Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials
BMJ 2025; 388 doi: https://doi.org/10.1136/bmj-2024-079971 (Published 19 February 2025) Cite this as: BMJ 2025;388:e079971Linked Practice
Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline
Linked Editorial
Spinal interventions for chronic back pain
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
On behalf of the ASIPP Board of Directors, we were alarmed when we reviewed Wang and colleagues’ publication of their systematic review and network meta-analysis of randomized controlled trials of common interventional procedures for chronic non-cancer spine pain.
The Rapid Recommendation Guidelines were of such rapid nature that the authorship lacked relevant interventional expertise and what qualified as a “randomized” trial included those with as few as 10 patients, lacked differentiation between high and low-quality studies, what was referred to as a “sham procedure” was pharmacologically or physiologically active, clinically meaningful pain relief was simply dismissed and included studies that followed chronic pain patients for as short as one month while reviewing the treatment arm out of context to current practice of interventional pain medicine.
Interventional spine procedures serve diagnostic and therapeutic purposes and are part of an interdisciplinary approach to patient care. They form part of the interpreted workup similar to imaging workup that tends to be overly sensitive and nonspecific. In addition to appropriate diagnostic utilization, the effectiveness of the reviewed interventional procedures is dependent on proper patient selection, duration of underlying pathology, and proper technique based on the location of pathology (e.g., foraminal vs. central, unilateral vs. bilateral), after which interventional treatments are utilized as a part of a multimodal plan of care that may include physical, pharmacological and behavioral therapy coupled with lifestyle changes and expectation setting.
The reviewed interventions enjoy some of the lowest number of needed to treat (NNTs) in chronic pain for clinically meaningful pain relief relative other treatments. In appropriately selected patients, reviewed interventions offer substantial relief, improve function, enable physical therapy, allow performance of activities of daily living, maintenance of work or return to work, prevent detrimental effects of immobilization, reduce number of emergency department visits, and may delay or obviate the need for surgical interventions or long-term reliance on opioids.
The article’s conclusions diverge from a well-established body of high-quality research that supports the efficacy of interventional procedures, particularly with appropriate patient selection. The authors either completely discounted, misinterpreted or simply ignored several publications and peer reviewed data that demonstrate clinically meaningful relief that will be addressed in a subsequent publication.
The weaponization of the BMJ “Rapid Recommendation” against chronic pain sufferers notwithstanding, the article has the potential to impede access so that a meaningful duration of pain relief can be provided to sufferers of, for example, subacute or chronic cervical and lumbar radiculopathy and mitigate pain, suffering and chronification, contributing to unbalanced care that compensates by relying on peripheral and central analgesics with their adverse effects on a range of end organs and unmatched morbidity and mortality.
The Guideline, in sum and substance, states that such standard interventional procedures for chronic non-cancer spine pain are ineffective, should not be available to patients, and a financial reordering of their compensation is in order to bring how pain medicine was practiced back to the 1990s with endless failure of pharmacological and nonpharmacological therapies.
To be clear, the interventions need to be utilized as part of an interdisciplinary approach and applied appropriately based on the correct diagnosis. They are not intended for everyone with spinal pain. However, creating a nihilistic guideline that effectively denies appropriately selected patients for interventions does a great disservice to chronic pain suffers, and the outcome of treatment denials, unfortunately, may be pharmacological and surgical overutilization or persistence of avoidable pain and suffering that serves to promote chronic pain. The BMJ author’s message to spinal pain sufferers is that if a trial of conservative management fails, go for something riskier or suffer, will contribute to limiting patient care and choice, contributing to further deterioration of chronic pain outcomes, a metric that guidelines aim to improve, not antagonize.
Christopher Gharibo, MD, on behalf of the ASIPP Board of Directors
President, American Society or Interventional Pain Physicians
www.asipp.org
Competing interests: No competing interests
Dear Editor,
In light of the recent publication in BMJ “Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practical guideline” (1) it is imperative to critically appraise our research methodologies in pain management.
As physicians and scientists, we are trained to navigate challenges that often extend beyond the limits of our medical expertise and human capabilities. The COVID-19 pandemic of 2020 underscored this reality, presenting not only a biological crisis but also an epistemological one. Amid the urgent search for effective treatments, the scientific community witnessed a parallel outbreak, one of pseudoscience. Numerous medical and non-medical professionals hastily claimed curative treatment outcomes without sufficient evidence, despite the brief period since the pandemic's onset. Ultimately, time became the most impartial judge, revealing the virus multifaceted nature and the unpredictability of individual responses; some of which could have been attributed to the placebo effect rather than the intervention itself.
The scientific method teaches us that rigorous standardization in patient selection and treatment, coupled with reproducibility of outcomes, is essential for drawing valid conclusions. However, in the complex realm of pain management, this principle faces significant challenges. Pain is a multidimensional phenomenon influenced by anatomical variations, psychological factors, and clinician expertise in both diagnosis and procedural execution. Oversimplification of this complexity risks diminishing the depth and nuance required for effective pain management strategies.
A notable example of the importance of contextual understanding in research is the satirical study coincidentally published in The BMJ’s 2018 Christmas issue, titled "Parachute use to prevent death and major trauma when jumping from aircraft: randomized controlled trial" (2). The study humorously reported that parachute use did not significantly reduce death or major injury compared to a control group when participants jumped from stationary aircraft on the ground. The authors aimed to highlight the limitations of randomized controlled trials (RCTs) and caution against blind adherence to methodological rigor at the expense of real-world applicability.
As researchers and clinicians, we must cultivate a critical perspective on existing practices, continuously striving for robust and contextually appropriate evidence when determining optimal treatment approaches. However, this necessitates a clear understanding of the research landscape we must first ascertain whether the proverbial plane is in flight before evaluating the necessity of a parachute.
References
1. Busse J W, Genevay S, Agarwal A, Standaert C J, Carneiro K, Friedrich J et al. Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline BMJ 2025; 388 :e079970
2. Yeh RW, Valsdottir LR, Yeh MW, et al. Parachute use to prevent death and major trauma when jumping from aircraft: randomized controlled trial. BMJ. 2018;363:k5094. doi:10.1136/bmj.k5094
Competing interests: Proctor Medtronic
Dear Editor:
On behalf of 34 societies* representing physicians worldwide who treat patients with chronic pain, we write respectfully in response to the systematic review/network meta-analysis (NMA) by Wang et al. [1]. We have significant concerns regarding methodological flaws that compromise the reliability and applicability of their conclusions.
Heterogeneity in Study Selection
The analyses aggregate diverse patient populations, diagnoses, spinal regions, and procedures. The Cochrane Handbook states that a valid NMA requires comparable study sets to ensure meaningful conclusions [2]. However, this review includes heterogeneous procedures, such as thermal radiofrequency neurotomy (RFN) and intra-articular pulsed radiofrequency, and various epidural steroid injection (ESI) approaches, without appropriately distinguishing them.
For example, in eFig2, radiofrequency techniques for different spinal regions and pathologies are grouped together despite fundamental differences in anatomy and clinical outcomes. Similarly, in eFig6 and eFig8, medial branch RFN (an established technique with a defined mechanism of action guiding its application, though with technical variants that may impact efficacy) [3] and lateral branch RFN (a relatively new technique that applies the knowledge base of medial branch RFN but without widely-accepted selection criteria and technical parameters) are combined with intra-articular pulsed radiofrequency (an experimental treatment that is not similar to RFN without a clear mechanism of action that has not been incorporated into common practice).
There are no clinical insights that can be gleaned from statistics pooled from among unrelated treatments. Statistics aggregated from related but disparate treatments obscure the effects of established procedures in the noise of abandoned, experimental, and developing technologies or in the specific biomechanics and innervation of spinal regions.
The disappointing truth is that there are not enough high-quality RCTs to perform a well-powered meta-analysis. We do not, therefore, pool dissimilar studies. Instead, we rely on carefully reasoned systematic reviews that incorporate the breadth of available literature [4]. We are confident that the authors support our call for increased research funding in the field of spine pain to provide clarity and iteratively improve clinical decision-making for clinicians and patients.
Omission and Inaccuracy in Data Extraction
Key studies were omitted from the systematic review, including a high-quality RCT demonstrating the efficacy of lumbar transforaminal ESI [5]. Additionally, data from the strongest RCT on cervical medial branch RFN were inaccurately extracted [6]. The authors incorrectly reported the diagnostic block threshold as "unclear" despite explicit details provided in the original study. Such errors raise concerns about the accuracy and comprehensiveness of the data synthesis.
Technical Fidelity
Proper assessment of interventional procedures requires attention to technical execution [7,8]. Many studies in this review failed to document essential procedural elements, such as imaging guidance, needle placement verification, and RFN lesion parameters. Deviations from technical standards significantly impact clinical outcomes and should have been considered in the review’s methodology.
Recommendations
To improve the validity of future systematic reviews/NMAs, we recommend:
• Grouping studies based on consistent patient populations, spinal regions, diagnoses, and procedural techniques.
• Ensuring data extraction accuracy and transparency.
• Assessing the technical execution of interventional procedures to ensure valid comparisons.
Conclusion
Given these concerns, we do not support developing clinical guidelines or coverage policies based solely on the findings of this systematic review/NMA. We also stress the importance of selecting, analyzing, and aggregating appropriate studies when developing systematic reviews/NMAs and clinical guidelines [9,10]. We call upon researchers, clinicians, and policymakers to recognize the complexity of chronic spine pain and to support expanded research and ongoing access to interventional procedures underpinned by rigorous clinical standards.
* Access the full response, which includes the list of 34 signatories, at https://cdn.ymaws.com/www.ipsismed.org/resource/resmgr/advocacy/25/Multi....
________________________________________
References:
1. Wang X, Martin G, Sadeghirad B, et al. Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials. BMJ 2025;388:e079971. doi:10.1136/bmj-2024-079971
2. Chaimani A, Caldwell DM, Li T, Higgins JPT, Salanti G. Chapter 11: Undertaking network meta-analyses [last updated October 2019]. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.5. Cochrane, 2024. Available from www.training.cochrane.org/handbook (http://www.training.cochrane.org/handbook).
3. Schneider BJ, Doan L, Maes MK, Martinez KR, Gonzalez Cota A, Bogduk N. Systematic review of the effectiveness of lumbar medial branch thermal radiofrequency neurotomy, stratified for diagnostic methods and procedural technique. Pain Med 2020 Jun 1;21(6):1122-1141. doi: 10.1093/pm/pnz349. PMID: 32040149
4. IPSIS Systematic Reviews. https://www.ipsismed.org/page/SystematicReviews [Accessed 3/18/2025]
5. Ghahreman A, Ferch R, Bogduk N. The efficacy of transforaminal injection of steroids for the treatment of lumbar radicular pain. Pain Med 2010 Aug;11(8):1149-68. doi: 10.1111/j.1526-4637.2010.00908.x. PMID: 20704666
6. Lord SM, Barnsley L, Wallis BJ, McDonald GJ, Bogduk N. Percutaneous radio-frequency neurotomy for chronic cervical zygapophyseal-joint pain. N Engl J Med 1996 Dec 5;335(23):1721-6. doi: 10.1056/NEJM199612053352302. PMID: 8929263
7. Practice Guidelines for Spinal Diagnostic and Treatment Procedures, 2nd edition. Bogduk N, ed.San Francisco: International Spine Intervention Society; 2013.
8. Maus TP, Cohen I, McCormick ZL, Schneider BJ, Smith CC, Stojanovic MP, Waring PH (Eds). Technical Manual and Atlas of Interventional Pain and Spine Procedures. International Pain and Spine Intervention Society; 2024.
9. Multisociety letter to the agency for healthcare research and quality: serious methodological flaws plague technology assessment on pain management injection therapies for low back pain. Pain Med 2016 Jan;17(1):10-5. doi: 10.1111/pme.12934. PMID: 26400156
10. McCormick ZL, Vorobeychik Y, Gill JS, et al. Guidelines for composing and assessing a paper on the treatment of pain: a practical application of evidence-based medicine principles to the Mint randomized clinical trials. Pain Med 2018 Nov 1;19(11):2127-2137. doi: 10.1093/pm/pny046. PMID: 29579232
Competing interests: No competing interests
Dear Editor
We thank readers for their feedback and welcome the opportunity to respond.
Robinson et al. claim we categorized epidural lidocaine injections in one trial (1) as a sham control. We did not. As per Supplement Table 9 in our review, we categorized this arm as an epidural injection of lidocaine. Another study compared an extradural lumbar injection of steroids vs an extradural lumbar injection of saline, in which both arms first received infiltration of local anesthetic to minimize injection site pain.(2) No intervention categorized as sham in our review included delivery of medication.
We pooled epidural injections whether they were administered in the cervical or lumbar region, or whether they were administered transforaminal, intralaminar, or caudally. We presented results separately for chronic axial and chronic referred spinal pain and pooled across different types of conditions (e.g. disc herniation, lumber stenosis) within these categories. These decisions were made after subgroup analyses showed no difference in treatment effects based on location, or approach, of administration. These results are provided in Supplementary Table 24 in our review.
Current guidance from Cochrane advises that an I2 of 30% to 60% may represent moderate heterogeneity, and an I2 50% to 90% may represent substantial heterogeneity.(3) Interpretation requires consideration of effect size, direction, and strength of evidence. Our approach followed the GRADE framework,(4) where unexplained heterogeneity led to downgrading the certainty of evidence.
Further, the four highest I2 values highlighted by Robinson et al. include comparisons with radiofrequency ablation procedures. We found a credible subgroup analysis based on provider blinding status for conventional, pair-wise meta-analysis of radiofrequency ablation vs. sham procedures, that fully explained heterogeneity. Specifically, 7 trials that did blind providers showed a pooled effect on pain relief of -0.23 cm on a 10 cm VAS (-0.60 to 0.14) that was associated with an I2 of 0%. (Supplement Table 24) We were underpowered for this association in our network meta-regression, but the same trend emerged. (Supplement Table 32)
We acknowledged the possibility that future research may identify differential effects of interventional procedures based on subtypes of patients.
Dr. Sharvill points out that our findings do not apply to acute spine pain, which is correct and highlighted in the associated guideline.(5)
We enthusiastically endorse the call by Dr. Johnson for greater support to promote the development of clinician-scientists in the field of pain medicine, and for large, rigorously conducted randomized trials to inform effectiveness of commonly administered interventions.
Xiaoqin Wang, Liang Yao, and Jason W. Busse; On behalf of review authors
Department of Anesthesia, McMaster University, Hamilton, ON, Canada
References
1. Friedly JL, Comstock BA, Turner JA, et al. A Randomized Trial of Epidural Glucocorticoid Injections for Spinal Stenosis. New England Journal of Medicine [Internet]. 2014; 371(1): 11–21.
2. Helliwell M, Robertson JC, Ellis RM. Outpatient Treatment of Low Back Pain and Sciatica by a Single Extradural Corticosteroid Injection. Int J Clin Pract. 1985;39(6):228–31.
3. Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0. Chapter 9.5.2 Identifying and measuring heterogeneity (Available at: https://handbook-5-1.cochrane.org/chapter_9/9_5_2_identifying_and_measur...)
4. Guyatt G, Zhao Y, Mayer M, et al. GRADE guidance 36: updates to GRADE's approach to addressing inconsistency. J Clin Epidemiol. 2023; 158: 70-83.
5. Busse JW, Genevay S, Agarwal A, et al. Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline. BMJ. 2025 Feb 19;388:e079970. doi: 10.1136/bmj-2024-079970. PMID: 39971339.
Competing interests: No competing interests
Dear Editor,
We understand that the field of clinical chronic pain is inherently subjective. The prior prescription opioid crisis has shifted pain medicine from medical to interventional management given the weariness of physicians to prescribe opioids. This shift is driven by concerns that the risks of opioid prescriptions significantly outweigh the benefits long-term. Compared to other medical disciplines, chronic pain is rather young, yet this does not justify with the prevalence of insufficient controls or shams, as seen in this study involving interventional procedures. This study has fundamental issues. First, in a sham, there should be no active or any medication injected, but the authors included a study where lidocaine was injected as part of the sham Lidocaine has anti-inflammatory properties, meaning that the participants in the sham group experienced improvements.[1–5] The extent of these improvements remains unclear as no real sham or control was used.[1]
As the authors of the paper must know, in network meta-analysis, the controls should be comparable. However, the comparison of a study using lidocaine versus saline as a sham violates this consistency assumption, and this is just one mere example.[1,6] The authors provide an overgeneralization without differentiation by grouping procedures such as cervical, lumbar, caudal, and transforaminal epidurals, surmising that all these procedures are equivalent. Grouping non-equivalent procedures clearly dilutes the effect size. Furthermore, it leaves one to question if the “trained reviewers” were indeed the most appropriate and what this training entailed. The methodologists were purported to have “general medicine” knowledge, which is quite vague. It must be noted that to even be an internist with general medicine knowledge requires a minimum of 7 years of further education and training following college.
Moreover, grouping procedures such as spinal stenosis and radiculopathy dilutes the effect size. Now, how can conclusions be made when the comparisons are fraught with heterogeneity. In the supplemental data, heterogeneity was measured in 8 comparisons and results were 0%, 4.6%, 42.3%, 55.3%, 59.4%, 87.9%, 88.8%, and 91.1%. Five out of 8 are greater than 50% which is considered highly heterogeneous. Finally, the methodology was sufficiently lenient allowing all patients from the selected studies to be included. Overall, by grouping, the authors overlook the prognostic factors that can affect outcomes such as lower pain score, body mass index, and depression.[7]
While this meta-analysis raises important discussion points, its methodological flaws, inappropriate generalizations, and failure to reflect real-world clinical practice limit the validity of its conclusions. Instead of broadly questioning the efficacy of interventional procedures, future research should focus on refining patient selection criteria, standardizing procedural techniques, ensuring proper training of providers, and assessing long-term outcomes to better understand the durability of these treatments. The authors reported that the procedures are costly, but to do nothing or do surgery, is by no means a comparison of the cost given the outsized financial burden pain has on society.[8] Otherwise, like the controls used, this may just be a sham.
References:
1. Friedly JL, Comstock BA, Turner JA, Heagerty PJ, Deyo RA, Sullivan SD, et al. A Randomized Trial of Epidural Glucocorticoid Injections for Spinal Stenosis. New England Journal of Medicine [Internet]. 2014 Jul 3 [cited 2025 Feb 23];371(1):11–21. Available from: https://www.nejm.org/doi/full/10.1056/NEJMoa1313265
2. Weinschenk S, Weiss C, Benrath J, von Baehr V, Strowitzki T, Feißt M. Anti-Inflammatory Characteristics of Local Anesthetics: Inhibition of TNF-α Secretion of Lipopolysaccharide-Stimulated Leucocytes in Human Blood Samples. Int J Mol Sci [Internet]. 2022 Mar 1 [cited 2025 Feb 23];23(6). Available from: https://pubmed.ncbi.nlm.nih.gov/35328706/
3. Su D, Gu Y, Wang Z, Wang X. Lidocaine attenuates proinflammatory cytokine production induced by extracellular adenosine triphosphate in cultured rat microglia. Anesth Analg [Internet]. 2010 [cited 2025 Feb 23];111(3):768–74. Available from: https://pubmed.ncbi.nlm.nih.gov/20686009/
4. Yuan T, Li Z, Li X, Yu G, Wang N, Yang X. Lidocaine attenuates lipopolysaccharide-induced inflammatory responses in microglia. J Surg Res [Internet]. 2014 [cited 2025 Feb 23];192(1):150–62. Available from: https://pubmed.ncbi.nlm.nih.gov/24952412/
5. Van Der Wal SEI, Van Den Heuvel SAS, Radema SA, Van Berkum BFM, Vaneker M, Steegers MAH, et al. The in vitro mechanisms and in vivo efficacy of intravenous lidocaine on the neuroinflammatory response in acute and chronic pain. Eur J Pain [Internet]. 2016 May 1 [cited 2025 Feb 23];20(5):655–74. Available from: https://pubmed.ncbi.nlm.nih.gov/26684648/
6. Helliwell M, Robertson JC, Ellis RM. Outpatient Treatment of Low Back Pain and Sciatica by a Single Extradural Corticosteroid Injection. Int J Clin Pract. 1985;39(6):228–31.
7. Cohen SP, Doshi TL, Kurihara C, Reece D, Dolomisiewicz E, Phillips CR, et al. Multicenter study evaluating factors associated with treatment outcome for low back pain injections. Reg Anesth Pain Med [Internet]. 2022 Feb 1 [cited 2025 Feb 23];47(2):89–99. Available from: https://pubmed.ncbi.nlm.nih.gov/34880117/
8. Gaskin DJ, Richard P. The economic costs of pain in the United States. J Pain [Internet]. 2012 Aug [cited 2024 Mar 28];13(8):715–24. Available from: https://pubmed.ncbi.nlm.nih.gov/22607834/
Competing interests: No competing interests
Dear Editor
One major point that needs raising here is NHS care. Services for acute severe non cancer back pain with disabling radicular pain are very unlikely to see any action or intervention within 3 months. There is a risk that the headline of this article may lead those controlling funding of treatments (Integrated care boards in England) may therefore stop funding acute pain treatments that will help because of the timeline and the pain is then defined as 'chronic'. Is it possible to know how many of the UK population with Chronic back pain missed the boat with help before their pain became chronic?
Once chronic pain has set in there is an expectation that interventions will help despite the evidence but we are often not brave enough to advise that the solution in reducing pain is not surgery, injections or drugs (or referral to poi clinics often) but exercise and lifestyle. As a result many life years of active enjoyable life (and work) are lost as people wait for yet another pain clinic appointment and intervention expecting a cure.
Competing interests: I have had disabling acute back pain but was able to afford to pay for urgent private MRI
Dear Editor,
I am writing in response to the meta-analysis “Common Interventional Procedures for Chronic non-cancer spine pain: a systematic review and meta-analysis of randomised controlled trials” by Wang X, Asif MS, Khalid MF, et al. The authors have made an excellent attempt at distilling a heterogeneous collection of studies into a more coherent analysis, but I think it is unfortunate that the number of studies and their sizes was filed in the supplementary information rather than available in the body of text. When these supplementary data sets are reviewed in detail there are some procedures with a very limited number of studies many studies with small patient cohorts. This should not take away from the work of the authors, instead this illustrates of how we in the sub-specialty of Pain Medicine have failed our patients by the paucity of large-scale studies of interventional procedures within the field.
Until we reach a time where every pain department in the country has the enthusiasm to help design and participate in large randomised controlled trials of interventions, we will continue to find our practice restricted to an ever-shrinking pool of procedures, not because none of our interventions benefit our patients, but because few have been studied appropriately rigorously.
Within the UK anaesthetic workforce there is not same culture of postgraduate research doctorate education that exists in other specialties such as oncology. Anaesthetic trainees often decide ten years or more after graduating medical school to subspecialise within the field, at which point they may feel they have “missed the boat” in terms of developing research portfolios and have commitments both in and outside of work that preclude them taking multiple years out of training to undertake a PhD.
I would call on the Royal College of Anaesthetists and the Faculty of Pain Medicine to develop a program of support for senior trainees and early years consultants who wish to upskill in research and academia. This will ensure in the context of both chronic and acute pain management we are able to deliver the best in evidence-based medicine to our patients in the years to come.
Competing interests: No competing interests
Re: Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials
Dear Editor
We thank readers for their comments.
Rittenberg advises that no outcome data between any of the 132 randomized trials our review identified should be pooled, as differences in how the same types of procedures are administered, where along the spine they are administered, and types of complaints will introduce unacceptable heterogeneity.
Our panel experts were less certain, and we therefore elected to explore pooling to maximize power to detect effects, improve generalizability, and facilitate subgroup analyses. (1) Example subgroup analyses of pairwise meta-analyses included: 1] whether pain relief afforded by epidural injection of local anaesthetic vs. local anaesthetic + steroids differed based on approach (interlaminar, caudal, transforaminal, unclear, or a combination of transforaminal and interlaminar) (Figure 1; test of interaction p-value 0.7), and 2] whether pain relief afforded by joint radiofrequency vs. joint targeted injection of local anaesthetic + steroid differed based on whether joint radiofrequency was pulsed or non-pulsed (Figure 2; test of interaction p-value 0.3). We found no evidence of credible subgroup effects based on these or other factors. (Supplementary Table 24)
We then explored if pooling introduced problematic heterogeneity to our network meta-analyses. We found no evidence of credible subgroup effects (e.g., Supplementary Tables 28, 38, 48, 58), statistical variability across treatment effects in our outcome networks, or in assessments of between-study variances within the closed loops of evidence. (2)
Rittenberg claims that we omitted an eligible trial from our review (Ghahreman et al., 2010). (3) We did not. Our review focussed on chronic spine pain, whereas this trial enrolled mostly patients with acute pain (80 of 150), and did not report results separately for patients with chronic pain.
Rittenberg suggests that the 24-patient trial by Lord et al. (1996) is the “strongest RCT on cervical medial branch radiofrequency neurotomy” with clearly defined diagnostic blocks.(4) However, the Bone and Joint Decade 2000–2010 Task Force on Neck Pain concluded this trial was scientifically inadmissible for their review due to critical limitations, including: 1] selection of patients “on the basis of a nonvalidated response to facet blocks”; 2] small sample size failed to balance critical prognostic factors – 83% of subjects in the sham group were involved in litigation due to their neck injury vs. 33% of patients in the intervention group (5); and 3] blinding was in doubt as 42% of patients in the active treatment arm developed long-term anesthetic or dysaesthetic areas of skin vs. 0% of control patients. (6)
We extracted technical details reported for administering interventional procedures from all eligible trials (Supplementary table 6).(2) We agree with Rittenberg that trial authors should clearly report such details, and that there is variability in this regard.
Our previous reply addressed concerns raised by Gharibo. No intervention designated as sham in our review was pharmacologically active. We assessed risk of bias for all included trials and conducted subgroup analyses to explore for differences in treatment effect between trials at high and low risk of bias. We found a credible subgroup effect for radiofrequency ablation vs. sham procedures that fully explained heterogeneity. Specifically, 7 trials that did blind providers showed a pooled effect on pain relief of -0.23 cm on a 10 cm VAS (-0.60 to 0.14) that was associated with an I2 of 0%. (Supplement Table 24)
Figure 1: Example subgroup analysis for chronic radicular spine pain [https://mcmasteru365-my.sharepoint.com/:f:/g/personal/wangx431_mcmaster_...
Figure 2: Example subgroup analysis for chronic axial spine pain[https://mcmasteru365-my.sharepoint.com/:f:/g/personal/wangx431_mcmaster_...
We agree with Tafur that researchers and clinicians should critically evaluate current practices and seek reliable and relevant evidence to guide treatment choices. We hope our work (2,8) will prove helpful in this regard.
Xiaoqin Wang, Liang Yao, and Jason W. Busse; On behalf of review authors
References
1. Gotzsche PC. Why we need a broad perspective on meta-analysis. It may be crucially important for patients. BMJ. 2000; 321(7261): 585-6.
2. Wang X, Martin G, Sadeghirad B, et al. Common interventional procedures for chronic non-cancer spine pain: a systematic review and network meta-analysis of randomised trials. BMJ. 2025; 388: e079971.
3. Ghahreman A, Ferch R, Bogduk N. The efficacy of transforaminal injection of steroids for the treatment of lumbar radicular pain. Pain Med 2010; 11(8): 1149-68.
4. Lord SM, Barnsley L, Wallis BJ, McDonald GJ, Bogduk N. Percutaneous radio-frequency neurotomy for chronic cervical zygapophyseal-joint pain. N Engl J Med 1996; 335(23): 1721-6.
5. Harris I, Mulford J, Solomon M, van Gelder JM, Young J. Association between compensation status and outcome after surgery: a meta-analysis. JAMA. 2005; 293(13): 1644-52.
6. Carragee EJ, Hurwitz EL, Cheng I, et al; Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. Treatment of neck pain: injections and surgical interventions: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its Associated Disorders. Spine (Phila Pa 1976). 2008; 33(4 Suppl): S153-69.
7. Akl EA, Sun X, Busse JW, et al. Specific instructions for estimating unclearly reported blinding status in randomized trials were reliable and valid. J Clin Epidemiol 2012; 65: 262-7.
8. Busse J W, Genevay S, Agarwal A, Standaert C J, Carneiro K, Friedrich J et al. Commonly used interventional procedures for non-cancer chronic spine pain: a clinical practice guideline BMJ. 2025; 388: e079970
Competing interests: No competing interests