TPI Frequency
There is no general consensus regarding the optimal frequency of TPI.1,12 Repeated injections are generally not recommended if two or three prior sessions have been unsuccessful.8,13
The 2021 American College of Occupational and Environmental Medicine (ACOEM) guidelines13 recommend at least 3-4 weeks between injections. They also supported that repeated injections are not recommended if no improvements are seen after the second session. The 2001 American Society of Interventional Pain Physicians (ASIPP) practice guidelines1 recommend at least one week, but preferably two weeks, between TPIs during the initial diagnostic phase. And supported a maximum of 4 sessions per year or rolling 12-month period. If a patient is in the therapeutic phase, the society recommendation is that injections should be at least two months apart, with a maximum of 6 sessions per 12 months. However, it is important to note that these ASIPP recommendations were adapted from epidural steroid injection protocols. The application of epidural steroid injection protocols to TPIs lack a scientific basis or empirical justification, as the targets, techniques, injectate type, and dosing differ substantially. Additionally, there is no clear distinction between diagnostic and therapeutic phases in the current literature for TPI.
There is no clear consensus from the literature or society guidelines as to the exact frequency for TPI. However, there is general agreement in the literature and these guidelines that some frequency limitation should exist for how often a person receives TPIs.1,8,13 It is widely reported in the literature that TPI can be associated with adverse events such as subcutaneous hemorrhage, dizziness, muscle soreness, transient hypertension, insomnia, coldness, injection site pain or burning, paresthesias, cervical muscle spasm, localized hematomas, and bleeding.14 Limitations to the frequency of TPI serve to ensure patient safety and avoid adverse events and harm from repeated TPI sessions.
Recommendations from the literature and society guidelines are variable. The 2001 ASIPP guidelines offer two recommendations of either a maximum of 4 TPIs per 12 months during the diagnostic phase, or a maximum of 6 TPIs per 12 months during the therapeutic phase. The 2021 ACOEM guidelines state that repeated injections are not recommended if no improvements are seen after the second session. The SMEs from the multi-jurisdictional CAC, representing a consensus of the medical community, agree that typically injections are not being administered more frequently than every three months.7(pp21-29) We also conducted a separate survey of SMEs within our jurisdiction, and when asked about the optimal frequency of TPIs, the majority of responses also stated approximately 3 months between injections. The 3-month interval between injections is recommended by the SMEs from the CAC and our SME survey.
To minimize potential adverse effects from repeated injections and based on the best available evidence and expert consensus (in the absence of formal society guidelines), no more than four (4) trigger point injection (TPI) sessions are considered reasonable and necessary within a rolling 12-month period when used as adjunctive treatment for myofascial pain. In alignment with ACOEM guidelines, routine administration of four or more TPI sessions per year would not be expected. Only under exceptional clinical circumstances should a patient require the maximum of four (4) sessions as part of a multimodal therapeutic program.
US and Imaging Guidance for TPI
There is currently no guideline supporting the routine use of US or imaging guidance for TPI. The diagnosis of a trigger point itself is reliant on physical examination with direct palpation and identification of the trigger point, which is a discrete, focal, hyperirritable area located within a taut band of skeletal muscle. Direct palpation is generally performed for diagnosis and identification prior to injection.15 The use of US has been investigated to determine if it would improve the safety and accuracy of needle placement.12,16-18 It has been theorized that US may help differentiate normal tissue compared to myofascial trigger points and minimize complications such as vascular injury, pneumothorax, and injury to adjacent structures. However, current society guidelines point to overall weak evidence and recommend that TPI can be conducted based on palpation alone. Further research is needed to determine the long-term efficacy and safety of image guidance compared to direct palpation and landmark-based approaches.10,16,19,20
There have been studies which have identified the US characteristics of myofascial trigger points (MTrPs). Sidkar et al21 evaluated nine patients with MTrPs in one or both upper trapezii in an exploratory descriptive study. Participants underwent a physical examination, pressure algometry, and three types of US imaging: grayscale (2D US), vibration sonoelastography (VSE), and Doppler. The US imaging team was blinded to the clinical findings to reduce bias. Results showed that MTrPs appeared as focal, hypoechoic regions on 2D US. Active MTrPs were also more likely to have higher blood flow score (BFS) than latent sites (P<0.021). Authors concluded that US imaging techniques were able to visualize and distinguish myofascial tissue with trigger points from normal tissue. However, the certainty of the findings is limited due to its preliminary nature of results, limited generalizability due to its small sample size, lack of control group, lack of standardization for operator techniques, possible observer bias, and lack of inter- and intra-observer variability assessment.
A systematic review evaluating the benefit and safety of US-guided interventions for myofascial trigger points included 11 RCTs.22 While some evidence supported pain and functional improvements with minimal or zero adverse events, the overall effectiveness of US-guided interventions compared to direct palpation remains unclear for MPS treatment. The studies examined comparisons between US-guided and blinded interventions (2 studies, n=174), head-to-head comparisons of different US-guided techniques (8 studies, n=483), and US-guided vs non-interventional therapy (1 study, n=21). Risk of bias was assessed by the author using the RoB 2.0 tool, which identified high risk in seven studies, some concerns in three, and low risk in only one. Most issues stemmed from problems with the randomization process and selection of reported results. There is a low certainty of evidence as this review is limited due to small sample sizes, clinical heterogeneity (intervention, comparison, target intervention sites, varying US-guided techniques), search parameters, and ratings on clinical relevance.
Kang et al23 conducted a preliminary pilot study involving 41 patients diagnosed with trapezius MPS at a medical institution in Korea. Patients were randomly assigned to receive either US-guided TPI with shear wave elastography (SWE) (n=21) or blind TPI with SWE (n=20). Patients in both groups showed improvements in pain scores at 4-week follow-up, measured by the visual analogue scale (VAS) scores, Neck Disability Index (NDI) scores, and Shoulder Pain and Disability Index (SPADI). The US-guided TPI with SWE group showed significantly greater improvements in VAS scores, NDI scores, and SPADI (P=0.003, 0.012, and 0.018, respectively) comparatively to the blind TPI group. No adverse events were reported. Study limitations include its small sample size, short follow-up, preliminary nature, lack of blinding, and a high risk of bias (RoB 2.0). Additional limitations include selection bias due to recruiting participants from a single rural institution, limited generalizability given that it is non-U.S. based with a young population (mean age 44.2 years), and the absence of other treatment arms such as US-guided injection without SWE.
A prospective study by Rha et al24 recruited patients with active myofascial trigger points (MTrP) in their upper-trapezius (n=41, mean age 51.8 years) and lower-back muscles (n=62, mean age 58.6 years) at a university hospital in Korea. While participants received TPI, the local twitch response (LTR) was observed by investigators, one using US and the other using visual examination, with effectiveness assessed prior to and immediately post injection. Both methods were effective at detecting all LTRs in the upper trapezius muscles, however, for the deeper lower back muscles, LTRs were more often detected by US, resulting in a significantly higher detection rate than upper trapezius (OR, 1.26; 95% CI, 1.22–1.30; P<0.001). Ultrasound was more effective at detecting the LTR in deeply located muscles compared to superficial ones (OR, 3.01; 95% CI, 2.40 –3.77; P<0.001). The group who had a LTR during the TPI experienced significantly more pain alleviation than those who did not, with VAS scores post injection (P<0.001 for both lower-back and upper-trapezius muscles). Limitations of this study include generalizability issues to the Medicare population (non-U.S. based, younger age), lack of long term follow up, not comparing US and palpation for LTR detection, not providing a clear definition of LTR on US, and the absence of inter- and intra-reliability assessment of LTR detection on the US.
A technical report by Botwin and colleagues25 on US-guided injection techniques describes the potential difficulty of palpating trigger points in obese patients and recommends image guidance to improve accuracy of needle placement and to avoid injecting adipose tissue, which would minimize effectiveness of the TPI. The report also notes that image guidance could help prevent potential complications such as pneumothorax or inadvertent intrathecal injection. However, the primary limitation is that this is a technical commentary rather than a clinical study, and its recommendations have not been evaluated in real patient studies.
A few case reports describe patient experiences with US-guided TPI. The patients ranged from 35-58 years old and showed pain reduction and improved range-of-motion following treatment. A patient with chronic MPS of the upper trapezius received US-guided TPI once weekly for 4 weeks, resulting in significant pain reduction and improved cervical spine range of motion.26 Another case involved a patient treated with US-guided 3-in-1 TPI for a persisting pain related to MPS, with only slight residual pain one month later.27 A third patient, referred after failed responses to relaxants, anti-inflammatory agents, and prior TPI's experienced immediate pain relief (>80%) and improved range of motion following US-guided TPI.28 Limitations include generalizability issues (small sample, young age, non-U.S. based), lack of long term follow up, and the nature of case reports and its low certainty of evidence.
A 2024 guideline by the American Society of Regional Anesthesia & Pain Medicine in collaboration with other societies3 regarding adult chronic pain interventions and trigger points provided the following recommendations related to image guidance and the use of US for performing TPI:
“TPI can be conducted based on palpation alone or with US which may improve accuracy of injection. Grade C”
“Clinicians may consider US guidance for TPI conducted in areas near high-risk tissues (risk of neural, vascular, pulmonary, or visceral injury) or in trigger points located in deeper anatomic locations. Grade C”
A Grade C recommendation from the United States Preventive Services Task Force (USPSTF) is defined as “selectively offering or providing this service to individual patients based on professional judgment and patient preferences. There is at least moderate certainty that the net benefit is small”.
Overall, US-guided TPI demonstrate potential benefits in terms of visualization, improved pain outcomes, and reducing adverse events. However, current evidence remains insufficient to demonstrate improved outcomes compared to landmark-based and direct palpation based TPI. Randomized controlled trials are limited by small sample sizes, short follow-up periods, methodological heterogeneity, and varying risks of bias. Remaining studies offer low certainty of evidence and are limited to non-blinded designs, retrospective data, or case reports. Generalizability to the Medicare population is limited due to younger, non-USA based study populations. No studies evaluating the efficacy of other image guided modalities for TPI, such as CT, MRI, or fluoroscopy were identified. For US-guided TPI, current guidelines provide a grade C recommendation, indicating moderate certainty that potential net benefit is small.3 The diagnosis and identification of trigger points is based on physical examination, and there is no clear standardization in US techniques. There is a lack of data to support any long-term benefit to using US-guided or image-guided TPI over landmark-based techniques. There is a lack of data to support that US improves outcomes for TPI, and therefore the use of US or other imaging modalities for performing TPI is not considered reasonable and necessary.
Injectates
Overall, a wide variety of injectates for TPI have been studied across trials. No single injectate has shown clear superiority,29 and the certainty of evidence remains low.
All Injectates
A systematic review assessed the effectiveness of TPI for myofascial neck and back pain.30 Ovid Medline, Embase, Cochrane Library, and Scopus were searched from database inception to April 2020 for randomized trials, cohort studies, and case-control studies. Fourteen studies were included (13 RCTs and two cohort studies, including one article with both). Randomized controlled trials assessed effectiveness and cohort studies contributed data on the harms of TPIs. Pain relief was measured using NRS (Numerical Rating Scale) and VAS (Visual Analog Scale), along with functional outcomes and harms. The risk of bias (RoB) was assessed using the Cochrane Back Review Group (CBRG) 12 item checklist which evaluates selection bias, performance bias, detection bias, attrition bias, and selective outcome reporting bias. The CBRG checklist provides Yes/No/Unknown as responses across 12 bias domains, generating an overall quality score. Out of 13 randomized studies, RoB assessment indicated five as high quality (as defined by the authors as having low RoB in at least 8 out of 12 items) and eight moderate quality (low RoB in 4–7 items). Due to heterogeneity in study populations and outcome reporting, the authors did not perform pooled comparisons for pain and functional outcomes. Instead, results were provided as a descriptive summary. Six studies compared TPI of Botulinum toxin A with normal saline, five of which used Onabutulinum toxin A and one Abobotulinum toxin A. In two studies, greater pain improvement was observed in the Onabotulinum toxin A group at short (7 days to less than 6 weeks) and intermediate (6 weeks to less than 3 months) follow up compared to NS. One Abobotulinum toxin A study reported superior pain relief at short, intermediate, and long-term follow ups (3 months to less than 6 months). The remaining three Onabutulinum toxin A studies found that both groups experienced pain relief over time, but no statistically significant differences were found between BTX and NS. No differences in effectiveness were found in two studies comparing Onabotulinum toxin A to local anesthetic, one study comparing Onabotulinum toxin A to methylprednisolone (steroid), and one comparing Onabotulinum toxin A to DN. Another study compared Ozone to Lidocaine and DN, with all three groups demonstrating significant short-term improvement in pain, with no statistically significant differences between them. Two studies compared sterile water to NS and reported no significant differences in short term pain reduction. However, one study observed greater pain relief with sterile water at intermediate, long, and longest term (>6 months) outcomes. Lastly, one study compared Tropisetron to a local anesthetic (Prilocaine) and found no significant differences in short term pain improvement. The trial was discontinued in non-responding patients, resulting in inconclusive findings.
Due to the mixed results of the studies included in the systematic review,30 there is no definitive evidence which demonstrates that any single injectate is superior to another in terms of pain relief and functional outcomes. The aforementioned studies comparing injectates showed improvements in pain scores in all injectates studied, with no significant differences between the injectates. No consensus was found regarding optimal needle size, injectate doses, and volumes. The systematic review did note that adverse events included transient flu like symptoms, local muscle weakness, non-severe skin redness in the Onabotulinum toxin A group and muscle soreness in the Abobotulinum toxin A group. All injectate modalities had injection soreness. This review is limited by its inconclusive findings, lack of comparisons for commonly used LA vs steroid injections, and inability to determine whether imaging guidance influences outcomes. Given these limitations, the certainty of evidence is low and insufficient to determine the clinical utility of individual injectates.
Corticosteroids
Corticosteroids are synthetic steroid hormones (also called glucocorticoids or steroids) which are used as prescription medications to reduce inflammation or suppress the immune system. Within the context of TPI, corticosteroids are intended to reduce inflammation and the presence of inflammatory markers surrounding the trigger points.3
There is limited evidence to support the use of corticosteroids as an injectate in TPI, and society guidelines which do not support the use of corticosteroids for these injections.
The American Society of Regional Anesthesia & Pain Medicine (ASRA), in collaboration with other societies, published a 2024 guideline on corticosteroid usage for adult chronic pain interventions.3
The guideline states that the addition of corticosteroid to a local anesthetic does not result in increased benefit that outweighs the potential risks (Level of certainty: moderate) and recommended that "The use of local anesthetic alone should be considered for TPI Grade B."3
The United States Preventive Services Task Force (USPSTF) defines a Grade B recommendation as one where there is high certainty that the net benefit is moderate, or there is moderate certainty that the net benefit is moderate to substantial. This definition provides helpful context for interpreting the Grade B rating used in the ASRA guideline.
The guideline further notes that while corticosteroids are commonly used in TPIs, evidence does not support their benefit on treatment success, and their use may increase the risk of infection and cause systemic steroid related adverse effects.
A randomized controlled non-inferiority trial enrolled 48 patients with MPS (mean age 42.5 years) diagnosed in two urban hospital emergency departments (ED) in Texas.31 Participants were blinded and randomized to receive TPI with either a conventional active drug mixture (CADM; lidocaine + triamcinolone acetonide; n=25) or a normal saline (NS; n=23). Pain relief was assessed using a 0-10 NRS by a blinded physician at multiple time points: ED arrival, immediately before and after treatment, before discharge, and at 2 two week follow up. "The mean pain scores were as follows: immediately before TPI, 7.59 (NS) and 7.44 (CADM); immediately after TPI, 2.22 (NS) and 1.76 (CADM); prior to discharge, 1.52 (NS) and 1.76 (CADM). At two week follow-up, the mean pain scores were 4.29 (NS) and 4.14 (CADM)." Both groups experienced significant pain reduction after treatment, with similar outcomes at two weeks. The authors concluded that the similarity in pain scores between these groups at the two week follow-up point suggest that the injectate used for TPI (NS or CADM) was irrelevant for pain control. No adverse events were reported. Risk of bias as per the RoB 2.0 tool determined a low risk of bias. Limitations include small sample size, short follow up, ED setting, heterogeneity in pre-injection analgesic use, lack of compliance monitoring for muscle stretching, and a younger population limiting generalizability.
A RCT enrolled 45 patients with myofascial pain (ages 18-65) and at least one trigger point in the orofacial or cervical region linked to headaches at a Brazilian dental school.32 Patients were randomly assigned to one of three groups: DN (control), lidocaine, and lidocaine + corticosteroid. All patients were assessed at baseline, ten minutes, one week, four weeks, and 12 weeks post injection for pain intensity, post injection sensitivity, duration of relief, and ibuprofen use, using the modified Symptom Severity Index (SSI) and patient diaries. All three treatment groups showed significant improvements in pain scores over time (P<0.001). The lidocaine + corticoid group demonstrated fewer days of post injection pain (P=0.0232) and a trend towards reduced ibuprofen use, though differences were not statistically significant between groups. No significant differences were found between groups for injected site pain intensity. No adverse events were reported. Study limitations include a small sample size, short follow up duration, lack of blinding, no true placebo group, limited methodological and demographic detail, some concerns for bias (RoB 2.0), and a non-U.S. based population limiting generalizability. Additional limitations include standardized but restricted injection sites, absence of guidance on stretching, home therapies, and parafunctional habits that may have influenced symptom relief.
At this time, there is a lack of guideline endorsement and insufficient evidence to support the use of corticosteroids as an injectate for TPI and therefore will not be considered reasonable and necessary.
Botulinum Toxin
The 2010 American Society of Anesthesiologists (ASA) Task Force and the American Society of Regional Anesthesia and Pain Medicine (ASRA) practice guidelines for chronic pain management recommend against the routine use of botulinum toxin for myofascial pain.2 This is based on RCTs comparing botulinum toxin type A to saline placebo, which yielded equivocal findings (Category C2 evidence).
A systematic review and meta-analysis compared the effectiveness of LAs and botulinum toxin-A (BTX-A) on pain intensity in patients with myofascial pain.14 Pain outcomes were measured using the VAS and the Neck Pain and Disability Scale (NPAD). Databases including EMBASE, CENTRAL, and Medline were searched from database inception to May 2017 for RCTs, controlled trials, and randomized trials, identifying 33 studies in which 18 assessed LA, 16 assessed BTX-A, and one evaluated both. Risk of bias assessment indicated that 18 of 33 studies had low risk of bias on 4-6 of the six items assessed. Qualitative assessment found that both LA and BTX-A showed inconsistent effectiveness in reducing pain across all follow-up periods. However, meta-analyses revealed LA to be more effective at reducing pain compared to BTX-A, with multiple injections sessions proving more beneficial than single sessions. Reported adverse events (AEs) varied across studies. Local anesthetic AEs included subcutaneous hemorrhage, dizziness, muscle soreness, transient hypertension relieved after brief rest, insomnia, coldness, injection site burning, paresthesia’s, injection site pain, cervical muscle spasm, localized hematomas, and minimal bleeding. Adverse effects associated with BTX-A included transient pain and weakness, muscle soreness, minor chewing discomfort, surrounding injection site redness, feeling feverish, injected site tightness, shoulder stiffness, fatigue, headache, heaviness, and numbness. The certainty of evidence is low to moderate, primarily due to high heterogeneity among the included studies, stemming from differences in anesthetic types, study design (RCT, controlled trial, randomized trial), injected muscles, adjunct treatments, and comparator groups. Additionally, this review is distinct from the broader systematic review by Debrosse et al30 assessing all injectates. While five studies overlapped between the two, they were re-analyzed in this review to address a focused comparison between BTX-A and LA.
A RCT conducted at a school of dentistry in Brazil enrolled 45 patients (aged 18-65) with myofascial pain associated with headaches reproduced by activating at least one trigger point.33 Participants were randomized to receive TPI with DN (group 1), lidocaine (group 2), or botulinum toxin (group 3), with no additional therapeutic instructions. Patients were evaluated over a 12-week period for pain intensity, frequency and duration, local postinjection sensitivity, time and duration of relief, and use of rescue medication (ibuprofen 200 mg). All three groups demonstrated significant improvements across all measured parameters (P<0.05), except for the use of rescue medication and post injection sensitivity, where the botulinum toxin group demonstrated superior results. No adverse events were reported. Study limitations include the small sample size, lack of blinding, lack of detailed patient demographics, high risk of bias (RoB 2.0), and the non-U.S. based population limiting generalizability.
A single-center double-blind randomized crossover study enrolled 18 patients (mean age 51.1 years) with MPS involving trigger points in the neck, shoulder girdle, hip girdle or back regions.34 Participants were randomized to begin treatment with either bupivacaine or BTX-A injections targeting up to eight of their most painful trigger points. Subjects were followed until their pain returned to 75% or more of their pre-injection pain for two consecutive weeks, followed by a two week wash-out period before crossover to the alternate injection. Both treatment modalities significantly reduced pain compared to baseline (P=0.0067). However, no significant differences were found between the BTX A and bupivacaine groups in duration or magnitude of pain relief, function, or satisfaction. Adverse events included regional weakness and dizziness after bupivacaine injections, numbness, limb coldness, sore throat, and nausea after BTX A injections. Limitations of this study include the small sample size, short follow up period, crossover design, selection of patients already responsive to bupivacaine, home exercise program potentially influencing results, high risk of bias (RoB 2.0), and a younger, non-U.S. based population limiting generalizability.
A retrospective study reviewed clinical records of 82 patients (mean age 35.6 years) with myofascial trigger points in the masseter muscle treated at a hospital in Turkey.35 Patients were divided into three treatment groups: group I LA injections (n=27), group II BTX (n=26), and group III platelet-rich plasma (PRP) injections (n=29). Pain at rest and while chewing was assessed, and all groups showed improvement at 1- and 3-months, with LA and BTX showing greater improvements than PRP at month three. Improvements in VAS pain, jaw functional limitation scale (JFLS), and Oral Health Impact Profile questionnaire (OHIP-14) values were significantly better in BTX group than LA at three months (P=0.009; P=0.004; P=0.002). Significant improvement in VAS pain, JFLS, and OHIP-14 (P=0.008; P<0.001; P<0.01) values were recorded only in the BTX group at six months. All injection groups were effective at treating symptoms of trigger points in the masseter muscle at 1- and 3-months, with LA and PRP providing relief for a shorter period compared to BTX. No adverse events were reported. Limitations of this study include its retrospective design, small sample, lack of a control group, limited generalizability due to younger, non-U.S. based population, and failure to account for etiologic factors that may have influenced treatment outcomes.
While some studies have shown short term benefits, the overall evidence for botulinum toxin is limited due to inconsistent results, small samples, and lack of generalizability to the Medicare population. Additionally, societal guidelines recommend against its routine use for myofascial pain.2 There is insufficient evidence at this time to support botulinum toxin for TPI, therefore it is not considered reasonable and necessary and will not be covered.
Saline
Saline is a mixture of sodium chloride (salt) and purified water. The use of sterile normal saline (NS) in TPI has been attributed to its minimal side effects and overall safety compared to conventional active medications.15
A double-blinded RCT conducted in Switzerland enrolled 107 patients (average age 48 years) with MPS and masticatory trigger points of the head and neck.36 Participants were randomized to receiving injections of bupivacaine (n=40), lignocaine (n=33), or saline (n=34). Pain ratings were assessed through detailed interviews rather than VAS. No significant differences were found between groups regarding symptom reduction or therapeutic benefit. Overall, 53 patients (49%) became symptom free, 40 (38%) reported substantial relief, and 14 (13%) experienced no change. The number of injections required to reach end of treatment did not differ significantly among groups. Three patients in bupivacaine and two in the lignocaine group experienced transitory facial palsy during treatment. Study limitations included the short follow up period, missing data for one participant, outcome measuring methods, high bias risk assessed by the RoB 2.0 tool, and limited generalizability due to the younger, non-U.S. based population.
A randomized double-blind crossover study enrolled 15 patients with MPS who were regularly receiving lumbar or cervical trigger-point injections with bupivacaine at a chronic pain clinic in Arizona.37 Each patient received all three types of injections with bupivacaine, etidocaine, and physiological saline in different treatment sessions spaced 1-3 weeks apart. Patients were asked to subjectively rate six pain-related variables on a VAS: average pain, percentage of time pain felt, effect of pain on average physical activity, muscle tension, effect of pain on sleep, and effect of pain on mood. Ratings were collected before treatment and at 15 minutes, 24 hours, and seven days after each injection. Overall, TPIs with bupivacaine and etidocaine were generally preferred over saline across several pain-related measures. No adverse events were reported. Study limitations include the small sample size, short follow up period, crossover design, inclusion of trigger point experienced patients, lack of demographic information, limited detail on methodology, and some concerns for bias (RoB 2.0).
A double-blinded RCT enrolled 80 patients (mean age 40.4 years) with MPS in the upper trapezius at an outpatient clinic in Thailand.38 Patients were randomly assigned to receive either an US-guided interfascial injection with physiological saline (group 1, n=40) or US-guided TPI with lidocaine (group 2, n=40). The interfascial injection was performed at the area beneath the trigger point. Pain was assessed ten minutes post-procedure using the VAS, with the lidocaine group demonstrating significantly greater immediate pain reduction (P=0.037). However, no significant differences between saline and lidocaine groups were observed at the 2- and 4-week follow-up. No adverse events were reported, and patients only experienced mild post injection soreness and dizziness. Study limitations include small sample size, short follow up period, lack of placebo group, multiple treatment variables between groups (saline vs lidocaine, injection volumes, needle techniques), high risk of bias (RoB 2.0), and limited generalizability due to young, non-U.S. based patients.
A double-blinded RCT enrolled 70 female Fibromyalgia patients aged 23-70 from a hospital in Brazil to evaluate the effects of TPI in the temporal muscles for those with masticatory MPS and headache.39 Patients were assigned to received either saline (n=26), anesthetic (n=21), or no injection (n=23). After accounting for dropouts, final analysis included 14 in the saline group, 17 in anesthetic, and 16 in the control group. Both saline and anesthetic treatments showed significantly reduced facial pain intensity compared to control (P=0.004 and P<0.001), though there was no significant difference between them (P=0.003 and P=0.005). Pain reduction was 87.71% for saline and 100% with anesthetic. TPI with saline or anesthetic significantly reduced facial pain, headache frequency, and intensity, with no significant differences between the injected substance. Adverse effects include post procedure soreness and lack of pain improvement at the temples or above ears. Limitations include small sample size, high dropout, low completion rates, high risk of bias (RoB 2.0), and limited generalizability due to the all-female, non-U.S. based population.
Electronic health reports were retrospectively reviewed and analyzed for 142 patients receiving US-guided physiological saline injection (PSI) for myofascial pain at a Thailand hospital.40 The most injected muscles include upper trapezius (19.5%), multifidus (10.0%) quadratus lumborum (9.5%), rhomboid muscle on the chest wall (6.8%), and most common in the upper and lower limbs deltoid (8.4%) and pyriformis (8.4%). Pain reduction was experienced in 72.8% of patients. Acceptable pain periods lasted 63 days, with 43.9% having acceptable pain over three months. No adverse events were reported. Limitations of this study include its retrospective nature, lack of control group, selection bias, and non-U.S. based patients with mean age 55 years limiting generalizability to the Medicare population.
Overall, some studies have shown that saline, when used as a comparator, can have similar effects to LAs. However, other studies report slower or less effective pain relief, with no clear evidence of superiority. The evidence is limited due to small sample sizes, short follow up periods, and limited generalizability. There is insufficient evidence at this time to support the use of saline for TPI, therefore it is not considered reasonable and necessary and will not be covered.
Bicarbonate and Hyaluronidase
Sodium Bicarbonate solutions can be injected into muscle to reduce metabolic acidosis. Hyaluronidase is an enzyme that breaks down naturally occurring hyaluronic acid which can accumulate in muscles. One study explored their effects in patients with MPS.
A single-blinded randomized study enrolled 56 patients (ages 25-75) with MPS in the trapezius and upper back, referred to two hospitals in Iran.41 Patients were allocated into groups receiving either bicarbonate (n=16), hyaluronidase (n=19), or lidocaine TPI (n=19). Pain reduction using the VAS and range of motion (ROM) were assessed before, immediately after, and at the second- and fourth-week post injection. The main effects of group and week were significant for VAS (P<0.05), with VAS values differing significantly between treatment groups by week four. Patients in the hyaluronidase group had the greatest reduction in pain during all assessment periods. Lidocaine significantly reduced pain immediately after injection, but effects did not last until the four week follow up mark. No adverse events were reported. Study limitations include a small sample size, lack of control group, no demographic information, short term follow up, and a high risk of bias (RoB 2.0).
The role of these agents in TPI is not well established. There is insufficient evidence to support the use of bicarbonate or hyaluronidase in TPI and therefore it will not be considered reasonable and necessary.
Ozone
Oxygen-ozone is a gas mixture that may improve tissue oxygenation, reduce inflammation, and provide mild pain relief.
Because of these effects and emerging evidence of benefit in musculoskeletal disorders (knee osteoarthritis, low back pain, tendon pathologies, carpal tunnel syndrome), it has also been studied as a potential injectate for TPI in patients with MPS.42,43
A single center, single-blinded, RCT was conducted in Turkey enrolled 46 patients (mean age 44.7 years) with upper trapezius MPS.42 Participants were randomized to receive either oxygen-ozone injections (n=23) or lidocaine injections (n=23) and were asked to continue exercise programs during follow up. The primary outcome was pain severity, measured with VAS, while secondary outcomes included pain score (PS), and Neck Disability Index (NDI). At the four- and 12-week follow-up marks, all measures significantly improved compared to baseline in both groups (P<0.001). Visual analog scale scores showed a significant group x time interaction, with lidocaine showing a greater pain reduction over time compared to oxygen-ozone group (P=0.028). Both treatments effectively improved pain and function, with lidocaine being superior in pain reduction, but not function and PS. No adverse events were reported. Study limitations include small sample size, lack of patient blinding, absence of a control group, some concerns for bias (RoB 2.0) and a younger, non-U.S. based population limiting generalizability.
A single-blinded RCT conducted in Iran enrolled 72 patients (mean age 39.4 years) diagnosed with MPS of the upper trapezius.43 Participants were randomized into three equal groups to receive either DN, ozone injection (OI), or lidocaine injection (LI) administered in three weekly sessions, alongside instructions for trapezius stretching exercise and muscle relaxation training. Outcomes were assessed at baseline and four weeks post treatment, including VAS for pain, cervical lateral flexion, pain pressure threshold (PPT), and NDI. Ten patients did not complete the study. All groups demonstrated significant improvements in pain and PPT at the four week follow up. Specifically, VAS scores decreased (mean difference [MD]= -3.6±1.4, P=0.001) and PPT increased (MD=7.2±5.1, P=0.001). Neck disability index scores also improved significantly (MD=–9.9±8.7, P=0.001), though no significant increase was observed in lateral flexion range (MD=1.7±3.4, P=0.05). A significant difference in efficacy was found between the treatment groups for VAS, NDI, and PPT (P=0.02, 0.01, and 0.04, respectively), favoring the OI and LI groups over DN. No significant differences were noted between the OI and LI groups. Adverse events included transient local flare reactions unspecified minor events. Study limitations include the small sample size, short follow up period, lack of an objective functionality assessment tool, participant dropouts, high risk of bias (RoB 2.0) and the younger, non-U.S. based population limiting generalizability.
The role of oxygen-ozone in TPI is not well established. There is insufficient evidence to support the use of oxygen-ozone in TPI and therefore will not be considered reasonable and necessary.
Granisetron
Granisetron is a serotonin 5-HT3 receptor antagonist commonly used to treat nausea and vomiting. Its proposed use in TPI is based on the idea that blocking 5-HT3 receptors may reduce chronic muscle pain.44
A double-blinded RCT included 40 patients (mean age 50.1 years) with MPS of the upper trapezius from an Iran hospital.44 Participants were randomized to receive TPI of either lidocaine (n=20) or granisetron (n=20) and were instructed to perform muscle stretching exercises throughout the study. Outcomes were measured using the NDI)and NDPS prior to injection and at 1- and 3-month follow-up visits. Both treatment groups demonstrated significant reductions in neck pain and disability (P<0.001), with lidocaine group showing more favorable responses than granisetron (P=0.001 for NDI, and P=0.006 for NDPS). Side effects were minimal and included transient injection site pain. Per the RoB 2.0 tool, there was a low risk of bias for this study. Limitations of this study include the small sample size, lack of a control group, younger, non-U.S. based population limiting generalizability, and the inclusion of stretching exercises preventing a pure comparison between injection treatments.
The role of this agent in TPI is not well established. There is insufficient evidence to support the use of granisetron for TPI and therefore will not be considered reasonable and necessary.
Nonsteroidal Anti-Inflammatory Drug (NSAID)
Diclofenac is an NSAID used to treat pain and inflammation by inhibiting the production of prostaglandins. One study explores the idea that this may reduce local inflammation and might help relieve muscle pain.45
A single-blinded randomized study enrolled 35 patients (median age 50 years) with localized myofascial pain who were referred to an outpatient rheumatological clinic in Denmark.45 Patients were randomized to receive TPI with either lidocaine (n=17) or diclofenac (n=18). Pain was assessed using a 10 cm VAS for up to five hours post injection, self-reported by patients. Eleven patients (six lidocaine, five diclofenac) were excluded from analysis due to protocol deviations, including the use of other analgesics or technical issues with pain recording, leaving 24 patients in the final analysis. Diclofenac showed significantly greater pain reduction than lidocaine at the four hour mark (P<0.05), with similar but insignificant trends at two and five hours (P<0.1). Compared to baseline, the diclofenac group showed significant pain relief after three hours, whereas the lidocaine group did not have any significant changes in pain levels. No adverse events were reported. Limitations include small sample size, short follow up time, lack of control group, high dropout rate, lack of methodological detail, high risk of bias (RoB 2.0) and a younger, non-U.S. based population limiting generalizability.
The role of NSAIDS as an injectate for TPI is not well established. There is insufficient evidence to support the use of NSAID as an injectate for TPI and therefore will not be considered reasonable and necessary.
Rationale for Injectate Determinations
Overall, a wide variety of injectates for TPI have been studied across trials. No single injectate has shown clear superiority,29 and the certainty of evidence remains low. However, LAs have shown consistent evidence of effectiveness, and in some studies have been more effective than BTX.14 Furthermore, guidelines recommend that BTX should not be used in the routine care of patients with myofascial pain, based on equivocal trial results and lack of expert consensus.2 Saline, when used as a comparator, has shown equivalent effects to LA in some studies,36 but generally provided slower or less complete relief, lacking evidence of superiority.37-39 Corticosteroids have been evaluated in very few studies31,32 and are not recommended for routine use based on minimal benefit, potential for harm, and lack of guideline support.3 There is insufficient evidence to support the use of platelet rich plasma (PRP), oxygen-ozone, bicarbonate, hyaluronidase, granisetron, or other biologics, and will not be considered medically reasonable and necessary at this time and consequently, are non-covered.
Based on the available evidence, LAs are the most appropriate and commonly used injectate for TPI.15 While studies across all injectate types share common limitations such as small sample sizes, short follow up periods, and heterogeneity in injectate type, dosing, injection techniques, outcome measures, and added interventions such as exercise therapy, local anesthetics have consistently demonstrated benefit across these trials. Generalizability is also limited due to the younger, non-U.S. based populations studied, which may not reflect the Medicare population. Despite these limitations, no other injectate has shown clear superiority in effectiveness or safety, and therefore the use of local anesthetics alone is considered reasonable and necessary.
Myofascial Pain Syndrome
Myofascial pain syndrome (MPS) is a painful condition characterized by the presence of myofascial trigger points. Primary treatment focuses on exercise and education, while medications, physical modalities, DN, and TPI are considered adjunct therapies appropriate for selected patients.46 A three-round Delphi survey established an expert based standardized definition of a trigger point, concluding that at least two of the following criteria must be present for diagnosis: a taut band, a hypersensitive spot, and referred pain.47
The American Society of Anesthesiologists (ASA) Task Force and the American Society of Regional Anesthesia and Pain Medicine (ASRA) 2010 practice guidelines for chronic pain management2 provide the following recommendations:
“TPI may be considered for treatment of patients with myofascial pain as part of a multimodal approach to pain management”.
While the guideline focuses on chronic pain management, the recommendations specifically mention TPI for patients with myofascial pain. The guideline notes there is “insufficient literature to evaluate TPI efficacy (i.e., compared with sham TPI) as a technique for providing pain relief for patients with chronic pain (Category D evidence)”. However, “observational studies suggest TPI may provide relief for patients with myofascial pain for assessment periods ranging from 1 to 4 months (Category B2 evidence)”. Consultants, ASA members, and ASRA members agree that TPI should be used for patients with myofascial pain.”2
A systematic review and meta-analysis compared the effectiveness of DN against TPI for patients with neck pain associated with myofascial trigger points.48 Databases searched included MEDLINE, CINAHL, PubMed, PEDro, Cochrane Library, SCOPUS, and Web of Science from inception to July 10, 2020. The primary outcomes were pain or pain related disability. Authors used The Cochrane Risk of Bias tool, the Physiotherapy Evidence Database score, and the GRADE approach to assess the quality of evidence. Six RCTs were included in analysis. TPI was found to reduce pain intensity (MD –2.13, 95% CI –3.22 to –1.03) with a large effect size (standardized mean difference [SMD] –1.46, 95% CI –2.27 to –0.65) compared with DN. No significant differences were found between TPI and DN for pain-related disability, pressure pain thresholds, cervical lateral-flexion, or depression. Common adverse events reported include post needling soreness, muscle pain, discomfort, paresthesia, fatigue, headache, hemorrhage, transient flare reaction, and dizziness. Although the risk of bias for studies was low, the evidence was downgraded by the author to low certainty due to heterogeneity and imprecision. Additional limitations include short follow up, varied age ranges, and the limited number of studies.
A single-blinded RCT enrolled 127 patients with shoulder girdle MPS from two hospitals in Colombia.49 Participants were randomized into three groups: Physical Therapy (PT) + Local Injection (LI) (n=43, mean age 37.2 years), PT (n=41, mean age 42.6 yrs), or LI (n=43 mean age 37.7 years). Pain ratings were assessed using the VAS one month after treatment. Secondary outcomes included VAS pain rating at three months, as well as function, quality of life and depressive symptoms evaluated at 1- and 3-months post treatment. A per protocol analysis found no significant differences in pain reduction between groups at one month (PT + LI, 40.8 [25.3] vs. PT, 37.8 [21.9], P=0.560 and vs. LI, 44.2 [24.9], P=0.545). No significant differences were observed in secondary outcomes. Six patients experienced complications, including localized hematomas (four in the PT + LI group, two in the LI group) and one minimal bleeding in the LI group. Limitations of this study include a small sample, absence of a placebo group, short follow up period, lack of patient blinding, per protocol analysis excluding missing participants, high risk of bias (RoB 2.0), different physical therapists and injectors, lack of patient centered functional measures, high variability across similar studies limiting comparability, as well as a younger, non-U.S. based population limiting generalizability.
A RCT enrolled 50 patients with noncardiac chest pain (NCCP) associated with MPS from outpatient clinics in Turkey.50 Patients were randomized into two groups: one receiving TPI into the pectoralis muscles combined with exercise (n=25; mean age: 42.8 years) and the other receiving exercise alone (n=25; mean age: 41.8 years). Pain intensity was assessed using the VAS at 1 and 3-months post treatment. Both groups showed significant improvements in VAS scores at follow up (P<0.001). The combination of TPI and stretching exercises resulted in significantly greater pain reduction compared to exercise alone, with results persisting at the 3-month follow-up (P<0.001). However, no significant differences between groups were found in quality of life, evaluated using the Nottingham Health Profile (P=0.522). No adverse events were reported. Study limitations include small sample size, short follow up duration, five patients lost to follow up and excluded from analysis, lack of blinding, high risk of bias (RoB 2.0), focus on patients with trigger points in the pectoralis muscle only, and a younger, non-U.S. based population limiting generalizability.
A randomized study enrolled 58 patients with MPS of the upper trapezius at a medical center in California.51 Patients were randomized to receive either TPI with lidocaine (group 1, n=26, mean age 42.2 years) or DN (group 2, n=15, mean age 41.7 years). Each group was further divided based on whether a local twitch response (LTR) was elicited, with group 1a (n=9, mean age 39.9 years) including patients receiving lidocaine without LTR, while group 2a (n=8, mean age 42.1 years) included patients receiving DN without LTR. Outcomes measured included subjective pain intensity, pain threshold, and range of motion of the cervical spine. Assessments were performed by blinded evaluators immediately after treatment and at two weeks. Patients in groups who did not have an LTR during treatment were not assessed at two weeks, as they received additional treatment. Significant improvement was observed immediately after treatment in groups 1 and 2 (P<0.05), however, at two weeks, group 1 had a significantly greater reduction in pain intensity compared to group 2 (P<0.05). Immediate effects also diminished in both groups after two weeks. In contrast, patients without an LTR showed minimal changes in pain, tenderness, or tightness. Adverse events include post injection soreness in both treatment groups, with the DN group reporting significantly higher soreness intensity and longer duration compared to the lidocaine group. Study limitations include a small sample size, no control group, short follow up duration, lack of blinding, lack of methodological detail, some concerns for bias (RoB 2.0), and a younger population, limiting generalizability.
A randomized study enrolled 30 patients with myofascial pain from a pain clinic in Brazil.52 Participants were randomized into two groups: group 1 (n=15, mean age 36 years) received bupivacaine TPI twice a week combined with daily cyclobenzaprine chlorhydrate and sodium dipyrone, while group 2 (n=15, mean age 32.2 years) received a combination of both classical acupuncture and trigger point acupuncture twice weekly. Both groups were provided instructions for physical exercise. Pain intensity, number of trigger points, and quality of life were assessed at baseline and four weeks after treatment. Both groups showed a significant reduction in pain scores and number of trigger points, with no significant differences between groups. Additionally, improvements in quality of life scores were reported for some functional domains in both groups, although no improvements were observed in the general health status domain or the emotional domain in group 1. Reported adverse events include sleepiness, local pain, dry mouth, lipothymia, and epigastria. Limitations of this study include small sample size, lack of blinding, high bias risk (RoB 2.0), absence of a control group, short follow up period, and a younger, non-U.S. based population limiting generalizability.
A single-blinded randomized trial enrolled 39 patients with chronic MPS of the upper trapezius muscle from a hospital in Korea.53 Participants were randomized to receive either acupuncture needling (n=18, mean age 79.22 years) or lidocaine injection (n=21, mean age 75.9 years) at all trigger points on days 0, 7, and 14. Patients were instructed to perform self-stretching exercises three times daily until the next treatment session. Outcomes assessed included pain scores, range of neck movement, pressure pain intensity, and depression, measured four weeks after first treatment. Both groups demonstrated significant reductions in pain compared to baseline, but no statistically significant differences were observed between groups at any time point. Overall, the range of cervical movement improved similarly in both groups except for extension in the acupuncture needling group, which did not improve. Adverse events include 17 cases of soreness across both groups, one case of dizziness, and one case of visible subcutaneous hemorrhage in the lidocaine group. Study limitations include a small sample size, lack of blinding for treatment providers and outcome assessors, some concerns for bias (RoB 2.0), use of thumb pressure rather than algometer for pain intensity, lack of long term follow up, as well as a non-U.S. based population limiting generalizability.
A randomized study enrolled 104 patients with acute MPS in the cervical region from outpatient clinics in Turkey.54 Patients were randomized into three groups: Group 1 (n=35, mean age 36.9 years) received kinesiotape (KT), Group 2 (n=35, mean age 37 years) received a single TPI of lidocaine into the trapezius muscle, and Group 3 (n=34, mean age 41.2 years) received neural therapy (NT) injections with the same lidocaine mixture. Pain was assessed using the VAS and PPT, while disability was measured using the Turkish version of the NPDS. Two patients dropped out prior to study completion. All groups showed significant improvements in pain and disability at the three day and seven day follow up points (P<0.001), with no statistically significant differences between the groups. Significant improvements in PPT were observed over time in the TPI and NT injection groups, but not in the KT group. No adverse events were reported. Study limitations include lack of blinding, some concerns for bias (RoB 2.0), lack of control group, short term follow up, focus on immediate pain relief for acute MPS, as well as a younger, non-U.S. based population limiting generalizability.
A single-blinded RCT enrolled 60 patients with myofascial trigger points in the upper trapezius muscle from a hospital in Thailand.55 Patients were randomized to receive either repeat lidocaine TPI (n=30, mean age 22.93 years) or radial extracorporeal shock wave therapy (rESWT) (n=30, mean age 23.05 years). Both groups completed three weekly treatment sessions. Pain severity was assessed using a VAS at baseline (T1), 15–30 min after the first treatment (T2), before the second treatment (T3) and one week after the third treatment (T4). Additional outcomes included muscle elasticity index, PPT and NDI. No significant differences in outcomes were observed between T1 and T4. However, within group comparisons (T3, T4) they showed significant improvements in all outcomes when comparing baseline and post treatment measurements (P<0.001). Reported adverse events include pain and skin reddening in some cases. Study limitations include small sample size, lack of patient blinding, some concerns for bias (RoB 2.0), no subgroup analysis for different rESWT settings, and a younger, non-U.S. based population limiting generalizability. Additionally, this pilot study was not a non-inferiority trial and could not establish that rESWT is as effective as lidocaine injections.
A randomized trial enrolled 54 patients with low back pain due to myofascial trigger points in the quadratus lumborum (QL) muscle at an Iran hospital.56 Participants were randomized to receive either radial shock wave therapy (ESWT; n=27, mean age 44.7 years) in five weekly sessions or a single session of corticosteroid trigger point injection (TPI; n=27, mean age 45 years). Outcomes were blindly assessed using the Oswestry Disability Index (ODI), VAS, PPT, and short form (36) health survey (SF-36), measured at baseline, week two, and week four. All patients were instructed to perform stretching exercises with tracking. At the two week follow-up, the corticosteroid TPI group showed significantly greater improvements in ODI (P<0.01), VAS (P<0.001), and PPT (P=0.001) compared to the ESWT group. However, by week four, the ESWT group demonstrated significantly better outcomes in ODI (P< 0.01) and SF-36 (P<0.001). "Patients in the ESWT group were 1.46 times more likely to achieve 30% reduction in VAS, 2.67 times more likely to achieve 30% reduction in ODI, and 2.30 times more likely to achieve 20% improvement in SF-36 compared to corticosteroid TPI group." The reported effect sizes for ESWT were large across all outcomes (d=4.72, d=1.58, d=5.48, and d=7.47 for ODI, PPT, SF-36, and VAS, respectively). No adverse events were reported. As per risk of bias assessment (RoB 2.0), there is a low risk of bias. Study limitations include small sample size, short follow up period, absence of a control group, lack of patient blinding, and a younger, non-U.S. based population limiting generalizability.
A single-center, open-label RCT by Farrow and colleagues57 evaluated ED patients in Mount Sinai Medical Center, Florida presenting with neck and back pain associated with MPS. Patients were selected from a convenience sample and randomized in advance to receive either US-guided TPI with bupivacaine (n=102) or standard medications (n=105). Participants were given the option to request rescue therapy if pain persisted, except additional injections on those who had already received TPI during initial treatment. Of the 207 enrolled, outcomes of 196 patients were analyzed after accounting for dropouts. The primary analysis focused on the most common therapy used in the standard medication group, a combination of NSAID plus muscle relaxant (MR) (n=56). No significant differences in pain reduction were found in the groups during ED discharge, However, TPI group experienced greater reduction at the first assessment (15 to 30 minutes post procedure) and required less rescue therapy overall. No serious adverse events occurred, and no patients returned due to complications within two weeks. Limitations of this study include selection bias due to convenience sampling, the ED setting, short term follow up, lack of blinding, lack of medication standardization, variability in pain reassessment timing, missing data, and a high risk of bias as assessed by RoB 2.0. Generalizability is also limited due to the young sample (median age 44) not being reflective of the Medicare population.
A retrospective study included 30 patients with quadratus lumborum MPS treated at a hospital in Korea.58 Patients selected their preferred treatment of either extracorporeal shock wave therapy (ESWT) (n=15, mean age 55 years) or TPI (n=15, mean age 53 years). All patients received conservative treatment including analgesics, rest, and therapeutic exercises. Treatments were administered three times at three-day intervals. Pain and disability outcomes were measured using the VAS, PPT, and disability assessment at baseline, after the third treatment, and at one month follow up. Both ESWT and TPI groups showed significant improvements in pain and disability compared to baseline. However, ESWT demonstrated significantly greater reduction in pain compared to TPI (P<0.05), while improvements in disability measures were similar between groups with no significant differences. No adverse events were reported. Study limitations include its retrospective nature, small sample size, lack of a control group, short follow up period, potential bias due to patient selection of treatment, and a younger, non-U.S. based population limiting generalizability.
Overall, evidence suggests that TPI are effective for reducing pain and improving functional outcomes in patients with MPS. Guidelines recommend TPI as part of a multimodal approach to pain management.2 Multiple studies have compared TPIs to other modalities including DN, acupuncture, kinesiotaping, neural therapy, and extracorporeal shock wave therapy, with all interventions demonstrating some benefit. While no single modality has shown clear superiority, TPIs have consistently shown improvements in pain intensity and are a commonly used treatment option. Despite these findings, certainty of evidence is limited by small sample sizes, short follow up durations, lack of blinding, absence of control or placebo groups, and variability in study techniques. Additionally, most of the studies included younger, non-U.S. based populations, limiting generalizability to Medicare beneficiaries.
Fibromyalgia
The most common feature among patients with fibromyalgia syndrome (FMS) and MPS is chronic musculoskeletal pain. While both conditions involve local tenderness, FMS is primarily associated with tender points, while MPS involves trigger points. A tender point refers to a localized area of tenderness in a muscle, muscle tendon joint, fat pad, or a bursal region. In contrast, myofascial TPs are palpable as discrete, focal, hyperirritable spots located within a taut band of skeletal muscle or its associated fascia. Clinicians recognize FMS and MPS as distinct clinical entities that require different management strategies.9
Sixty-two female patients (mean age 45.8 years) recruited from FMS support groups were enrolled in a double-blind randomized controlled study at the University of Florida Department of Medicine in a 2014 study.59 Patients with musculoskeletal pain who met the American College of Rheumatology (ACR) criteria for FMS were randomized into three groups: Group 1 received four lidocaine injections (n=20), Group 2 received two lidocaine and two saline injections (n=21), and group 3 received four saline injections (n=21). Patients received intramuscular injections, two into the center of each trapezius and two into the upper medial quadrants of both gluteus maximus muscles. Pressure and heat hyperalgesia along with clinical pain were assessed at 30- and 60-minutes post injection using the VAS. Lidocaine injection significantly reduced primary hyperalgesia in the trapezius and gluteal muscles (P=0.004) and secondary heat hyperalgesia in the arms (P=0.04) compared to saline. Pain pressure thresholds at the arms and legs were not significantly increased in any group, although gluteal muscles showed an increase in PPT with lidocaine (P<0.001). Pain pressure thresholds also increased in the shoulder muscles for both groups, with greater improvement in the lidocaine group (P=0.004). Clinical FMS pain significantly declined by 38%, but no statistical differences were observed between groups. No adverse events were reported. Risk of bias assessment per the RoB 2.0 tool determined low risk. However, study limitations included the use of intramuscular rather than targeted trigger point or tender point injections, small sample size, selection bias, and limited generalizability due to the all-female, younger study population.
A prospective study in 1996 included 18 female patients with active TrPs of the upper trapezius from a pain control clinic in Taiwan.60 Group 1 (n=9, mean age 36.6 years) had both MPS and FMS while Group 2 (n=9, mean age 36.3 years) had MPS only. The investigator performed TPI with lidocaine, and a blinded assistant recorded outcomes. Immediately after injection, the MPS + FMS group showed significant improvements in range of motion (ROM), while the MPS-only group showed improvements in ROM, pain intensity (PI), and pain threshold (PT) (P<0.05). At the two week follow up, both groups had improvement in all parameters (P<0.05), but the MPS + FMS group showed fewer immediate effects. The two groups did not differ in PT and ROM improvements; however, post injection soreness was more severe, developed sooner, and lasted longer in the MPS + FMS group. No adverse events were reported. Study limitations include small sample, lack of long term follow up, limited generalizability to the Medicare population due to an all-female, younger, non-U.S. based population, as well as the factors influencing results, such as stretching and home exercises.
An evidence-based clinical review commissioned by the American Pain Society (APS) focused on evaluating treatments for FMS.6 Multiple databases were searched to collect evidence from 1966-2004 and included all human, RCTs and meta-analyses of RCTs of FMS. Literature was then reviewed by an APS selected expert panel. It was found that trigger point injection efficacy for FMS has not been adequately evaluated, as there were no available RCTs at that time, therefore there is no evidence of efficacy for trigger point and tender point injections for patients with FMS.
There is insufficient evidence at this time to support the efficacy of TPI in the treatment of fibromyalgia. The small number of existing studies primarily addressed treating MPS in patients who also have fibromyalgia or using intramuscular injections in general muscle regions rather than targeting tender points specifically. Studies are also limited due to small sample sizes, limited generalizability due to all female, younger, and non-U.S. based populations. Using TPI solely for fibromyalgia treatment is not medically reasonable or necessary, and therefore, not covered.
Low Back Pain
Evidence-based clinical guidelines from the North American Spine Society (NASS) for the diagnosis and treatment of back pain4 provide the following recommendation regarding TPI:
"There is insufficient evidence to make a recommendation for or against the use of TPI in the treatment of low back pain. The type of injectate does not influence outcomes.
Grade of Recommendation: I. The work group does not have any recommendations for future research on this topic". Grade I recommendation is defined by the authors as “Insufficient or conflicting evidence not allowing a recommendation for or against intervention”.
Similarly, the American Association of Neurological Surgeons (AANS) and Congress of Neurological Surgeons (CNS) Joint Guidelines Committee (JGC), in their update on fusion procedures for degenerative disease of the lumbar spine state that there is insufficient evidence to support or refute the use of TPI for chronic low back pain without radiculopathy.5
The update provides a Grade B recommendation for lumbar TPI, concluding that interventions including injections using anesthetics alone, with steroids, or techniques such as DN are not recommended in patients with chronic low-back pain without radiculopathy from degenerative disease of the lumbar spine, as long-lasting benefits have not been demonstrated (Level II evidence).
Additionally, the guideline notes that there is no evidence to support the use of TPI with anesthetics alone, accompanied by steroids, or DN techniques, in the management of patients suffering from chronic low-back pain secondary to degenerative lumbar disease (Level IV evidence).
A systematic review aimed to assess the effectiveness of injection therapy compared to a placebo or other treatments for patients with subacute or chronic low back pain.61 Databases searched included the Cochrane Central Register of Controlled Trials, MEDLINE, and EMBASE through March 2007, and encompassed studies published in English, French, German, Dutch, and Nordic languages. Eighteen RCTs involving 1179 participants were included, evaluating a range of injection sites from epidural sites and facet joints (i.e. intra-articular injections, periarticular injections and nerve blocks) to local sites (i.e. tender and trigger points) and a variety of drugs such as corticosteroids, LAs, and other agents. Pain relief was the primary outcome assessed, as either a percentage of patients reporting improvement, or a mean change in pain relief on a continuous scale, with both short term (<6 weeks) and long term (>6 weeks) effects evaluated. The methodological quality of included trials was limited with ten out of 18 studies rated as having a high methodological quality. Due to clinical heterogeneity among studies, statistical pooling was not possible. Overall, the results indicated that there is no strong evidence for or against the use of any type of injection therapy for subacute or chronic low back pain. Reported adverse events across studies included headache, dizziness, transient local pain, paresthesia, and nausea. Overall, this review provides low to moderate certainty of evidence, as it is limited by a small sample size and study heterogeneity.
A randomized double-blind trial enrolled 63 patients in the United States (average age 38 years) with low back strain who received conservative therapy four weeks prior to entering the study.62 Patients were randomized to receive either TPI with lidocaine (group A, n=13), TPI of lidocaine and steroid (group B, n=14), acupuncture (group C, n=20), or vapocoolant spray with acupressure (group D, n=16). Pain improvement was assessed at the two week follow-up mark through a subjective patient response categorized as “improved” or “not improved”. Results showed a 63% improvement rate in patients receiving therapy without injected medication and a 41% improvement rate in those who received injected medications, though this difference was not statistically significant (P=0.09). Study limitations include a small sample size, a high attrition rate of 20%, the use of subjective outcome measures, inability to use an analog scale as planned, inadequate statistical power, short follow up period, lack of recruitment site details, high risk of bias (RoB 2.0), and a younger population limiting Medicare generalizability.
A randomized study enrolled 54 patients admitted to the ED due to low back pain caused by TrPs in Turkey.63 Patients were randomized during time of admission to receive either intravenous drug NSAID therapy (n=32; mean age 40.9 years) or TPI (n=22; mean age 45.1 years). Pain scores were measured using the VAS 60 minutes post procedure. Results showed a significant decrease in pain scores in the TPI group, with a MR of 0.41 ± 1.30, compared to a MR of 2.59 ± 2.37 in the NSAID group (P<0.001). Additionally, response to treatment was higher in the TPI group compared to the NSAID group (21/22 vs 20/32 respectively, P=0.008). No adverse events were reported during the study. This study was limited by its small sample size, short follow up period, lack of blinding, focus on acute pain in an ED setting, lack of assessment on the reliability or reproducibility of TrP identification, select patient groups, some concerns for bias (RoB 2.0), and the younger, non-U.S. based patient populations limiting Medicare generalizability to the U.S. Medicare population.
A prospective study included 98 patients with lumbosacral radiculopathy referred to an orthopedic clinic in Iran.64 All patients initially received conservative treatment for one week, including bed rest, NSAIDs, and physiotherapy. Sixty-four patients who continued to experience symptoms and had trigger points were divided into two groups, with group TP (n=32; mean age 46 years) receiving TPI combined with continued conservative therapy and group N (n=32; mean age 49.1 years) having continued conservative therapy alone for three more days. Pain scores were assessed on day seven and ten of therapy, using VAS assessment and straight leg raise (SLR) tests, with a positive SLR indicating pain felt and a negative score indicating no pain when raising the leg between 0 to 70 degrees. Pain scores (Mean ± SD) in the TP group were 7.12 ± 1.13 and in the N group were 6.7 ± 1.16, P=0.196. Following the treatment, pain scores were 2.4 ± 1.5 in the TP group and 4.06 ± 1.76 in the N group, P=0.008. SLR test became negative in all patients in the TP group, but only in six (19%) patients in the N group, P=0.001. No adverse events were reported. The TPI group had lower pain scores and almost full SLR recovery compared to those receiving conservative treatment alone. Limitations of this study include selection bias, small sample size, and a younger, non-U.S. based population limiting Medicare generalizability.
Overall, there is low certainty of evidence to support the use of TPIs for the treatment of low back pain. Guidelines and systematic reviews report insufficient evidence to recommend for or against their use.4,5,61 Studies that suggest possible benefits are limited by small sample sizes, short follow up periods, and methodological weakness. Additionally, younger, non-US based populations further limit generalizability to the Medicare population.
Due to the lack of endorsement from specialty societies and insufficient supporting evidence, TPIs are not considered reasonable and necessary for low back pain and therefore will not be covered.
Evidence Submitted During the Comment Period
No evidence was submitted during the comment period. However, references cited within submitted comments were reviewed, when publicly accessible, and assessed to address the comments received. Responses to all submitted comments are provided in the article Response to Comments: Trigger Point Injections (A60436/A60437).
A total of 102 references were cited across all comments. After accounting for duplicate citations, 28 references were identified that were not already included in the LCD bibliography. In the absence of submitted full-text materials, the contractor screened references that were identifiable through database searches and publicly accessible (13). References were excluded from detailed review when they were outside the scope of this LCD, including those addressing pelvic or pelvic floor conditions (e.g., pelvic pain, pelvic floor dysfunction, urogenital or gynecologic conditions), non-trigger point related conditions (e.g., thoracic outlet syndrome, piriformis syndrome, and abdominal wall pain conditions, including abdominal cutaneous nerve entrapment syndrome or nerve-related abdominal wall injections), narrative reviews evaluating ultrasound-guided thoracic or chest wall interventions, medico-legal analyses, or procedures not applicable to TPIs (e.g., transversus abdominis plane blocks).
Three references were reviewed in full text, including two guidelines (Sayed, 2022; Finnoff, 2015) and one RCT (Chen, 2025). All were excluded from the evidence analysis.
The American Society of Pain and Neuroscience (ASPN) Evidence-Based Clinical Guideline (Sayed et al, 2022) supports the general use of trigger point injections for myofascial pain but is primarily focused on low back pain and contains only a limited subsection on TPIs. The guideline includes descriptive discussion of potential complications based largely on case reports and observational data but does not evaluate or include studies comparing image-guided and non-guided TPIs, does not assess whether imaging improves clinical outcomes or reduces complications, and does not provide recommendations regarding the use of imaging guidance.
The American Medical Society for Sports Medicine (AMSSM) position statement (Finnoff et al, 2015) broadly evaluates US-guided musculoskeletal interventions across joints and soft tissues but does not address TPIs or MPS specifically and does not provide outcome-based evidence supporting image-guided TPIs.
The RCT (Chen et al, 2025) evaluated US-guided myofascial hydrodissection compared to TPI; however, hydrodissection was the primary intervention, and both study arms were performed under ultrasound guidance. The study did not include a comparison between image-guided and non-guided techniques and did not isolate the effect of imaging guidance.
As these references do not evaluate TPIs as the primary intervention, do not provide outcome-based evidence assessing the clinical utility or medical necessity of imaging guidance, and do not contribute additional information beyond that already included in the existing evidence base, they were not relied upon in the development of the coverage criteria for this LCD and did not affect the evidence analysis or coverage determinations.