third, the optimal dose of botulinum toxin infiltration into the piriformis muscle remained unclear and the choice of modality to guide the injection process needed more clarity.⁷

A systematic review was conducted to address these gaps. The focus of this systematic review was to evaluate the pain and relevant functional outcomes from botulinum toxin into piriformis muscle rather than clinical or imaging methodologies to diagnose piriformis syndrome. Therefore, this systemic review directly evaluated studies where the presumptive or clinical diagnosis of piriformis syndrome had been made regardless of underlying etiologies. The themes of interest include (1) the therapeutic effectiveness of pain reduction in patients with piriformis syndrome who were treated with botulinum toxins; (2) to determine the improvement in function after botulinum toxin injections; (3) adverse effects (if any) after the injection and if any intervention was needed to manage these adverse effects; (4) dosages of botulinum toxin used during each injection and (5) the various imaging or non-imaging modalities to guide needle placement into the piriformis muscle.

2. Methods

A systematic review of relevant studies was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines.¹²

2.1. Search strategy

PubMed/Medline, Embase and CINAHL databases were searched by the two authors (MHK, YLT) from October 1, 2002 to October 6, 2020. A combination of keywords and controlled vocabularies were used to identify relevant studies. For PubMed/Medline, the terms used for All Fields were “botulinum toxin”, “piriformis syndrome” AND “piriformis muscle”. The MeSH terms were “botulinum toxin”, “muscle syndrome, piriformis” and “piriformis syndrome”. The same terms were applied to Embase and CINAHL databases.

Publications were restricted to those published in the English Language. Decision was made by the authors not to include the etiologies of piriformis syndrome into the search strategies due to the overwhelming lists of possible etiologies.

2.2. Eligibility criteria

Studies were included in the systematic review if they: (1) were randomized clinical trials, case control or cohort studies; (2) evaluating patients diagnosed with piriformis syndrome regardless of underlying etiologies and being treated with botulinum toxin; (3) included studies describing the dosages of botulinum toxins used for each injection; (4) described the various peri-procedural modalities used to guide the botulinum toxin into the piriformis muscle; (5) described pain and functional outcomes after botulinum toxin injection into piriformis muscle; and finally (6) any adverse effects after this procedure and if any intervention was needed to manage the adverse effects.

Studies were excluded if they: (1) consisted of patients who were not diagnosed with piriformis syndrome; (2) mentioned gluteal, hip pain or term such as deep gluteal syndrome without specifically mentioning piriformis syndrome; (3) did not use botulinum toxins in the treatment of piriformis syndrome; (4) consisted of narrative review articles describing anatomical variants of piriformis muscle and the sciatic nerve; (5) were anatomical studies on cadavers; (6) were not published in English; (7) were publications in case reports, case series or book chapters or as poster abstracts; (8) included articles expressing opinions on piriformis syndrome and botulinum toxin in the form of Letter to the Editor or comments in the Editorial; and (9) described buttock or posterior hip pain from lumbar radiculopathies or any other etiologies not related to piriformis syndrome.

2.3. Study screening, data abstraction, and outcomes

The studies from the initial search were screened independently by the two reviewers. Each reviewer screened the titles and abstracts, and the studies were selected based on the inclusion and exclusion criteria. The reference lists of all included studies were screened for additional relevant articles.

The data extracted included year of study, authors, study design, number of patients, dosages of botulinum toxin used and number of injections received per patient, modalities to guide injection, physical therapy and duration applied after the intervention, outcomes to assess pain and function, and adverse effects. The two reviewers met to come to consensus after the second reviewer examined the data extraction table for completeness and accuracy.

2.4. Quality assessment

The National Heart, Lung and Blood Institute (NIH) quality assessment tool was used due to its comprehensive ability to appraise randomized studies, observational cohort and cross-sectional studies in this systematic review.¹² The tools included items for evaluating potential flaws in the study methodology, including sources of bias, confounding, study power, the strength of causality in the association between interventions and outcomes, and other factors. The choice for the applied tool was based on the study design. Specifically, the NIH tools for randomized trials, case control studies and cohort studies comprise of 14 questions, 12 questions and 14 questions to assess the risk of bias respectively. For the randomized trials and cohort studies, two reviewers (MHK and YLT) scored each article on the 14 questions as “yes”, “no”, “cannot determine”, “not applicable” or “not reported”. The NIH quality ratings are “good”, “fair”, “poor”, “not reported”, “cannot determine” or “not applicable”. Similar methods of scoring and rating were applied to the 12 questions found in NIH tools for case control studies. In the case of disagreement, consensus was reached through discussion.

2.5. Statistical analysis

Descriptive statistics were extracted for all reported outcomes which were described individually. Due to the presence of heterogeneous intervention techniques and dosages, as well as variable outcomes measures taken at different time periods, a decision was made not to perform a meta-analysis and results were described qualitatively.

3. Results

The original search yielded 174 results after duplicates were removed. Following the title screen, 23 studies were eligible. 16 studies were further removed after the full text review, leaving seven studies for analysis. The Prisma flow diagram is depicted in Fig. 1. The final seven included studies comprised of 152 patients in three randomized trials, two case-control studies and two cohort studies. These studies span a period of 18 years from 2002 to 2019 and their respective study designs were illustrated in Table 1. Five out of the seven studies had physical therapy after botulinum toxins (see Fig. 2).

Fig. 1.

Prisma flow diagram.

Table 1.

Study design and description of relevant studies which met the inclusion criteria.

First author	Year of study	Study design	Number of patients who received BoNT	BoNT dose(U) per injection into PM	Total number of BoNT per patient	Frequency/interval between each injection	Modalities to guide injection	Needle size and length	Physical therapy after intervention with botulinum toxin	Outcome measures	Adverse effects
H Najdi et al.	2019	Retrospective case control	7	7 patients each received BoNT A (Botox) ranging from 100 to 300	1	-	CT	Not mentioned	Nil	No specific pain scale mentioned; 7 patients who received BoNT had transient pain relief from one week to one month; no other functional outcomes reported.	No adverse effects reported; no intervention needed
Fishman	2017	RCT comparing botulinum toxin with normal saline	56	A 4-site injection of 3 cc containing 300 units of incobotulinum toxin A	1	-	EMG	22 to 25G 3.5 inch	One physical therapy per week for 12 weeks	Reduction of VAS over a period of 12 weeks; reduction of VAS of 0.36 at week 2, 0.39 at week 4, 0.55 at week 6, 0.65 at week 8, 0.55 at week 10, 0.62 at week 12	Five mild adverse effects: injection site pain, flu-like symptoms, wobbly neck in botulinum toxin group; no intervention needed
Al–Al-Shaikh et al.	2014	Retrospective case control study	12	100 of botulinum toxin A (Botox) each into PM and obturator internus	1 to 4	Minimum of 3 months between each injection with a mean of 2.1 injections	Either US or CT	Not mentioned	Nil	Reduction of VAS buttock pain from 7.5 to 3.2 and sciatic pain from 7.04 to 1.7.	No adverse effects reported
Yoon et al.	2007	Prospective cohort study	20	150 of botulinum toxin A	1	Only once	CT	3.5 inch, 22G	Stretching exercises 20 times each day for 12 weeks after treatment	Drop in baseline mean pain score 0f 7.06 to 4.45 at 4weeks, 3.55 at 8 weeks and 3.48 at 12 weeks; Improvement in all subdomain scores of SF-36 4 weeks after treatment	Mild and transient adverse effects; one case of flu-like symptoms for 2days, transient numbness of less than 72 h in the other, lower limb ecchymosis in two patients; no intervention needed
Fishman	2004	Prospective cohort study	26	5000, 7500, 10,000, 12,500 of BoNT B in 4 groups	Seven patients had 2 injection, 19 had 1 injection	Not mentioned	EMG	23G of 3.5 inch	Physical therapy twice weekly for 3months	Overall mean reduction of VAS of 3.6 in 5000 group, 2.12 in the 7500 group, 2.89 in the 10,000 group and 4.51 in the 12,500 group; improvement in FAIR tests in most groups at 12 weeks	One case: monocular blurred vision,; two cases each: gastroesophageal reflux; lump in throat; One case: constipation; two cases: difficulty in swallowing; none needed medical attention; no intervention needed
Fishman	2002	RCT comparing botulinum toxin versus lidocaine and triamcinolone versus normal saline	21	200 of botulinum toxin A with 2 cc saline dilution	1	Only once	EMG detection of H reflex	Not mentioned	Standard physical therapy protocol twice weekly for 12 weeks	VAS assessment into two groups: 50% improvement or more versus less than 50%; 65% improvement observed in the Botox group	No adverse effects reported
MK Childers et al.	2002	RCT comparing botulinum toxin versus normal saline	10	100 of botulinum toxin A	1	Only once	Fluoroscopic or EMG guidance	5.5 inch, 20G needle	Had concurrent physiotherapy and home stretching but changes in physiotherapy program not permitted	Greater reduction in VAS, distress spasm and pain interference with activities over 10 weeks when compared with saline group	No adverse effects reported

Abbreviations: CT, Computed Tomography; FAIR, Flexion, adduction and internal rotation, MRN; Magnetic Resonance Neurography, MRI; Magnetic Resonance Imaging, mm: millimeters,ms:milliseconds, PM: Piriformis muscle; PE, physical examination; QOL: quality of life, RCT, randomized control trial, US, Ultrasound; VAS:Visual Analogue Scale.

Fig. 2.

Techniques of Botulinum Toxin Injections in the included studies.

3.1. Results relating to the study quality appraisal

Overall, two out of the three randomized trials were rated of good quality.^13,14 One had fair quality.¹³ All three randomized trials did not assess if assessors were blinded to the outcomes. Two trials did not report if there was high adherence to the interventional protocols.^15,16 One study had a drop-out rate of 20% or more.¹⁵ Otherwise “yes” ratings were given to the rest of the domains as shown in Table 2. Sample size justifications were not reported in all three studies (see Table 3).

Table 2.

Checklist for assessing quality of randomized trials using the National Institutes of Health (NIH) study quality assessment tools.

	Fishman et al., 2017	Fishman et al., 2002	Childers et al., 2002
1. Was the study described as randomized, a randomized trial, a randomized clinical trial, or an RCT?	Yes	Yes	Yes
2. Was the method of randomization adequate (i.e., use of randomly generated assignment)?	Yes	Yes	Not reported
3. Was the treatment allocation concealed (so that assignments could not be predicted)?	Yes	Yes	Yes
4. Were study participants and providers blinded to treatment group assignment?	Yes	Yes	Yes
5. Were the people assessing the outcomes blinded to the participants' group assignments?	Not reported	Not reported	Not reported
6. Were the groups similar at baseline on important characteristics that could affect outcomes (e.g., demographics, risk factors, co-morbid conditions)?	Yes	Not reported	Yes
7. Was the overall drop-out rate from the study at endpoint 20% or lower of the number allocated to treatment?	Yes	No	Yes
8. Was the differential drop-out rate (between treatment groups) at endpoint 15% points or lower?	Yes	No	Yes
9. Was there high adherence to the intervention protocols for each treatment group?	Yes	Not reported	Not reported
10. Were other interventions avoided or similar in the groups (e.g., similar background treatments)?	Yes	Yes	Yes
11. Were outcomes assessed using valid and reliable measures, implemented consistently across all study participants?	Yes	Yes	Yes
12. Did the authors report that the sample size was sufficiently large to be able to detect a difference in the main outcome between groups with at least 80% power?	No	No	No
13. Were outcomes reported or subgroups analyzed prespecified (i.e., identified before analyses were conducted)?	Yes	Yes	Yes
14. Were all randomized participants analyzed in the group to which they were originally assigned, i.e., did they use an intention-to-treat analysis?	Yes	Yes	Yes
Quality Rating	Good	Fair	Good

Table 3.

Checklist for assessing quality of case control and cohort studies using the National Institutes of Health (NIH) study quality assessment tools.

Case control study	Najdi et al., 2019	Al–Al-Shaikh et al., 2014	Cohort Study	Yoon et al., 2007	Fishman,2004
1. Was the research question or objective in this paper clearly stated and appropriate?	Yes	Yes	1. Was the research question or objective in this paper clearly stated and appropriate?	Yes	Yes
2. Was the study population clearly specified and defined?	Yes	Yes	2. Was the study population clearly specified and defined?	Yes	Yes
3. Did the authors include a sample size justification?	No	No	3. Was the participation rate of eligible persons at least 50%?	No	No
4. Were controls selected or recruited from the same or similar population that gave rise to the cases (including the same timeframe)?	Yes	Yes	4. Were all the subjects selected or recruited from the same or similar populations (including the same time period)? Were inclusion and exclusion criteria for being in the study prespecified and applied uniformly to all participants?	Yes	Cannot determine
5. Were the definitions, inclusion and exclusion criteria, algorithms or processes used to identify or select cases and controls valid, reliable, and implemented consistently across all study participants?	Yes	Yes	5. Was a sample size justification, power description, or variance and effect estimates provided?	No	No
6. Were the cases clearly defined and differentiated from controls?	Yes	Yes	6. For the analyses in this paper, were the exposure(s) of interest measured prior to the outcome(s) being measured?	Yes	Yes
7. If less than 100% of eligible cases and/or controls were selected for the study, were the cases and/or controls randomly selected from those eligible?	Not reported	Cannot determine	7. Was the timeframe sufficient so that one could reasonably expect to see an association between exposure and outcome if it existed?	Yes	Yes
8. Was there use of concurrent controls?	Yes	Yes	8. For exposures that can vary in amount or level, did the study examine different levels of the exposure as related to the outcome (e.g., categories of exposure, or exposure measured as continuous variable)?	No	Yes
9. Were the investigators able to confirm that the exposure/risk occurred prior to the development of the condition or event that defined a participant as a case?	Not reported	Yes	9. Were the exposure measures (independent variables) clearly defined, valid, reliable, and implemented consistently across all study participants?	Yes	Yes
10. Were the measures of exposure/risk clearly defined, valid, reliable, and implemented consistently (including the same time period) across all study participants?	Not reported	Yes	10. Was the exposure(s) assessed more than once over time?	Yes	Yes
11. Were the assessors of exposure/risk blinded to the case or control status of participants	No	No	11. Were the outcome measures (dependent variables) clearly defined, valid, reliable, and implemented consistently across all study participants?	Yes	Yes
12. Were key potential confounding variables measured and adjusted statistically in the analyses? If matching was used, did the investigators account for matching during study analysis?	Yes	Not reported	12. Were the outcome assessors blinded to the exposure status of participants?	Not reported	No
Quality rating	Fair	Fair	13. Was loss to follow-up after baseline 20% or less?	Yes	Yes
			14. Were key potential confounding variables measured and adjusted statistically for their impact on the relationship between exposure(s) and outcome(s)?	Not reported	Not reported
			Quality rating	Fair	Good

Both case control studies had overall fair ratings.^3,6 The study eligibility pertaining to cases receiving botulinum toxins and controls were either not reported or unable to be determined.

For the two cohort studies, one had good rating while the other had fair, and sample size justifications were also not reported.^10,16

3.2. Results relating to pain outcome after botulinum toxin injection

All the seven studies reported improvement in pain outcome as depicted in Table 1. Most of the studies used Visual Analogue Scale (VAS) as their choice of pain assessment. Absolute reduction in pain score or mean reduction in pain were commonly described. Four studies reported at least a three-point improvement in pain score.^5,10,13,16 However, two studies reported improvement in pain using descriptive terms or percentages; Najdi et al. evaluated pain improvement with 7 patients having transient pain relief from botulinum toxin lasting one week to one month while Fishman 2002 reported 65% of the patients who received botulinum toxin having 50% or more pain improvement compared to 32% of patients in the triamcinolone and lidocaine group.^2,15 Only one study by Childers reported less than 3-point change of VAS from baseline for activities, intensity, spasm and distress.¹⁴ For the randomized trials comparing botulinum toxins with other injectable medications, all the three studies demonstrated piriformis injection with botulinum toxin reduced pain to a greater extent than normal saline.

3.3. Results relating to other clinical outcomes other than pain after botulinum toxin injection

Only two studies described other clinical outcomes other than pain reduction. 36-item Short Form Survey (SF-36) was used by Yoon et al. with improvement in all sub-domain scores of the SF-36 at 4 weeks after treatment.¹⁰ Childers et al. evaluated four categories of distress, spasm, pain interfering with activities over 10 weeks relating to VAS and observed that botulinum toxin A was superior to saline injection only in the category of interference with activities.¹⁴

3.4. Results relating to dose of botulinum toxin used and administration of botulinum toxin

A total of 152 patients within the seven included studies received botulinum toxin. Six studies used botulinum toxin A while the study from Fishman used botulinum B. There were wide ranges of botulinum toxin A (100–300 U) and B (5000–12,500 U) used for the included studies. Specifically, for each injection, those studies which used botulinum toxin A reported a range of 100–300 U to each patient per visit. Likewise, studies which used botulinum toxin B reported a range of 5000 to 12,500 U used per visit. The number of botulinum toxin injection per patient was also heterogenous within studies and between studies.^2,16 In most of the seven studies, only one injection was administered to each patient. Nevertheless, the study by Al–Al-Shaikh et al. described up to 4 injections per patient while the study by Fishman 2004 had seven patients who received two injections.^5,16

Results relating to modalities to guide the botulinum toxin infiltration.

All seven studies used modalities such as ultrasound (US), electromyogram (EMG), computed tomography (CT) or fluoroscopy to guide the botulinum toxin injections. EMG was the modality most commonly described followed by CT. Only one study used either US or CT guidance.⁵ The study by Childers used either US or fluoroscopic guidance to guide botulinum toxin into the piriformis muscle.¹⁴

4. Results relating to adverse effects from botulinum toxin injection

Four studies did not report any adverse effects after botulinum toxin infiltration.^2,5,14,15 For the remaining three studies which did report, the adverse effects described were flu-like symptoms (two cases), transient numbness (one case), ecchymosis (one case), monocular blurred vision (one case), gastro-esophageal reflux (two cases), constipation (one case) and wobbly neck (one case).^10,13,16 No interventions were needed to manage these adverse effects. There was no description with regards to whether these adverse effect reported were dose-related.

4.1. Physical therapy administered after intervention with botulinum toxin

Five out of the seven included studies indicated provision of physical therapy.^10,13, 14, 15, 16 Amongst the five studies, four provided physical therapy sessions for 12 weeks while the remaining one did not specify the duration of the exercises.¹⁴

5. Discussion

This systematic review focused on studies which specifically mentioned piriformis syndrome rather than a review of the entire spectrum of conditions under the disease deep gluteal syndrome.¹⁷ The purpose of this article was to give practitioners from the Orthopedic, Pain and Rehabilitation arena further insights into evidence revolving around the use of botulinum toxins in piriformis syndrome in terms of clinical outcomes, safety and modalities to guide the intervention.

One of the main aims of this systematic review was to establish the pain outcome of patients who received botulinum toxin injection. This review suggests there was a mean reduction in the pain score of at least three points in patients diagnosed with piriformis syndrome treated with botulinum toxin. However, given the context of heterogeneity among the included studies, it is difficult to interpret the validity of the absolute pain reduction after the intervention. Nevertheless, we postulated that satisfactory pain reduction could still be achieved if botulinum toxin injection is included as part of the multidisciplinary approach to the overall pain management of this group of patients. Specifically, practitioners managing patients with piriformis syndrome should consider botulinum toxin as one of the injectable modalities for treatment in addition to traditional methods of pain relief such as slow stretch exercise, massage, heat, or ultrasound treatment.¹⁰ With improvement of pain, the physical function of patients may improve with adequate physiotherapy. Majority of the studies did provide physiotherapy for a duration of 12 weeks but each study in our review did not describe its exercise protocol in detail. Neither were there mention of specific outcome evaluation such as outcomes relating to ambulation, joint range of motion or activities of daily living. Only one study in our review attempted to distinguish pain levels with respect to interference in activities, but it was unclear precisely which physical activities were evaluated.¹⁴ Therefore, it was difficult to conclude if botulinum toxin injection in patients with piriformis syndrome followed by physiotherapy led to improvement in specific functional outcomes due to paucity of information available. As part of the comprehensive rehabilitation program for piriformis syndrome, we suggest that future research studies consider inclusion of appropriate functional scales for pre- and -post botulinum toxin surveillance and describe in more details their rehabilitation exercise protocol. Examples of specific functional scales reported previously included Lower Extremity Functional Scale questionnaire, hip range of motion, and Oswestry Disability Index.^18,19

In terms of medications used to inject the piriformis muscle, there were three studies in this systematic review comparing the pain outcomes between botulinum toxin and other injectable solvents such as normal saline or steroids with lignocaine. Only two randomized trials compared botulinum toxin with normal saline while the other study compared botulinum toxin with normal saline and steroid with lignocaine injections. Given the focus of this study was to evaluate the pain outcome with botulinum toxin injection, we did not further evaluate if botulinum toxin was superior to normal saline or steroid with lignocaine injection.

Nevertheless, given the insufficient data and the limited number of studies comparing different medications in this review, it might be difficult to make any robust conclusion anyway. The second aim of this systematic review was to identify and map out the peri-procedural practice by practitioners during botulinum toxin infiltration in patients diagnosed with piriformis syndrome. The piriformis muscle courses deep to the gluteus maximus with physiologic variations of the sciatic nerve innervating and penetrating the piriformis muscle. Hence it was not surprising that all the included studies used at least one type of imaging or non-imaging modality to guide injection.19, 20, 21 In the seven studies reviewed, EMG was most commonly mentioned followed by CT guidance of botulinum toxin injection. None of these studies justified the reasons for choosing a particular modality to guide the process of botulinum toxin infiltration. EMG machine and CT imaging services are generally available in the hospitals. For US, we postulate that not all rehabilitation or pain units around the world have access to US machines and availability of practitioners with sonography skillsets. This could be two of the many reasons why US is less mentioned than EMG or CT. At present, there are many published trials comparing different modalities to guide needle placement into the piriformis, however there is lack of systematic reviews to compare the efficacy of different injection modalities for botulinum toxins.^22,23 We hope the current mapping of various choices of modality guidance for botulinum toxin injections can spur future studies in the direction of comparing therapeutic effectiveness of botulinum toxin between the use of different modalities.

There was a wide dose range of botulinum toxin A used from 100U to 300U without clear justification of the choice of dosage used in piriformis muscle. In general, dosage recommendations are available for botulinum toxin injections into the upper and lower limb muscles. Specifically, the European Consensus Table on the use of botulinum toxin type A for various muscle groups with respective dose recommendation was more for adult spasticity management rather than pain, and piriformis muscle was also not mentioned in this consensus.²⁴ Based on our current knowledge, there is no dosage guide for pain management using botulinum toxin for patients suffering from piriformis syndrome. There is a need to explore the optimal dose range per injection for pain management in piriformis syndrome for various reasons such as therapeutic effectiveness and cost. We suggest future randomized control trials include high quality methodologies comparing pain and functional efficacy of different doses of botulinum toxins with a control group such as the use of normal saline.

The adverse effects from botulinum toxins were all reported to be mild and transient without the need for any specific interventions.¹⁰ The adverse effects described in all the included studies assist in guiding future practitioners during consent taking prior to botulinum toxin injection. From the results of this systematic review, it is reasonable to conclude that botulinum toxin injections into the piriformis muscle is generally safe.

In terms of study quality, more than half of the seven studies (n = 4) had fair quality which made the overall results susceptible to some bias but not sufficient to be deemed invalid.¹² The remaining three studies had good quality ratings which indicate low risk of bias. We conclude it is still plausible to make two recommendations using Strength of Recommendation Taxonomy (SORT) where an algorithm was used to assign the strength of the following recommendations using Grade A,B or C.²⁵

The two recommendations can be summarized as follow:

• 1.Botulinum toxin injection is a reasonable treatment option to reduce pain intensity in patients diagnosed with piriformis syndrome; however, it is unclear regarding the absolute pain reduction given the heterogeneity of available studies (Grade B)
• 2.Botulinum toxin injection for the treatment of pain in piriformis syndrome is relatively safe with mild adverse effects which generally needed no further intervention. Practitioners should counsel patients who intend to undergo this procedure appropriately prior to the procedure. (Grade B)

In terms of functional outcomes after botulinum injection, there is insufficient evidence from this systematic review to conclude that injection of botulinum toxin into patients with piriformis syndrome led to specific functional improvement in joint range of motion, ambulation and activities of daily living. The optimal dose and number of botulinum toxin injections remain unclear as studies included are only of fair quality, subject to some bias and are heterogeneous in the number of injections and injection techniques. For modalities to guide the process of botulinum toxin injection, there is insufficient data to suggest one modality is superior to another although all studies use some form of imaging guidance or EMG during the procedure.

5.1. Limitations

The heterogeneity of studies included in this systematic review limited our ability to proceed further with meta-analysis and results were summarized qualitatively. On reviewing all included studies, different types and dosages of botulinum toxin, with different number of visits for botulinum toxin injection were described. In addition, the application of physical therapy pre- and post-botulinum toxin treatment were varied, and the specifics of the physical therapy were not well described. As such, these were potential confounding factors affecting pain outcomes.

This review is also limited by the low number of included studies. In addition, the overall results from included studies are of fair quality which suggested some bias and firm conclusions cannot be made for functional outcomes, optimal botulinum dosing, and modalities to guide injection. In addition, as this review was conducted with studies of patients who were already diagnosed with piriformis syndrome, there was a lack of quantitative analysis of the criteria used by each included studies to derive at the diagnosis of piriformis syndrome. We also acknowledged that deep gluteal syndrome was not included in the search term. Given that piriformis syndrome is considered as one of the conditions under deep gluteal syndrome, articles mentioning deep gluteal syndrome with subset of patients diagnosed with piriformis syndrome were not included in this systematic review.¹⁷ Conclusions from this review were also limited to the diagnosis of piriformis syndrome and we could not extrapolate the findings to other diseases under the umbrella of deep gluteal syndrome.

6. Conclusion

From this systematic review, there was fair quality of evidence to suggest botulinum toxin has a potential role to reduce pain in piriformis syndrome. There was, however, insufficient evidence to establish the reduction in pain quantitatively. The procedure of botulinum neurotoxin infiltration was generally safe with mild adverse effects reported. When adverse effects occurred, no clinical interventions were needed. For functional improvement, there was insufficient evidence to describe improvement in specific functional outcomes after the procedure. The botulinum toxin A dosage used generally ranged from 100 to 300 U but the optimal dose remains unclear due to heterogeneous practice in the number of injections and injection techniques. The choice of modality to guide needle placement of botulinum into the piriformis muscle remains heterogeneous and more future robust studies are needed to define this area.

Funding sources

No funding received or any other support.

Declaration of competing interest

We have no conflict of interest.