2025 国际多专业工作组共识实践指南：骶髂关节复合体疼痛

Abstract

Background: The past two decades have witnessed tremendous growth in the appreciation and treatment of sacroiliac joint (SIJ) complex pain, including anatomical dissections that shed light on innervation, an appreciation for the contribution of extra-articular components to SIJ complex pain, the advent of radiofrequency ablation (RFA) and a host of minimally invasive surgical techniques. Yet, there is no standardization on diagnosis and treatment paradigms.

Methods: In February 2023, the Boards of Directors for the American Academy of Pain Medicine (AAPM) and American Society of Regional Anesthesia & Pain Medicine (ASRA-PM) approved the development of multispecialy guidelines on SIJ complex pain. Thirty partner organizations with clinical and scientific interests in SIJ complex pain were identified, and formal letters of request-for-participation were sent to each, along with a request for nominees to serve on the committee. Twenty five organizations agreed to participate in addition to the Departments of Defense and Veterans Affairs. A steering committee developed 21 questions, which spanned criteria for diagnosis, non-interventional and inter-ventional treatments including surgery, technical parameters on how to optimize results, and what constitutes positive outcomes. Questions were methodically assigned to specialized modules comprising 4-5 members with complementary expertise, who collaborated with the Subcommittee Lead and one of three Committee Chairs to develop preliminary drafts. Following thorough revisions, these drafts were subsequently submitted to the full committee for comprehensive review. A modified Delhi method was used in which the answers to questions were sent to the committee en bloc and comments were returned in a non-blinded fashion to the Chairs, who incorporated the comments and sent out revised versions until consensus was achieved. During a committee meeting before commencement, it was agreed that recommendations would be noted when there was >50% agreement among committee members, but that a formal recommendation would require ≥75% consensus.

Results: Twenty-one organizations formally endorsed the guidelines. The American Society of Anesthesiologists, and American Academy of Physical Medicine & Rehabilitation, and the North American Spine Society affirmed the benefit of the guidelines but did not officially endorse them. The American Academy of Neurology declined to affirm the benefit of the guidelines citing "lack of relevance to their membership." Per policies, while the Departments of Defense and Veterans Affairs did not formally review the guidelines for endorsement, their representatives approved them. In addition to being endorsed or the benefit affirmed by all voting organizations, complete consensus from committee members was obtained on all 21 questions. On 2 recommendations, there were dissensions from 3 societies who thought that selecting patients for sacral lateral branch RFA and minimally invasive fusion should be based on ≥75% relief from 2 blocks instead of at least 50% relief from a single block before RFA, and greater than 50% pain relief with documented functional improvement after a single block before fusion. One additional society (Latin American Society of Regional Anesthesial) abstained on the statement that the evidence is stronger for extra-articular than intra-articular injections. The committee found that a battery of physical tests has reasonable sensitivity, but lower specificity, for identifying intra-articular but not extra-articular pathology, with negative tests having greater predictive value than positive ones. Intra-articular injections have diagnostic validity for SU intra-articular, but not extra-articular pain. There is unclear or negative evidence for imaging. The prevalence rates of intra-articular and extra-articular pathology are comparable, with both intra- and extra-articular steroid injections providing at least 4 weeks of relief in well-selected patients. However, the evidence is slightly stronger for extra-articular corticosteroid injections to provide short-term relief. The evidence base for non-interventional therapies is indirect, extrapolated mostly from low back pain studies. There is weak evidence supporting dextrose-based prolotherapy and platelet-rich plasma to provide at least 3 months of pain relief. There is strong evidence for sacral lateral branch RFA to provide relief for at least 6 months in individuals with extra-articular pathology, with face validity and indirect evidence from randomized trials supporting sacral lateral branch blocks as a prognostic tool. There is stronger evidence for larger lesions or more aggressive lesioning strategies than for less stringent techniques. There is weak evidence to support non-steroidal anti-inflammatory drugs to prevent neuritis after RFA, and in most cases anticoagulation does not require cessation in the periprocedural period. With an aggressive lesioning strategy, sensory stimulation provides minimal therapeutic benefit, with weak, extrapolated evidence that motor stimulation can provide safety benefit. The cutoff to designated diagnostic or prognostic blocks as positive is most commonly set at 50%, with higher values not shown to improve outcomes for more definitive procedures; for therapeutic treatment outcomes, the evidence supports a lower threshold of ≥30% pain relief or meaningful benefit on pain-pain outcomes (eg, opioid cessation) for designation of a positive response. For carefully selected patients with intra-articular SIJ complex pain based on controlled blocks who have failed conservative therapies, there is weak or very weak evidence that minimally invasive SIJ fusion can provide benefit for at least one year.

Conclusions: SIJ complex pain remains an underappreciated source of chronic low back pain, affecting between 15% and 30% of patients with axial pain predominantly below L5. Answers to many questions were limited by low-quality evidence, indicating the need for better research. SIJ complex pain is a multifarious condition (ie, pain can be from different portions of both the intra- and extra-articular components of the joint) for which an interdisciplinary, multimodal treatment plan can optimize treatment outcomes.

Keywords: sacroiac joint; radiofrequency; lateral branch; minimally-invasive fusion; corticosteroid injection; extra-articular; provocative test.

Introduction

The sacroiliac joint (SIJ) complex is a multiplex structure. The SIJ itself is composed of the lateral parts of the sacrum and medial portions of the iliac bones, which form the joint space enclosed by a fibrous capsule and synovial lining. The superior and dorsal portions of the joint itself are amphiar-throdial with the sacral and iliac surfaces separated by

fibrocartilage. The small antero-inferior portion of the joint is diarthrodial, more mobile, and contains hyaline cartilage on the bony articular surfaces. The SIJ is supported by an extensive network of posterior and anterior ligaments, which in combination with the joint itself, make up the SIJ complex, all of which contain nociceptors and can serve as pain generators.$^{1-4}$ Given this complexity, sacroiliac joint (SIJ) pain, or

Pain Medicine, 2025, Vol. 00, No. 0

more fittingly SIJ complex pain, remains one of the most challenging etiologies of axial low back pain (LBP) to understand and manage.

Data on the prevalence of SIJ complex pain vary widely, as there is no reference standard besides diagnostic injections, which are subject to false-positive and false-negative results. Nearly all diagnostic epidemiological studies evaluated intra-articular (IA) injections, which are associated with inherent diagnostic limitations related to capsular defects, structural communication with the S1 formen, and anatomical variations presenting procedural challenges for consistent intra-articular access.$^{5}$ The prevalence rate of extra-articular (EA) pathology is probably similar, perhaps even higher in younger individuals with post-traumatic pain.$^{6,7}$ Prevalence data are further confounded by the thresholds used to classify diagnostic injections as positive or negative, with more stringent criteria (ie, multiple injections and higher cutoff thresholds) being associated with a lower prevalence (positive rate.$^{8}$ This conceptual framework categorizes SIJ complex pain as a dichotomous variable, whereas in clinical practice, pain originating from spinal structures may be more accurately conceptualized as existing on a continuum due to the presence of multiple concurrent pain generators. Similarly, prevalence rates derived from diagnostic injections are contingent upon clinical selection criteria, with the majority of studies employing diagnostic injections selecting patients based on suggestive clinical presentations (eg, positive provocative maneuvers or mechanical pain localized to the SIJ complex); more rigorous selection criteria invariably elevate the positive rate of diagnostic injections, resulting in higher prevalence estimates. In well-selected populations with mechanical back or buttock pain predominantly below L5 spinous process, SIJ complex pain appears to account for the index pain in about 15%-30% of patients.$^{1,8}$

Difficulties in diagnosis and discrepancies in prevalence rates have translated to a lack of standardization among clinical trials evaluating treatment interventions. There are very few placebo-controlled trials evaluating SIJ injections, with only one small trial (n = 10) evaluating IA injections and two small studies with less than 25 people each, by the same group of investigators, evaluating EA injections, with the longest study following blinded patients for two months.$^{9-11}$ There is only one large study that compared EA to IA injections in a randomized fashion,$^{6}$ widespread disparities in injectate composition and technique in clinical studies, and no randomized-controlled trial(s) (RCT) investigating an IA corticosteroid injection that prescreened patients with diagnostic injections. For joint injections with other substances such as platelet rich plasma and hyaluronic acid, the extant studies are of poor quality and also contain widely divergent techniques, selection criteria and outcome measures.$^{12,13}$

The strongest evidence may be for radiofrequency ablation (RFA), which has shown superiority over sham RFA, standard of care, exercise therapy and IA corticosteroids. However, similar to joint injections, the results of these studies vary considerably, and there is no consensus on patient selection, technique and nerves targeted, or outcome measures.$^{14-18}$ Clinical studies have evaluated pulsed radiofrequency, internally-cooled RFA (icRFA), bipolar RFA, conventional RFA (RFA) with a wide variety of methods, and combination RFA (eg, bipolar and unipolar or internally-cooled and conventional) strategies.$^{17}$ In addition to branches of the dorsal rami, RFA has been performed inside of the

joint and in the posterior ligaments.$^{19,20}$ Studies have used both IA and EA injections as well as sacral lateral branch injections (SLBBs) in isolation to select patients, with one case report describing the performance of SLBB in isolation before icRFA in a cancer patient.$^{21-25}$ Radiofrequency ablation studies have targeted nerves ranging from the L4 dorsal ramus to the S4 lateral branch, in multiple permutations including selecting nerve levels based only on sensory stimulation.$^{17,21,25}$ These studies have used a variety of different cutoff thresholds ranging from ≥2-points on an 11-point numerical rating (pain) scale (NRS) to ≥80% relief for designating screening injections as positive,$^{21,22,27}$ with a similar range from ≥2-points or 30% to >70% for treatment outcomes.$^{6,11,14,16,26,27}$

There are discrepancies in the interpretation of the literature as well, with some reviews touting positive evidence for injections and RFA, while others have been more skeptical in their analysis.$^{17,26}$ Collectively, this lack of standardization and consensus regarding procedural performance, evidence levels, what constitutes SIJ complex pain, which affects prevalence reports and selection for treatment, and even how to define procedural success, highlights the need for multispecality guidelines. The charge of this committee was therefore to synthesize the evidence to develop pragmatic guidelines that can be used to inform clinical care with an emphasis on optimizing patient outcomes; identify evidence gaps that can be used to steer research; serve as a foundation for standardizing care of the patient with SIJ complex pain; and to assist regulators with developing authorization standards and clinical practice pathways.

Methods

The decision to convene a multispecialty and multinational Sacroliac Joint Complex Pain Working Group (SIJ-WG) to develop SIJ complex intervention guidelines was approved by the American Academy of Pain Medicine (AAPM) and American Society of Regional Anesthesia and Pain Medicine (ASRA-PM) in July 2023. Thirty stakeholder academies and societies, as well as other organizations (eg, U.S. Departments of Defense and Veteran Affairs) with a vested interest in SIJ complex interventions were identified, and formal request-for-participation and member nomination letters were sent to those organizations in August 2023. Twenty-seven organizations agreed to participate, and three invited professional societies declined to participate (Lumbar Spine Research Society, LSRS; American Society for Interventional Pain Physicians, ASIPP; and American Society of Spine Radiologists, ASSR). Organizations were asked to consider a candidate's expertise, clinical experience, academic interests, and diversity in their nomination process. Each society nominated a single representative member (See Appendix SA for a list of participating academies, societies, and respective representatives). For the Departments of Defense and Veterans Affairs representatives, the Chairperson of the Department of Anesthesia at the Unformed Services University of the Health Sciences and Director of the VA National Pain Management Program nominated individuals.

The SIJ-WG was charged with preparing guidelines on the use of IA SIJ injections, EA SIJ injections, SLBBs, sacral lateral branch radiofrequency ablation (SLBRFA), cryoablation, SIJ fusion, as well as conservative therapy that spanned the entire spectrum of care to include patient selection,

4

Pain Medicine, 2025, Vol. 00, No. 0

optimizing accuracy, interpreting results, and risk mitigation. Questions and formats were developed by the Working Group Committee Co-Chairs (Z.L.M., R.W.H., and S.P.C.) during iterative meetings. These questions and formats were refined based on feedback by the Executive Committee (D.T., A.C., L.K., and J.V.Z.). Once unanimous consensus was achieved by the Working Group Committee Co-Chairs and Executive Committee, input was gathered from the full Working Group Committee and further refined during the initial video-conference call. The resulting final unanimous consensus version was comprised of 21 distinct questions to be addressed within the guidelines.

The Working Group Committee was then divided into sub-committees composed of four to five members responsible for developing the answer(s) to each of the final 21 questions. One or two persons were designated as the “leads” responsible for task delegation. Once a subcommittee came to unanimous consensus on the answer to their assigned question, including the grade of evidence associated with relevant statements and recommendations, the Working Group Co-Chairs assisted with editing and provided feedback regarding content. Through iterative discussion of the supporting evidence, unanimous consensus was achieved amongst the subcommittee members and the Working Group Co-Chairs with regard to the content of the answer and the corresponding statements and recommendations. Next, the subcommittee's work product was sent to the entire Working Group Committee for open-forum comments and revisions. A modified Delphi method was used to tabulate comments, incorporate changes, and converge the answers toward consensus over rounds of teleconference or electronic correspondence. At the initial conference call, the Working Group Committee decided that >50% panel agreement was sufficient to report a statement or recommendation, but ≥75% agreement was required for consensus, consistent with the Lumbar Facet Intervention and Cervical Facet Intervention Guidelines.$^{28,29}$ After the Working Group Committee completed a draft of the guidelines, the document was sent to participating organizations' Boards of Directors for approval, with only minor changes permitted at this stage. For organizational agreement, the Working Group Committee agreed that consensus required at least ≥75% agreement, consistent with the Lumbar Facet Intervention and Cervical Facet Intervention Guidelines.$^{28,29}$ and at both the committee and organizational levels, dissensions and abstentions were tabulated for each question.

Search engines used during composition of the various sections included PubMed, Embase, Medline, Google Scholar, SCOPUS, and Cochrane Database of Systematic Reviews, in addition to examination of the reference sections of all manuscripts. Additional peer-reviewed articles were identified by searching in topically-related new journals that are not yet indexed by Science Citation Index or found within PubMed. There were no limitations on language or types of articles used to develop the guidelines, such that experimental studies were considered for the sections on physical exam, anatomy and technical parameters, and case reports were considered for sections pertaining to risk mitigation and complications. Keywords used to address guideline topics were tailored to individual questions and included, for example, "sacroiliac", "lateral branch", "radiofrequency", "pulsed radiofrequency", "denervation", "ablation", and "arthritis" for questions pertaining to radiofrequency. Searches were comprehensive but a formal systematic review was outside of the scope of the

|Certainty of net benefit|Magnitude of net benefit|||| |---|---|---|---|---| |Certainty of net benefit|Substantial|Moderate|Small|Zero/negative| |High|A|B|C|D| |Moderate|B|B|C|D| |Low|Insufficient||||

charter. Literature searches were performed through March 2025 to capture topical studies published following initial searches. In accordance with the Lumbar Facet Intervention and Cervical Facet Intervention guidelines,$^{28,29}$ conclusions for each topic were graded on a scale from A to D, or as insufficient, according to modified US Preventative Services Task Force grading of evidence guidelines, with the level of certainty rated as high, medium or low (Tables 1-3).$^{30}$ The guidelines followed the recommended domains of the Appraisal of Guidelines for Research & Evaluation II (AGREE-II) checklist (see Appendix SB).$^{31}$

This system, which has been modified for use in interventional pain and spine medicine guidelines drafted by the ASRA-PM, AAPM, American Society of Anesthesiologists (ASA), ASIPP, and the International Neuromodulation Society (INS),$^{32-35}$ was chosen over others because of its flexibility,$^{36,37}$ which permits high-grade recommendations in the absence of high-quality level I studies, which are challenging to conduct for invasive procedures.$^{38}$

Epidemiology

Q1. What is the prevalence of sacroiliac joint complex pain in individuals with chronic low back pain?

SIJ complex pain is commonly estimated to affect approximately 15%-30% of individuals with chronic, non-radicular LP predominantly below L5.$^{1}$ In individuals with suspected SIJ complex pain based on history and physical exam, the prevalence of true SIJ complex pain is similar, estimated at 20%-30% based on the studies using diagnostic injections.$^{8}$ Prevalence increases with older age, as chronic LBP is more common among older adults.$^{39}$ Radiologic imaging has been used to evaluate the SIJs; although it has high specificity, it has low sensitivity in identifying individuals with non-inflammatory, mechanical SIJ complex pain, making it a poor screening tool (see Question 3).$^{40}$ Despite their limitations, diagnostic injections with local anesthetic are currently the reference standard for estimating the true prevalence rate of SIJ complex pain.$^{5,41}$

It is important to note that while diagnostic IA injections are the current reference standard, this method is imperfect due to communication of the SIJ with the S1 foramen, the potential for dorsal leakage of injectate from the joint, and the presence of anterior capsular defects in a portion of individuals, which can result in anterior leakage of injectate.$^{5,42,43}$ Therefore, a technically precise IA SIJ injection can still result in off-target anesthetic spread, with the potential for a false-positive test outcome. Recognizing these inherent limitations, research employing single IA SIJ injections suggests the prevalence of SIJ pain to be between 29% and 63%, whereas studies using two injections have recorded rates ranging from 10%- to 33%. However, these rates reflect the prevalence of SIJ pain in individuals already suspected to have this diagnosis

Pain Medicine, 2025, Vol. 00, No. 0

Table 2. What the grades of evidence mean and suggestions for practice.

|Grade|Definition|Suggestions for practice| |---|---|---| |A|The committee recommends this treatment, test or strategy to improve outcomes. There is high certainty that the net benefit is substantial.|Offer or provide this service| |B|The committee recommends this treatment, test or strategy to improve outcomes. There is high certainty that the net benefit is moderate or there is moderate certainty that the net benefit is moderate to substantial.|Offer or provide this service| |C|The committee recommends selectively offering or providing this treatment, test or strategy to improve outcomes to individual patients based on professional judgment and patient preferences. There is at least moderate certainty that the net benefit is small.|Offer or provide this service for selected patients depending on individual circumstances| |D|The committee recommends against the test or intervention. There is moderate or high certainty that the service has no net benefit or that the harms outweigh the benefits.|Discourage the use of this service| |I|The committee concludes that the current evidence is insufficient to assess the balance of benefits and harms of the intervention. Evidence is lacking, of poor quality, or conflicting, and the balance of benefits and harms cannot be determined.|Read the clinical considerations section of the recommendation. If the treatment or service is offered, patients should understand the uncertainty about the balance of benefits and harms|

Table 3. Levels of certainty regarding net benefit.ª

|Level of certainty|Description| |---|---| |High|The available evidence usually includes consistent results from well-designed, well-conducted studies in representative populations with suspected SIJ complex pain and/or compelling evidence from non-randomized studies. The studies assess the effects of the treatment, test, or other intervention on treatment or other relevant outcomes. The conclusion is therefore unlikely to be strongly affected by the results of future studies.| |Moderate|The available evidence is sufficient to determine the effects of the intervention on outcomes, but confidence in the estimate is constrained by such factors as:The number, size, or quality of individual studies;Inconsistency of findings across individual studies;Limited generalizability of findings to individuals with suspected SIJ complex pain;High likelihood of bias;Lack of coherence in the chain of evidence.As more information becomes available, the magnitude or direction of the observed effect could change, and that change may be large enough to alter the conclusion.| |Low|The available evidence is insufficient to assess effects on treatment and other outcomes of interest. Evidence is insufficient because of:The limited number or size of studies;Important flaws in study design or methods;Inconsistency of findings across individual studies;Gaps in the chain of evidence;High likelihood of bias;Findings not generalizable to individuals with suspected SIJ complex pain;Lack of information on important outcome measures.More information may allow estimation of effects on treatment outcomes.|

ª Note, the levels of certainty described in the definitions for the grading of evidence in Table 2 refer solely to the magnitude of benefit attributed to the intervention, while the levels of certainty in Table 3 consider the basis of evidence for the recommendation and the likelihood the recommendation will be affected by future studies. Whereas the two are related, the grading of evidence and rating of certainty were considered separately.

based on history and physical examination that included positive SIJ provocation tests (Patrick's Gaenslen's, Yeoman's, distraction, compression, sacral pressure, and sacral thrust tests), and do not estimate the prevalence of SIJ pain in the overall population of patients with LBP. Furthermore, relaxing the criteria from ≥75% to ≥50% pain relief increases the prevalence estimates. The addition of corticosteroids to diagnostic injections also significantly increases the reported prevalence, with one systematic review finding that when corticosteroids are added to local anesthetic, the positive rate increased from 35% to 49% using a 75% index pain relief threshold.$^{8}$ This is likely due mostly to greater potential for off-target spread of injectate, as adding corticosteroid to the IA injectate increases the total volume and therefore spread outside of the joint space itself though in some cases systemically absorbed

steroids can provide quick pain relief, and local steroids can also enhance the effects of local anesthetic. In studies using sedation, the estimated prevalence was higher, which is likely the result of a higher false-positive rate due to analgesia, anxiolysis, and muscle relaxation.$^{44}$

The prevalence of SIJ pain in the presence of concomitant pathology is less well-defined. In a 2020 study of individuals with lumbar disc herniations, the prevalence of SIJ dysfunction was estimated to be 33% based on three or more positive provocation tests, suggesting that SIJ pain may be present in one-third of individuals with lumbar radicular pain and MRI findings consistent with disc herniation.$^{45}$ However, these study results must be interpreted with caution given that the reference standard of anesthetic injections were not used to confirm “true” SIJ pain. In addition, SIJ provocation tests

Pain Medicine, 2025, Vol. 00, No. 0

 may also exacerbate pain related to the herniated disc itself, thus resulting in a false-positive test for “true” SIJ pain. Nevertheless, it is likely that SIJ pain (and more comprehensively, sacroiliac joint complex pain) coexists with other sources of LBP in a subset of patients.

Dorsal ligament pain

The long dorsal sacroiliac ligament is anatomically connected to the erector spinae muscle, thoracolumbar fascia, and sacrotuberous ligament, potentially serving as a source of spinal pain.$^{46}$ The dorsal ligament associated with the SIJ has shared innervation with the lower portion of the posterior SIJ through the lateral branches of the dorsal rami of S1-3, with or without contributions from L5 and S4, via the nerve plexus termed the posterior sacral network(PSN)$^{47}$ (Figure 1). Given the shared innervation of the dorsal ligament, pain presentations could potentially vary from isolated dorsal ligament pain, dorsal ligament pain with concurrent SIJ-mediated pain, dorsal ligament pain with concurrent L5-1 facet-mediated pain, and dorsal ligament pain with both concurrent SIJ- and L5-S1 facet-mediated pain. By employing diagnostic injections, it becomes possible to estimate the prevalence of isolated dorsal ligament pain. This estimation could theoretically be performed by identifying patients in a comparative-effectiveness study who exhibited a positive response to SLBBs, with or without subsequent SLBRFA, who showed little to no response to IA SIJ injection. This subset would likely exclude patients experiencing concurrent lumbar facet-mediated pain, and assumes that the IA SIJ injections were administered accurately, with no EA spread.

To date, no published studies provide an accurate estimation of the prevalence of pain exclusively from the dorsal ligament or the coexistence of dorsal ligament and IA SIJ pain, as per the theoretical framework described above. A study conducted by Salman et al. compared the effectiveness of SIJ injection to L4-5 dorsal rami and S1-3 RFA. However, little information regarding prevalence can be extrapolated from this study due to the fact that response to RFA, which is not 100% effective, cannot be used to estimate the prevalence of SIJ dorsal ligament pain, just as the response to an IA steroid injection cannot be used to estimate IA SIJ pain. Moreover, patients in both groups were selected based on response to a high-volume (3 mL) IA SIJ block, which theoretically should not relieve purely IA joint pain. Additionally, data were not presented for patients who transitioned to the RFA group.$^{18}$ Another retrospective study by Yin et al.$^{25}$ reported the outcomes of SLBRFA in patients who responded positively to two deep interosseous ligamentous injections. However, the study did not provide information on the percentage of patients with respective positive responses to the first and second diagnostic injections, which would enable a more accurate estimate of the ligament-related pain prevalence as well as the incidence of false-positive blocks. Although this study does not offer the means to accurately estimate the prevalence of dorsal ligament pain, it does demonstrate that patients with such pain can respond to treatment with SLBRFA. Among the 14 patients included in this retrospective review, 9 (64%) experienced more than 50% pain relief, with 5 (36%) achieving complete pain relief at 6 months following RFA treatment. Given that prior studies evaluating outcomes

Pain Medicine, 2025, Vol. 00, No. 0

in SLBRFA had stringent inclusion criteria with positive diagnostic injections (some using IA SIJ injections), the outcomes in patients responding to SLBRFA (but showing a negative response to IA SIJ injection) remain unknown, necessitating further investigation.$^{48}$

Cohen et al. performed a randomized study in 125 patients with clinically suspected SIJ complex pain in which patients were allocated to fluoroscopically-guided or landmark-guided injections. Among the 60 patients in whom the 3 mL injection with local anesthetic and corticosteroid was limited to the EA dorsal ligaments, 57% experienced at least 50% pain relief lasting over 3 h vs 62% in those in for whom the injection was IA, and 69% in patients for whom injections were both IA and EA. Flaws in this study include the use of corticosteroids during uncontrolled single injections, and landmark-guided injections targeted to the area of maximal pain and tenderness rather distributed throughout the dorsal ligaments.$^{6}$

Sacroiliac joint complex pain following lumbar or lumbosacral fusion surgery

Prior lumbar fusion may increase angular motion and stress across the SIJ complex, and pain may also result from adjacent level disease, increasing the prevalence of SIJ complex pain in individuals who have undergone prior spine surgery.$^{49,50}$ Some patients may also experience positive diagnostic injections after lumbar fusion because a pre-surgical SIJ complex pain diagnosis was missed before fusion surgery.

In a computer simulation of a SIJ using a spine-pelvis model, the angular motion of the SIJ was modeled following L4-5, L4-S1, and L5-S1 fusion. The investigators found that motion increased steadily with each additional level fused and were especially substantial after an L5-S1 fusion (52% increased motion with flexion, 40% increase with extension, 28% increase with bending, and 51% increase with rotation). The most substantial increase in motion was observed with L4-S1 spinal fusion, with an estimated 168% increase in motion with flexion, 132% increase with extension, 95% increase with bending, and 69% increase with rotation.$^{49}$ Literature suggests the prevalence of SIJ complex pain to be between 33% and 59% in individuals who have undergone lumbar spinal fusion; thus, SIJ complex pain may be a contributor to the many etiologies of pain broadly categorized as "failed back surgery syndrome."$^{49}$,$^{49}$

Lumbosacral spinal fusion may also result in violation of both the synovial and ligamentous portions of the SIJ. In a CT study performed in 22 patients who underwent posterior iliac crest bone grafting and experienced persistent donor site pain, 16 were found to have violation of the posterior ligamentous structures, and three exhibited disruptions of the SIJ synovium. On imaging performed at least two years post-procedure, 10 of the 16 patients with ligamentous disruption had mild degeneration and six had severe degeneration vs 0 of 5 joints in patients who experienced no ligamentous or synovial violation. In the three patients who experienced synovial membrane breach, all experienced severe SIJ degeneration.$^{51}$ These findings suggest that iliac crest bone grafting, for which the estimated prevalence of donor site pain ranges between 6% and 39%, may be associated with pain from the dorsal ligament, the IA joint, and the donor site adjacent to the SIJ.$^{32}$

Kennedy et al. described several studies evaluating SIJ pathology in individuals who have undergone lumbar/

lumbosacral spinal fusion.$^{8}$ In a controlled study performed in patients with prior lumbar/lumbosacral spinal fusion, three of six positive provocation tests, and pain below L5 suggestive of SIJ complex pain who underwent two injections of local anesthetic (1 mL of 0.5% bupivacaine or 2% lidocaine) and corticosteroid (1 mL of 40 mg triamcinolone), ≥75% relief of pain within 1-4 h was reported in 21 out of 52 (40%) of patients.$^{53}$ A positive response to the injectate was predicted by unilateral pain that was different from preoperative pain, and more than three positive provocation tests.$^{53}$ These procedures were performed using fluoroscopic guidance and IA needle location confirmed with contrast dye.

One retrospective study using a threshold of >75% relief of index pain after a single injection of local anesthetic and corticosteroid under fluoroscopic guidance confirmed by contrast dye reported a positive rate of 59% (20 out of 34 patients) in individuals with a spinal fusion to the sacrum.$^{54}$ Another retrospective review of individuals with LBP included 28 individuals with a history of spinal fusion who underwent diagnostic IA SIJ injection. Patients evaluated with IA SIJ injection included those with either at least three out of five positive SIJ provocation maneuvers, or two or fewer positive SIJ provocation maneuvers but a negative zyg-apophyseal joint injection. IA SIJ injection was considered positive if at least 75% of the patient's pain was relieved after each injection with local anesthetic. The details regarding the number, type, and method of IA SIJ injection were not provided in this retrospective analysis. Among the 28 individuals with a history of spinal fusion identified in the analysis, 43% (12 out of 28) were identified to have SIJ complex pain, and 80% of those who had spinal fusion and SIJ complex pain had a history of spinal fusion to the sacrum, suggesting that spinal fusion to the sacrum is associated with higher rates of SIJ complex pain.$^{55}$

In a prospective study using local anesthetic IA SIJ injections in individuals with persistent LBP following successful lumbar spinal fusion, 35% of study participants experienced at least 75% relief from a single injection.$^{56}$ Patients who had unilateral pain >6 months in duration not extending below the knee, with tenderness at the sacral sulcus and no imaging evidence of a potential lumbar pain source were included. Injections were fluoroscopically guided using nonionic contrast to confirm IA placement, with 2 mL of 2% lidocaine used as the injectate. An injection was considered positive when contrast dye spread was contained within the joint and the patient experienced at least 75% pain relief within 15 min. The authors found that a pain-free interval of 3 months following surgery predicted a positive injection. This study also showed that L5-S1 spinal fusion patients had an increased frequency of positive injections, presumably because of the increased biomechanical load. Bone scintigraphy uptake in the SIJ was not associated with pain, when using IA SIJ injection as the reference standard.$^{56}$

Summary:

$\bullet$ The prevalence of SIJ-mediated chronic LBP is estimated to be 15%-30% in patients with axial pain predominantly below L5; Grade B evidence, moderate level of certainty.

* The prevalence of dorsal ligament-mediated pain is unknown but based on studies comparing IA and EA SIJ injection response, appears to be at least as common as exclusively IA pain; Grade I evidence.

Pain Medicine, 2025, Vol. 00, No. 0

$\bullet$ The prevalence of SIJ complex pain is suspected to be higher in patients with a prior history of lumbar or lumbosacral spinal fusion with estimated prevalence rates between 33% and 59%; Grade C evidence, low level of certainty.

* Chronic LBP following lumbar or lumbosacral spinal fusion surgery can be related to adjacent segment stress with increased loading and/or shear forces leading to accelerated degeneration, superimposed or missed SIJ-mediated pain prior to lumbar fusion, and/or violation of ligaments or synovial lining during iliac crest bone grafting; Grade C evidence, low level of certainty.

Recommendation:

* Further studies are needed to assess the prevalence of primarily dorsal ligament-mediated pain based on EA injections and SLBBs.

Diagnosis

02. Can history and physical exam identify a painful sacroiliac joint and can history and physical distinguish articular sacroiliac joint pain from dorsal ligament pain?

When assessing pain originating from the SIJ complex, it is important to distinguish between IA pain and dorsal ligament pain, as the treatments may differ (eg, IA injections and fusion for IA pain, EA injections and RFA for dorsal ligament pain). Some studies have identified historical and physical exam features that may help identify patients who will respond to IA diagnostic or therapeutic injections. However, no study has been designed using methodology that assesses the relative predictive value of elements of history and/or physical exam that help distinguish between IA SIJ vs dorsal ligament sources of SIJ complex pain. Since there are numerous parts of the SIJ complex that can be pain generators (eg, upper, middle, lower, dorsal and ventral ligaments, bony surfaces, synovial lining and fibrous capsule) with differences in referral patterns based on structure, location and stimulus intensity, the presentation of SIJ complex pain tends to be more variable than for individual facet joints, discs or muscles.

Historical features

There are some historical features that may help identify response to IA SIJ local anesthetic injection, suggesting IA pain, but studies have not typically distinguished between IA pain and dorsal ligament pain (see Table 4). The location of pain may have predictive value. Fortin et al. found that a point of maximal pain within 1 cm of a location just infero-medial to the posterior superior ilac spine (PSIS) correlated with response to IA SIJ injection in 16/16 subjects.$^{57}$ Another study found that tenderness within 2 cm of the PSIS was associated with positive response to EA (periarticular) injection$^{58}$; however, no IA SIJ injection was performed to assess the comparative response.

Pain patterns can be quite variable. Comparing SIJ IA responders vs non-responders, van der Würff et al. found no major differences except 10% of the responders vs 100% of the non-responders had isichial tuberosity pain.$^{59}$ Some studies have reported pain in the buttock and thigh without

symptoms distal to the knee,$^{60}$ while others have reported a comparable proportion of patients who experienced these pain patterns between responders and non-responders to IA SIJ injection.$^{43}$ The presence of groin pain has been shown to predict a positive response to an IA SIJ injection, though groin pain was only present in 14% of responders in one study.$^{43}$ Alternatively, in two subsequent studies, pain in the groin was found to be associated with a positive response to a diagnostic EA injection.$^{61,62}$ In another study, the positive correlation between response to an IA injection and pain in the posterolateral thigh and groin fell shy of statistical significance.$^{63}$

Characteristic painful activities and mechanisms of injury have been suggested as variables that may help distinguish different pain generators within the SIJ complex. In one retrospective cohort study, the authors identified several historical features that correlated with response to a therapeutic IA SIJ injection with local anesthetic and corticosteroid. These features were vehicle accidents, falls onto the buttock(s), post-partum state, American football injuries, and pelvis fracture.$^{64}$ Other inciting events included cumulative sports injuries such as lifting weights and running. Additional factors that may be associated with SIJ IA or ligamentous pain include leg length discrepancy, gait abnormalities, prolonged vigorous exercise, scolios, and spinal fusion.$^{65}$ Inflammation of the SIJ including a history of seronegative spondyloarthropathies with buttock pain suggests the presence of IA mediated pain. In one study, pain with rising from a chair was associated with IA SIJ pain but not lumbar facet joint pain.$^{66}$ However, none of these features have been shown to reliably distinguish between IA and EA sources of SIJ complex pain.$^{65}$

Features that are negative predictors for SIJ complex pain have also been described. In one study, only 4.4% of individuals with a positive diagnostic response to SIJ IA indicated that they had pain above L5 on a pain diagram that they completed at the time of baseline evaluation.$^{67}$ Several studies have also found that lumbar midline greater than paraspinal pain, or centralization of symptoms with provocation, is negatively associated with SIJ pain as diagnosed by IA SIJ injection as the reference standard.$^{57,66,68}$

Inflammatory arthritis represents a distinct variant of IA SIJ disease, being especially common in ankylosing spondyloarthopathy and to a lesser degree rheumatoid arthritis. Distinctive features that may distinguish inflammatory arthritis from osteoarthritis include prolonged morning stiffness (>30 min), pain improved with activity and not improved with rest, nocturnal pain, the presence of uveitis and/or enthesitis, and onset before 40 years of age.$^{69}$

Physical examination

There is currently no evidence that the physical exam can distinguish IA pain from EA ligament pain (see Table 5). Provocative tests such as Patrick's or flexion-abduction-external rotation (FABER) and Gaenslen's generally focus on stressing the anterior joint structures, while motion tests such as Gillet (aka Stork test) purport to assess mobility and sometimes alignment; mobility tests may be particularly useful for the identification of early inflammatory arthritis. In contrast, exam maneuvers that rely on tenderness may be more likely to detect dysfunction of the more superficial and palpable posterior ligament complex. A biomechanical study using a 3-dimensional finite element model found the SIJ provocation

|Author, year, type of injection|Study design and diagnostic criteria|No. of patients|Test|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Schwarzer $1995^{43}$ IA|Cross sectional study in patients with LBP below the L5-S1 spinal level and negative facet injections who responded to IA SIJ injections|43|Comparison of historical features|||No historical or PE test offered predictive value of response to injection, but the presence of groin pain was more common in patients with positive diagnostic injection.| |Dreyfuss, $1996^{67}$ IA|Prospective cohort study correlating 12 physical exam maneuvers with response to IA SIJ injection|85|Correlated PE & historical features with response to IA SIJ injection. These included pain drawings, maximal pain within 2 cm of posterior superior iliac spine, lifting butock off chair while sitting. Maximal pain within 1cm infero-medial to the PSIS|100%||Pain extending below knee showed comparable prevalence in those with and without SIJ complex pain.| |Fortin, $1997^{57}$ IA|Prospective cohort of patients with pain during arthrography|16|Patrick's, Ganslen's, shear and Yeoman's test, and pain with pressure over the sacral sulcus.|100%||No historical feature, none of the 12 SIJ tests, and no ensemble of these 12 variables demonstrated meaningful diagnostic value. Not able to distinguish between intra-and extra articular| |Slipman, $2000^{79}$ IA|Retrospective study correlating response to an IA SIJ injection with pain referral patterns|50|Patrick's, Ganslen's, shear and Yeoman's test, and pain with pressure over the sacral sulcus.|100%||All patients had positive provocative joint injections validating SIJ abnormalities. 94 described buttock pain, 36 patients described LBP, 14% had groin pain, and 50% had lower extremity pain. Pain extended into the lower leg in 28% and into the foot in 14%.| |Fukui, $2002^{60}$ IA|Prospective cohort study in patients who responded to an IA SIJ injection and had pain provocation with exam tests.|28, 32 joints|Patrick's and Gaenslen's tests|||All patients indicated pain over the SIJ, 68.7% had pain in the medial buttock, 37.5% experienced pain in the lateral thigh, 31.2% indicated the posterior thigh region as a pain site, and 9.3% reported groin pain.|

Table 4. (continued)

|Author, year, type of injection|Study design and diagnostic criteria|No. of patients|Test|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Chou, $2004^{64}$ IA|Retrospective study in patients with a negative EA injection with saline and a positive IA injection|194|Historical features, iliac compression test, Patrick's test, Yeoman's test, Gaenslen's test, midline sacral thrust and pressure to sacral sulcus Identified historical features of IA and EA sources of pain|||44% had a traumatic event, 21% had cumulative injury, and 35% had spontaneous or idiopathic onset.| |Cohen, $2005^{65}$|Evidence-based review|419, 160 SIJ patients|Correlated pain drawings to injection and RFA outcomes|||Factors associated with SIJ pain included leg length discrepancy, gait abnormalities, prolonged vigorous exercise, scoliosis, and spinal fusion. Inflammation of the SIJ, especially in those with bilateral symptoms, suggests IA pathology. Pregnancy predisposes women to EA pain. Pain localized to but-tock region was the only pattern associated with positive injections, with a diagnostic accuracy rate of 68.4%. Injections guided by pain provocation during 10% hypertonic saline injection.| |Jung, $2007^{63}$ IA/ligamentous|Prospective cohort in patients with lumbar facet or SIJ complex pain confirmed by IA or ligamentous injection|419, 160 SIJ patients|Correlated pain drawings to injection and RFA outcomes|||44% had a traumatic event, 21% had cumulative injury, and 35% had spontaneous or idiopathic onset.| |Murakami, $2008^{58}$ Ligamentous|Prospective study in patients with SIJ pain confirmed with placebo-controlled EA injections|46|Pain within 2cm of PSIS|78%|70%|Factors associated with SIJ pain included leg length discrepancy, gait abnormalities, prolonged vigorous exercise, scoliosis, and spinal fusion. Inflammation of the SIJ, especially in those with bilateral symptoms, suggests IA pathology. Pregnancy predisposes women to EA pain. Pain localized to but-tock region was the only pattern associated with positive injections, with a diagnostic accuracy rate of 68.4%. Injections guided by pain provocation during 10% hypertonic saline injection.| |Kurosawa, $2015^{61}$ Ligamentous|Prospective study in patients with SIJ complex pain confirmed by positive response to EA injections and negative response to control injections|50|Worst pain divided into 4 regions: upper but-tock, around the PSIS, middle buttock, and lower buttock.|Pain situated around the PSIS had the highest sensitivity (96%) with middle buttock pain having the lowest (69%).|Pain around the PSIS had the lowest specificity (82%) while lower buttock pain had the highest (99%).|22 (44%) patients complained of groin pain which was provoked in only 1 patient but relieved by local anesthetic in all patients| |Kurosawa, $2017^{62}$ Ligamentous|Prospective study to determine distinguishing factors between SIJ ligament pain, lumbar disc herniation and lumbar spinal stenosis.|62 with SIJ complex pain, 59 with disc herniation or spinal stenosis|Two pain areas (PSIS detected by the one-finger test, ie, pointing to the most painful area, and groin pain), sitting on a chair, shear provocation test, and 4 tender points|90% sensitivity for a positivity cutoff point of 4 (out of a risk score of 0-9) based on regression analysis.|86% specificity|When the cutoff point was increased to five, the sensitivity decreased to 85% and specificity increased to 93%|

EA, extra-articular; IA, intra-articular; PSIS, posterior superior iliac spine; SIJ, sacroiliac joint.

10

|First author, year|Study design|No. of patients|Tests|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Dreyfuss, $1996^{67}$|Prospective cohort study correlating 12 physical exam maneuvers with response to IA SIJ injection|85|Various pain drawings, maximal pain within 2 cm of posterior superior iliac spine, lifting buttock off chair while sitting, Gillet test, thigh thrust test, FABER/Patrick test, sacral thrust test, Gaenslen's test, joint play|||No historical feature, none of the 12 SIJ tests, and no ensemble of tests provided meaningful diagnostic value.| |Maigne, $1996^{80}$|Prospective study assessing prevalence of IA SIJ pain using dual injections|67|Distraction, compression, sacral pressure, Patrick's and Gaenslen's tests, pain with resisted external hip rotation and pain with pressure applied to the public symphysis|||There was no association between any clinical variable or pain provocation test and a (+) response to injections.| |Broadhurst, $1998^{81}$|Placebo-controlled double-blind study|40||77% for Patrick's, 80% for posterior shear and 87% for resisted abduction tests|100% for Patrick's, posterior shear and resisted abduction tests|No patient had ≥70% pain relief after saline.| |Slipman, $1998^{82}$ IA|Prospective cohort study examining tests associated with response to an IA SIJ injection|50|Patrick's, Ganslen's, shear and Yeoman's tests, and pain with pressure over the sacral sulcus.|||SIJ injections were positive in 60% of patients with 3 positive tests.| |Vleeming, $2002^{75}$ IA/Ligamentous|Cross sectional analysis in women with lumbopelvic pain that began during or right after pregnancy|178|Tenderness with palpation over the bilateral long dorsal sacroiliac ligament. Posterior pelvic pain provocation test and active straight leg raising test used as reference criteria.|76% for tenderness to palpation of long dorsal sacroiliac ligament for entire cohort, 86% in those diagnosed based on positive posterior pelvic pain provocation and active straight leg raising test, and 98% in those with severe pain on both tests.||Did not use injections as reference criteria, limited validity| |Laslett, $2003^{68}$ IA|Prospective, observational study evaluating physical exam tests associated with response to dual diagnostic IA SIJ injections|48 patients with lumbopelvic pain|Thigh thrust, Gaenslen's, distraction, compression and sacral thrust tests, as well as McKenzie assessments|78% for 3 positive tests|78% for 3 positive tests|9 subjects had peripheralization or centralization during the course of movements with none having positive response to injections.|

|First author, year|Study design|No. of patients|Tests|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Young, $2003^{66}$ IA|Prospective study evaluating physical exam tests associated with outcome in patients who experience pain provocation with cap-sular distension and relief with IA SIJ injection|102 with lumbopel-vic pain|Distraction, compression, sacral thrust, thigh thrust (posterior shear) and Gaenslen's tests|||3 positive tests, unilateral pain, and pain worsened when rising from sitting, strongly associated with response to injection. Negative relationship between midline lumbar pain and positive injection. Positive and negative predictive values were 77% and 87%, respectively, for 3 or more positive tests.| |Van der Wurff, $2006^{59}$ IA|Prospective study in patients who responded to a dual diagnostic SIJ injection|140|Distraction test, compression test, thigh trust test, Patrick's test, Gaenslen's test|3 or more positive tests had 85% sensitivity|79% for 3 or more positive tests|Positive and negative predictive values were 77% and 87%, respectively, for 3 or more positive tests.| |Szadek, $2009^{83}$|Systematic review of 18 studies||Various exam maneuvers, historical signs and pain patterns, and radiological studies|||No major differences between responders and non-responders except 10% of the responders vs 100% of the non-responders had ischial tuberosity pain. Using positive response to double-injections as the reference standard, the odds ratio of the thigh thrust test was 18.5, compression test was 3.9, and 3 or more positive stressor tests was 17.2.| |Stanford & Burnham, $2010^{84}$ IA|Prospective cohort study evaluating exam tests associated with response to IA SIJ injection|34|Patrick's, ipsilateral and contra-lateral Gaenslen's, thigh thrust, lateral compression and sacral thrust tests|3 or more positive tests had 82% sensitivity|3 or more positive tests had 57% specificity|3 or more positive tests had 47% positive predictive value and 87% negative predictive value.| |Liliang, $2011^{53}$ IA|Prospective cohort study evaluating physical exam tests associated with response to dual IA SIJ injections|52 with prior lumbar fusion|Pain elicited with pressure over the sacroiliac liga-ments, Patrick's, Gaenslen's, and shear (thigh thrust) tests, Yeoman maneuver, and standing-extension test|||Normalization of tests not predictive of injection response. 4 or more provocative tests predicted response to SIJ injections. Postoperative pain different from baseline pain and unilateral pain also associated with positive injections. No single physical exam maneuver is predictive of response to diagnostic injection, but when at least three physical exam findings are present, the sensitivity and specificity increases| |Kennedy, $2015^{8}$ IA|Systematic Review|45 publications, 39 on exam findings|||||

|First author, year|Study design|No. of patients|Tests|Sensitivity|Specificity|Other|| |---|---|---|---|---|---|---|---| |Nejati, $2020^{85}$|Cross-sectional study evaluating correlation between physical exam tests and response to dual diagnostic IA SIJ injections|48|Motion palpation tests: Gillet and forward flexion tests. Pain provocation tests: Patrick's, thigh thrust (posterior shear), Gaenslen's, Yeoman, compression, distraction and Newton's tests.|95% for 3 or more positive tests. Gillet and forward flexion tests had 100% sensitivity, but 0% specificity.|11% for 3 or more positive tests. Patrick's test had 67% specificity. Gillet and forward flexion tests had 0% specificity.|Positive and negative predictive values of 3 or more positive tests 82% and 33%, respectively.|| |Schneider, $2020^{74}$|Prospective cohort study evaluating whether physical exam maneuvers correlate with response to IA SIJ injection|35|Thigh thrust, Gaenslen's, Patrick's, distraction, compression and sacral thrust tests|Highest for Patrick's and sacral thrust tests (77%), 29% for Gaenslen's, 29% for thigh thrust, 35% for distraction tests|Highest for distraction (83%) and compression (78%) tests, 44% for Gaenslen's, 28% for Patrick's and 72% for thigh thrust tests|No association found between a single or combination of SIJ provocation maneuvers and response to an injection. Positive (67%) and negative (58%) predictive values highest for distraction test.|| |Mekhail, $2021^{72}$|Prospective cohort study evaluating response to an IA SIJ injection|200|Patrick's, modified Gaenslen's alone and together|94% for combination|17% for combination|Patrick's and modified Gaenslen's tests had 81% positive predictive value, and 44% negative predictive value. Positive and negative predictive values 34.7% and 92.4%, respectively. Concluded a cluster of pain provocation tests provides 35% certainty of a painful SIJ. Negative tests are more predictive, providing 92% certainty an SIJ is not the primary pain generator. Patrick's test positive predictive value 43%, negative predictive value 38%. Gaenslen's test positive predictive value 47%, negative predictive value 41%. Combined positive predictive value 41%, negative predictive value 38%. All included studies evaluated IA injections|| |Saueressig, $2021^{76}$|Systematic review and meta-analysis of 5 studies||Clusters of pain provocation tests|80% for a cluster of positive tests|63% for a cluster of positive tests|Positive and negative predictive values 34.7% and 92.4%, respectively. Concluded a cluster of pain provocation tests provides 35% certainty of a painful SIJ. Negative tests are more predictive, providing 92% certainty an SIJ is not the primary pain generator. Patrick's test positive predictive value 43%, negative predictive value 38%. Gaenslen's test positive predictive value 47%, negative predictive value 41%. Combined positive predictive value 47%, negative predictive value 41%. Combined positive predictive value 41%, negative predictive value 38%. All included studies evaluated IA injections|| |Cohen, $2022^{73}$|Prospective multicenter cohort examining factors associated with 1-month outcome after SIJ injection|66|Patrick's test and Gaenslen's test|Patrick's test 59%, Gaenslen's test 71%, combined 53%|Patrick's test 26%, Gaenslen's test combined 37%|Patrick's test positive predictive value 43%, negative predictive value 38%. Gaenslen's test positive predictive value 47%, negative predictive value 41%. Combined positive predictive value 41%, negative predictive value 38%.|Patrick's test positive predictive value 43%, negative predictive value 38%. Gaenslen's test positive predictive value 47%, negative predictive value 41%. Combined positive predictive value 41%, negative predictive value 38%. All included studies evaluated IA injections| |Han, $2023^{71}$|Systematic review|168 415 319 213|Sacral thrust test Thigh thrust (posterior shear) test|57.3%|48.8%|All included studies evaluated IA injections|| ||||FABER/ Patrick's test Gaenslen's test|54.1%|53.7%||| |||||76.4%|32.3%||| ||||FABER/ Patrick's test Gaenslen's test|47.9%|47.9%|||

IA, intra-articular; SIJ, sacroiliac joint; FABER, flexion-abduction-external rotation.

Pain Medicine, 2025, Vol. 00, No. 0

tests that provided the highest contact pressure across the joint (>10 MPa) were the compression, thigh thrust, Patrick's, and Gaenslen's tests, with Gaenslen's test resulting in the highest contact pressure (27.9 MPa).$^{70}$ The posterior and interosseous ligaments were stretched in the compression test; the anterior and interosseous ligaments were stretched in the distraction test; all ligaments were stretched in the sacral apex pressure test; the posterior ligaments were stretched in the thigh thrust test; the anterior ligaments were stretched in Patrick's and Gaenslen's tests.$^{70}$ Collectively, this study demonstrates that provocation tests likely stress both soft-tissue (EA) and IA structures, with significant overlap between maneuvers leading to a lack of specificity.

Some studies and systematic reviews suggest that a positive response to physical exam maneuvers may help identify IA pain as verified by a positive response to IA SIJ injections,$^{59,68,71,72}$ while other studies have shown no significant correlation.$^{67,73,74}$ Although it might seem self-evident that meaningful pain relief with an IA SIJ injection indicates IA pain, diagnostic injections are confounded by high false-positive and false-negative rates.$^{8}$ Further muddling interpretation of these studies is the fact that >50% of attempted IA injections have been shown to extravasate injectate posteriorly into the dorsal ligaments.$^{6,58}$ Another study found that IA injection was associated with a 27% incidence of aberrant flow to unintended neural elements,$^{42}$ further undermining specificity and validity.

Vleeming et al. attempted to assess the prevalence of dorsal ligament pain in postpartum women with long dorsal sacroiliac ligament tenderness, but failed to use the widely-accepted criterion of pain relief following a diagnostic injection as the reference standard for identifying a painful SIJ complex.$^{75}$ In a systematic review involving five studies and 422 patients, all of whom received a battery of physical exam maneuvers and diagnostic IA injections, Sauessig et al. concluded that a positive cluster of SIJ provocation tests is associated with only 35% certainty of correctly identifying the joint as a pain generator, but that a negative cluster of tests can ascertain a non-painful joint 92% of the time.$^{76}$ Other studies evaluating McKenzie-based positional preferences have found those with centralization (reduction of referred pain from the axial spine toward an extremity) are unlikely to have SIJ complex pain based on response to injections.$^{66,68}$

Inter-rater reliability

The reliability of physical exam maneuvers is critical to assessing validity and has demonstrated mixed results in clinical studies. In a systematic review published in 2000 examining 11 studies that assessed inter- and intra-rater reliability for sacroiliac joint examination findings, van der Wurff et al. demonstrated poor reliability for joint mobility tests and inconsistent results for pain provocation tests, with an inverse relationship observed between study methodological quality and correlation coefficients.$^{77}$ In one prospective follow-up study using two independent examiners who performed exam maneuvers in 78 patients with back pain and asymptomatic controls, the weighted kappa was found to be 0.70 (95% CI = 0.45-0.95) for Gaenslen's, Patrick's, thigh thrust, compression and distraction tests, indicating substantial agreement.$^{78}$ In this study, the presence of three of five positive provocation tests was the threshold for a positive design- nation (see Table 6).$^{78}$

Summary:

$\bullet$ For assessing pain originating from the SIJ complex, there is low-quality evidence that a battery of historical (eg, groin pain) and physical exam (ie, provocative but not motion tests) features may help predict response to IA SIJ injection of local anesthetic with or without corticost器ids; Grade B evidence, moderate level of certainty

$\bullet$ There are minimal available data regarding their ability to predict results in patients with primarily EA dorsal ligamentous pathology; Grade C evidence, low-to-moderate level of certainty.

$\bullet$ The specificity for individual and clusters of tests is lower than the sensitivity, with negative tests having greater predictive value than positive tests; Grade A evidence, high level of certainty.

$\bullet$ Inter-rater reliability for SIJ provocation tests demonstrates a wide range, with some of the most common tests (eg, FABER ("Patrick's") test) having the highest reliability; Grade C evidence, low level of certainty.

$\bullet$ Historical variables and physical exam signs can be used to select patients for diagnostic injections, but in individuals for whom there is a strong clinical suspicion (eg, negative workup for other causes) for SIJ complex pain, IA SIJ injections may be reasonable, even in the absence of positive provocation tests; Grade B evidence, moderate level of certainty

Q3. Can radiological imaging (including SPECT, radiographs, CT, MRI) identify a painful sacroiliac joint and aid in distinguishing articular sacroiliac joint pain from painful dorsal ligaments?

Imaging is the reference standard for identifying sacroiliitis [inflammation of the SIJ(s)], most commonly occurring in the context of systemic rheumatologic disorders, albeit with certain limitations. In contrast, there is a scarcity of studies designed to use imaging for identifying painful SIJs, detecting SIJ dysfunction, or differentiating between pain originating from the SIJ itself vs pain from the surrounding dorsal ligaments. Notably, few studies have compared the efficacy of imaging techniques to the reference standard of diagnostic anesthetic injection for these purposes.

Imaging the sacroiliac joint

SIJ dysfunction is typically associated with few abnormal imaging findings.$^{91}$ Radiographs are the least costly imaging modality but are less sensitive than CT or MRI for detecting SIJ structural abnormalities, especially in early disease states. Besides anteroposterior (AP) radiographs, there is no consensus regarding the optimal additional radiographic views to best image the SIJ given its oblique orientation and sigmoid structure.$^{91}$ In contrast to radiographs, cross-sectional imaging with CT, MRI, and single photon emission computed tomography (SPECT) create multiplanar images of the SIJ. CT can demonstrate sclerosis and bone erosions, while MRI can identify synovitis and bone marrow edema.$^{91}$ MRI is the best modality for detection of early SIJ inflammation, as fluid-sensitive sequences can identify tissue changes such as subchondral bone marrow edema.$^{92}$ Meanwhile, SPECT can detect metabolic disturbances in and around the SIJ and its dorsal ligaments.$^{93}$

Pain Medicine, 2025, Vol. 00, No. 0

|Author, Year|Design|No. of subjects|Reliability|Comments| |---|---|---|---|---| |McCombe, $1989^{86}$|Prospective study assessing inter-rater reliability.|83|Kappa values for: Maitland test: 0.26-0.38 Compression test: 0.09-0.16 Distraction test: 0.11-0.36 Pain with hip flexion: 0.40-0.42 Pain on residented hip external rotation: 0.38-0.63 Kappa values highest for thigh thrust (0.88) and distraction, compression and pelvic torsion (0.76), and lowest for sacral thrust (0.59) and cranial shear (0.69)|Concluded pain with hip flexion is reliable, pain elicited by external rotation is potentially reliable, and the others are unreliable.| |Laslett, $1994^{87}$|Prospective study assessing inter-rater reliability|51|Kappa values highest for thigh thrust (0.88) and distraction, compression and pelvic torsion (0.76), and lowest for sacral thrust (0.59) and cranial shear (0.69)|All exams performed by the same 2 physical therapists.| |Van der Wurff, $2000^{77}$|Systematic review|||Poor reliability for joint mobility tests. Mixed results for pain provocation tests, with an inverse correlation found between study quality and correlation. Strongest evidence for Gaenslen's and thigh thrust tests. 3 of 5 positive provocation tests was the threshold for a positive designation.| |Kokmeyer, $2002^{78}$|Prospective study|78|The weighted kappa was found to be 0.70 for Gaenslen's, Patrick's, thigh thrust, compression and distraction tests, indicating substantial agreement|| |Potter & Rothstein, $1985^{88}$|Prospective study|17|Compression and distraction tests had a grandion, respectively, with the other 11 tests having <70%.|Since the tests with high agreement were based on subjective metrics (pain report), the authors concluded all tests contained significant limitations. Historical findings (buttock, groin and SI joint pain) had higher kappa values (>0.67) than any exam finding.| |Dreyfuss, $1996^{67}$|Prospective study|85|Exam findings with kappa values >0.6 were thigh thrust, Patrick's and Gaenslen's tests|Historical findings (buttock, groin and SI joint pain) had higher kappa values (>0.67) than any exam finding.| |Strender, $1997^{89}$|Prospective study|71|SIJ provocation tests had good reliability: 96% agreement for Patrick's, 79% (kappa 0.26) for compression and 97% for internal hip rotation|50 pts examined by 2 physiotherapists and 21 by 2 physicians.| |Bertilson, $2006^{90}$|Prospective study|50|The 3 physical examination tests (SIJ compression, paravertebral tenderness, and iliopsoas tightness) had the lowest kappa values for all tests (0.6 to 0.33)|All exams done by 2 physical therapists using Kirkaldy-Willis back pain classification.|

SIJ, sacroiliac joint.

Pain Medicine, 2025, Vol. 00, No. 0

Imaging sacroiliitis

Inflammatory sacroiliitis, which can be observed with ankylosing spondylitis, psoriatic arthritis, spondylitis associated with inflammatory bowel disease, and reactive arthritis,$^{92}$ may be imaged with x-ray (XR), CT, and MRI commonly used in clinical practice. Radiographic features of sacroiliitis take longer to become detectable than other means of diagnosing the underlying condition, wherein approximately 33% of patients with symptoms <1 year and 50% of patients with symptoms for 2-3 years demonstrate perceptible radiographic findings.$^{94}$ Interpreting radiographs of the SIJ in suspected sacroiliitis can be challenging. Inter-observer and intra-observer variations are substantial, and sacroiliitis can be missed or incorrectly diagnosed using radiographs.$^{95}$

MRI is considered the reference standard for detecting SIJ involvement in patients with spondyloarthropathies. Studies have demonstrated that MRI has over 80% sensitivity in detecting chronic sacroiliitis, including sclerosis, joint space abnormalities, and ankylosis, with the highest sensitivity achieved through contrast-enhanced and short tau inverted recovery (STIR) techniques.$^{96,97}$ Although MRI may be more sensitive in detecting active inflammation and associated structural changes, CT remains the preferred method for diseases involving bone destruction or ossification. When MRI is inappropriate or unavailable in the real world, visual scoring of SIJ to sacrum (SIS) ratios of SPECT/CT may serve to identify axial spondylarthritis (axSpA) with high sensitivity and a high negative predictive value. In one study, the sensitivity, specificity, and positive and negative predictive values of SIS ratio of SPECT/CT for axSpA were 87.5%, 56.5%, 16.6%, and 97.8%, respectively.$^{98}$

Identification of a painful sacroiliac joint with imaging

Imaging studies have not been found to be accurate in differentiating a painful from non-painful SIJ (see Table 7). In one prospective study performed in 364 patients evaluated in a neurosurgical clinic for mechanical LBP, the use of imaging added very little to the accuracy of an injection-confirmed SIJ complex pain diagnosis, with history alone identifying 85-86% of 150 cases, imaging added to history resulting in an 87% accuracy rate, and the combination of historical findings, physical exam, and advanced imaging increasing the accuracy rate to 96%.$^{99}$

Raduclide bone scans

Slipman et al.$^{100}$ and Maigne et al.$^{101}$ found that radionuclide bone scanning had sensitivities of 13% and 46%, respectively. In the study by Slipman et al., at least three SIJ provocation maneuvers were positive in included patients. The reference standard in this study was a single IA SIJ injection with either 2 mL of lidocaine mixed with 1 mL (6 mg) of betamethasone or 3 mL of lidocaine performed under fluoroscopic guidance. Iohexol 0.5 mL was injected to confirm IA needle placement, but rates of IA spread were not reported. A threshold of 80% pain relief was used to define a positive response. In the study by Maigne et al., patients were selected based on tenderness over the SIJ without use of SIJ provocation maneuvers. Needle placement was performed under fluoroscopic guidance with injection of iopamidol 1 mL followed by ~2 mL of lidocaine, with strictly IA spread reported in 32/39 patients. A threshold of 75% pain relief was used to define a positive injection. Although these studies showed specificities >90%, their low sensitivities indicate

that bone scan is not an effective screening tool for SIJ complex pain, as it fails to detect most cases.

CT scans

Elgafy et al.$^{102}$ conducted a comparison of CT findings between 62 patients experiencing pain relief after SIJ injections and 50 age-matched controls without pelvic or back pain. Patient selection relied on limited criteria: Pain with pressure 1 inch below the posterior superior iliac spine, absence of pain with lumbar movements, and negative lumbar MRI. The injection protocol included a 5 mL subcutaneous lidocaine injection for local anesthesia followed by a large volume IA administration of 5 mL 1% lidocaine with 1 mL betamethasone. Notably, the authors neither used contrast to confirm IA placement nor defined quantitative criteria for positive response beyond subjective “pain relief.” Results showed CT abnormalities in 57.5% of symptomatic patients vs 31% of controls, with 42.5% of symptomatic patients exhibiting no CT abnormalities. The calculated sensitivity (57.5%) and specificity (69%) indicate poor diagnostic utility, underscoring the methodological limitations of this investigation.

MRI

In a prospective observational study$^{103}$ performed in 60 patients with pain clinically suspected to originate from the SIJ who received MRIs, acute inflammatory arthropathy was found in 36.7%, nonspecific degenerative changes were found in 35%, stress fractures in 3.3%, and osteoid osteoma in 7%, with 23.3% having normal SIJs. The authors inadequately described the basis upon which patients were clinically suspected to have pain originating from the SIJs and did not report the use of physical exam maneuvers. Diagnostic injections were MRI-guided with 1 mL of a mixture of bupivacaine 1% and gado-pentetate dimeglumine, with spread assessed using a fat-saturated T1-weighted spin echo sequence. SIJs were successfully accessed in 94.7% (108/114), of which 95.4% (103/108) of cases of spread were restricted to the IA joint on imaging. Thirty-two patients with nonspecific degenerative changes or normal SIJs underwent diagnostic IA SIJ injections; 46.9% had positive diagnostic injections and 53.1% had negative diagnostic injections. Among the 46.9% with positive diagnostic injections, 66.7% had normal MRIs and 33.3% had nonspecific degenerative changes. This study demonstrates that SIJ pain, as diagnosed by the reference standard of anesthetic injection, may not consistently correlate with MRI findings. However, due to the study design, it was not possible to determine the sensitivity and specificity of MRI in identifying painful SIJs.

SPECT

Cusi et al.$^{93}$ conducted a study examining the utility of single-photon emission computed tomography/computed tomography (SPECT/CT) in 100 patients (72 females, 28 males) with pelvic girdle pain—52% post-traumatic and 48% postpartum—compared to 80 asymptomatic controls undergoing SPECT/CT for non-pain indications. Imaging findings were compared to those in 80 control patients who received SPECT/CT for non-pain related reasons. Diagnostic injections were not used in this study; rather, the reference standard for the diagnosis of SIJ complex pain was clinical exam (3 of 4 positive exam maneuvers). Nuclear medicine physicians interpreted imaging studies in a blinded fashion. The results demonstrated that soft-tissue uptake posterior to the joint [area under the curve (AUC) = 0.99] or

Downloaded from https://academic.oup.com/painmedicine/advance-article/dol/10.1093/pn/praaf/29:63:645 by guest on 01 December 2025

Pain Medicine, 2025, Vol. 00, No. 0

|First author, year, imaging modality|Study design and diagnostic criteria|No. of patients|Tests|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Elgafy, $2001^{102}$ CT|Retrospective case-control study evaluating the diagnostic value of CT in patients with SIJ complex pain. Diagnostic criteria were pain relief with CT-guided IA SIJ injection, exam consistent with SIJ complex pain, and negative L-spine MRI.|Cases: 62 Controls: 50|Compared CTs of 62 patients with pain relief after IA SIJ injections with local anesthetic (symptomatic group) to 50 age-matched patients who had CTs for reasons other than pelvic or back pain (control group)|57.5%|69%|| |Slipman, $1996^{100}$ Radionuclide bone scan|Prospective case-control study. 3 positive provocative maneuvers on exam, diagnostic IA SIJ injections were the reference standard.|N = 50 Cases: 31 Controls: 19|50 consecutive patients underwent nuclear scanning and IA SIJ injections|12.9%|100%|| |Maigne, $1998^{101}$ Quantitative radionuclide bone scanning|Prospective case-control study. Patients with LBP underwent imaging followed by IA SIJ injection.|Patients: 32 underwent injection [Positive (75% benefit) injection: 13 Negative injection: 19] Controls: 34|Unilateral LBP with tenderness on palpation of the SIJ line without x-ray or CT findings suggestive of lumbar etiologies.|46.1%|89.5%|| |Fritz, $2008^{103}$|Controls used to establish upper limits of normal for quantitative radionuclide bone scanning.|N = 60 Acute inflammatory orthropathy: 22/60 (36.7%)|All patients with suspected SIJ complex pain received MRI, and those with inflammatory changes received therapeutic IA corticosteroid SIJ injections while those without received diagnostic IA local anesthetic SIJ injections and if positive received therapeutic IA SIJ injections.|Unable to calculate with provided information|Unable to calculate with provided information|Among 32 without inflammatory changes on MRI who received technically successful diagnostic IA SIJ injections, 15 (46.9%) were positive. Ten patients (66.7%) had normal imaging and 5 (33.3%) had non-specific degenerative changes]. 17 (53.1%) had negative diagnostic IA SIJ injections.| ||LBP clinically suspected to originate from SIJ, >12 weeks duration despite nonsteroidal anti-inflammatory drugs and/or physical therapy.|Nonspecific degenerative changes; 21/60 (35%)|Stress fracture (2/60, 3.3%)|||| |||Stress fracture (2/60, 3.3%)|ostoid osteoma (1/60, 1.7%)|||| |||Normal in 14 of 60 (23.3%) patients.|||||

Table 7. (continued)

|First author, year, imaging modality|Study design and diagnostic criteria|No. of patients|Tests|Sensitivity|Specificity|Other| |---|---|---|---|---|---|---| |Cusi, $2013^{93}$ SPECT/CT|Prospective observational study Pelvic girdle pain with ≥3 positive SIJ provocation maneuvers|Cases: 100 with pelvic girdle pain with ≥3 positive SIJ provocation maneuvers|All patients received SPECT/CT|Soft-tissue uptake: 0.93 Ligament thesis uptake: 0.93 SIJ uptake: 0.93 SIJ sclerosis on CT: 0.75|Soft-tissue uptake: 1.0 Ligament thesis uptake: 0.99 SIJ uptake: 0.60 SIJ sclerosis on CT: 0.60|The best diagnostic criterion for SIJ incompetence was either soft-tissue uptake posterior to the joint (AUC = 0.99) or uptake at the sacroiliac ligament insertions (AUC = 0.99). The combination of these two criteria yielded an AUC of 1.0. No diagnostic injections were performed. Sensitivity and specificity were determined based on clinical exam (3 of 4 positive exam maneuvers) as reference standard. Patients with lateralized pain had higher standardized uptake value (SUV) on symptom-dominant side (mean 5.3, range 3.6-9.9) vs nondominant (mean 4.1, range 2.2-6.9) side. In the 10 patients with bilateral pain without lateralization, the mean SUV difference between sides was small ( 0.22 [range 0.0-0.6]). In the 3 patients with unilateral symptoms, the affected side had tracer accumulation, with mean SUV 3.6 (range 3.0-4.8) Increased uptake in the upper SIJ (S1-S2) with extension into the dorsal interosseous ligament in 100% of patients. Mean counts were 168 (SD: 91) on the affected side and 109 (SD 60) on the unaffected side.| |Tofuku, $2015^{104}$|Prospective observational "Severe SIJ dysfunction" unresponsive to 1 year of conservative treatment and pain >4/10 Positive diagnostic IA SIJ injection defined as 75% pain relief|Patients: 32 Bilateral: 29 (19 lateralized) Unilateral: 3 (all lateralized)|All patients received SPECT/CT|Unable to calculate with provided information|Unable to calculate with provided information|Patients with lateralized pain had higher standardized uptake value (SUV) on symptom-dominant side (mean 5.3, range 3.6-9.9) vs nondominant (mean 4.1, range 2.2-6.9) side. In the 10 patients with bilateral pain without lateralization, the mean SUV difference between sides was small ( 0.22 [range 0.0-0.6]). In the 3 patients with unilateral symptoms, the affected side had tracer accumulation, with mean SUV 3.6 (range 3.0-4.8) Increased uptake in the upper SIJ (S1-S2) with extension into the dorsal interosseous ligament in 100% of patients. Mean counts were 168 (SD: 91) on the affected side and 109 (SD 60) on the unaffected side.| |Saunders, $2018^{105}$|Either referred from private clinical practice or prospectively recruited into an open-label trial of prolotherapy for physiotherapy-resistant SIJ complex pain. All had ≥3 positive exam maneuvers|Patients: 1200|All patients received SPECT/CT|Unable to calculate with provided information|Unable to calculate with provided information|Pain Medicine, 2025, Vol. 00, No 0|

AUC, area under curve; CT, computed tomography; IA, intra-articular; LBP, lower bac pain; MRI, magnetic resonance imaging; SIJ, sacroiliac joint; SUV, standardized uptake value; SPECT, single photon emission computed tomograph.

18

Pain Medicine, 2025, Vol. 00, No. 0

uptake at sacroiliac ligament insertions (AUC = 0.99)—with the combination yielding an AUC of 1.0—showed significant association with clinically diagnosed SIJ complex pain. However, the absence of confirmatory diagnostic intra-articular SIJ injections or sacroiliac lateral branch blocks necessitates cautious interpretation of these findings. This study indicates that SPECT/CT may hold promise as an imaging modality capable of identifying a painful SIJ complex, but future study will require use of diagnostic anesthetic injections as the reference standard.

Differentiating a painful sacroiliac joint from painful dorsal ligaments by imaging

We found no studies evaluating the ability of radionuclide bone scans, CT, MRI, or SPECT to differentiate a painful SIJ from painful dorsal ligaments.

Summary:

$\bullet$ The sensitivity of radionuclide bone scan (13%-46%) is inadequate for identifying a painful SIJ, though specificity is high (>90%); Grade D evidence for use as a diagnostic tool, moderate level of certainty. $\bullet$ CT has inadequate sensitivity (57.5%) and specificity (69%) for identifying a painful SIJ; Grade D evidence for use as a diagnostic tool, low level of certainty. $\bullet$ The evidence is insufficient to determine the sensitivity and specificity of MRI for identifying a painful SIJ; the available evidence suggests that MRI has low sensitivity and specificity for identifying a painful SIJ; Grade I evidence.

$\bullet$ SPECT/CT appears to have high sensitivity (95%) and specificity (99%) for identifying a painful SIJ complex based on one study, but these estimates are limited by using a reference standard of physical examination (rather than anesthetic injection) to diagnose SIJ complex pain; Grade I evidence for use as a diagnostic tool.

$\bullet$ SPECT/CT can identify EA and/or IA uptake but there is insufficient evidence to suggest that it can be used to differentiate IA SIJ pain from dorsal ligament pain; Grade I evidence.

04. Are intra-articular sacroiliac joint injections and sacral lateral branch injections diagnostic or prognostic?

The SIJ complex is a dual-structured joint where the sacrum meets the ilium. It features an amphiarthrodial superior/dorsal region with fibrocartilage and a smaller, more mobile diarthrodial antero-inferior portion with hyaline cartilage.$^{1-4}$ The joint is enclosed by a fibrous capsule with synovial lining and supported by extensive anterior and posterior ligaments, all containing nociceptors that can generate pain. The neuroanatomy of the SIJ complex remains incompletely characterized, with variability observed between independent cadaveric dissection studies. This anatomical complexity has led to debate regarding the diagnostic and prognostic validity of IA SIJ injections and sacroiliac lateral branch blocks (SLBBs) in clinical practice.$^{3,25,106-118}$

Are sacroiliac joint injections diagnostic?

There is no reference standard beyond diagnostic IA SIJ injection for confirming SIJ complex pain, creating an inherent challenge in assessing its diagnostic value. Current literature reports one or more IA SIJ injections as the reference

standard to evaluate other diagnostic methods (history, physical examination, imaging). Multiple factors can affect the sensitivity and specificity of these injections, including placebo effects, referred pain, neuroplasticity, central sensitization, expectation bias, systemic absorption of local anesthetic, and psychological factors.$^{119}$ Furthermore, the SIJ may communicate with the S1 foramen, there is potential for dorsal leakage of injectate from the joint, and anterior capsular defects are present in approximately 40% of individuals with suspected SIJ complex pain.$^{43}$ As such, a technically sound IA SIJ injection may result in off-target anesthetic spread and the possibility of a false-positive test result.$^{5,42,43}$ Despite these limitations, multiple systematic reviews and meta-analyses provide evidence that IA SIJ injections are diagnostic for SIJ-mediated pain with the false-positive rate of uncontrolled injections averaging around 20%.$^{8,40,83,120}$ The Centers for Medicare and Medicaid Services (CMS) considers IA SIJ injections to have diagnostic value while acknowledging that the EA ligaments may be a pain generator.$^{120}$ A systematic review of 18 manuscripts by Simopoulos et al. found evidence that controlled comparative local anesthetic IA SIJ injections are the best means to diagnose SIJ-mediated pain.$^{40}$ Another systematic review by Kennedy et al. also supports the diagnostic validity of IA SIJ injections for SIJ-mediated pain.$^{8}$ A systematic review by Szadek et al. partially supports the diagnostic validity of IA SIJ injections, but recommended that further studies should be conducted secondary to a paucity of sensitivity and specificity data.$^{83}$

Are sacroiliac joint injections prognostic?

Intra-articular SIJ injections may be prognostic for the outcome of SIJ immobilization by fusion (see Table 8). All clinical studies assessing the effect of SIJ fusion to date have used IA SIJ injections to select patients for surgery. However, it is unclear what prognostic value IA SIJ injections have for predicting response to SLBRFA. Although most clinical studies evaluating SLBRFA have used IA SIJ injections to select patients and reported favorable outcomes,$^{14,21,27,119}$ SLBRFA targets nerves that innervate the posterior ligamentous complex and posterior aspect of the joint, but not the full capsule, synovum or most of the IA portion of the SIJ.$^{111,117}$ The current literature suggests there are minimal differences in the prognostic value of an IA SIJ injection compared to SLBB in predicting SLBRFA treatment outcomes.$^{121}$ The only randomized study that reported a negative result from SLBRFA used a single prognostic IA SIJ injection with 3 mL of 2% lidocaine, but also required only a 2-point decrease in pain to diagnose SIJ pain.$^{24}$ Because of the spread of local anesthetic (but not extension of an RFA lesion) to adjacent potential pain-generating structures, this cutoff is likely too lenient and results in a high rate of false-positive responses (low specificity). An additional confounding variable is that this study also used technology and a SLBRFA technique this is not anatomically valid.$^{121,122}$ Anatomically valid SLBRFA techniques are described in Questions 13 and 17.

Prognostic value of local anesthetic injections on steroid injection outcomes

There are few studies examining the prognostic value of local anesthetic injections on outcomes from corticosteroid injections. In a small, prospective study performed in 26 patients, Schneider et al. reported a modest correlation between

Pain Medicine, 2025, Vol. 00, No. 0

Pain Medicine, 2025, Vol. 00, No. 0

|Author, year|Patients|Design|Results|Comments| |---|---|---|---|---| |Patel, $2012^{22}$|51 pts with SIJ complex pain based on IA or ligamentous injections and ≥75% pain relief to dual L5 dorsal ramus and S1-3 lateral branch blocks|Randomized, double-blind, sham-controlled study|3-months post-procedure, NRS pain scores were lower in the RFA group than the sham group.|Second sham-controlled trial (high-level evidence) that L5 dorsal ramus and S1-S3 sacral lateral branch RFA may provide pain relief and functional benefit for SIJ complex pain. First large, randomized, multicenter study to demonstrated effectiveness for RFA compared to standard medical management for subjects with chronic SIJ complex pain. Only 11% had negative diagnostic injection. A non-anatomically valid SIJ RFA technique was used.| |Cohen, $2023^{14}$|210 patients selected based on positive response to an undescribed SIJ injection and ≥50% pain relief from SLBBs.|Randomized, multicenter comparative-effectiveness study|3 months post-treatment, the mean pain score for the RFA group decreased significantly vs no significant decrease in the standard medical management group.|First large, randomized, multicenter study to demonstrated effectiveness for RFA compared to standard medical management for subjects with chronic SIJ complex pain. Only 11% had negative diagnostic injection. A non-anatomically valid SIJ RFA technique was used.| |van Tilburg, $2016^{24}$|60 patients enrolled based on ≥2-point pain reduction after diagnostic SIJ injections. Included patients with “sciatica”, and did not describe the diagnostic SIJ injections.|Randomized, double-blind, sham-controlled, multicenter trial|No statistically significant difference in pain level over time between groups.|Only 11% had negative diagnostic injection. A non-anatomically valid SIJ RFA technique was used.| |Juch, $2017^{16}$|226 patients enrolled based on ≥50% pain relief from L5 dorsal ramus and S1-3 lateral branch blocks.|Randomized comparative-effectiveness study comparing exercise program alone to exercise and RFA. Included patients who responded to low-volume L5 dorsal ramus and S1-3 sacral lateral branch blocks.|Significant difference favoring RFA group at 3-month primary endpoint but not afterwards.|Patients could receive psychotherapy as indicated. Only 24% of patients had negative SLBBs. A non-anatomically valid SLBRFA technique was used in a subset of subjects.|

IA, intra-articular; RFA, radiofrequency ablation; SIJ, sacroiliac joint; SLBB, sacral lateral branch block (injection); SLBRFA, sacral lateral branch radiofrequency ablation.

Pain Medicine, 2025, Vol. 00, No. 0

immediate pain relief from the local anesthetic in low-volume (2 mL) lidocaine-steroid injections, with the negative predictive value of immediate pain relief being higher for a negative response 2-4 weeks post-injection than the positive predictive value of a positive response, which is consistent with diagnostic blocks performed for other conditions.$^{123,124}$ In two placebo-controlled trials evaluating IA facet steroid injections in which patients were selected based on a positive response to injections performed only with local anesthetic, neither reported positive outcomes.$^{125,126}$ In addition to the questionable prognostic value of local anesthetic blocks, differences in technique (ie, needle location) and injectate spread between injections limit reproducibility.

Are sacral lateral branch blocks diagnostic?

Randomized, double-blinded, placebo-controlled studies reveal that SLBBs are not diagnostic for IA SIJ-mediated pain.$^{117}$ There is anatomical evidence that the IA portion of the SIJ is innervated from both ventral and dorsal sources.$^{107,13,127}$ A randomized double-blind, placebo-controlled study in 20 asymptomatic subjects by Dreyfuss et al. revealed that multi-site, multi-depth SLBBs are approximately 70% effective in blocking nociception arising from the interosseous (IO) and dorsal sacroiliac (DSI) ligaments, as well as the inferior dorsal capsule of the joint, but do not effectively block pain associated with distension of the full SIJ capsule.$^{117}$ Although this study did not assess the ability of SLBBs to block nociception from the articular surfaces of the SIJ (ie, no noxious stimulus to the articular surfaces was applied), it is very unlikely that the full IA portion of the SIJ is innervated by sacral lateral branches when taking into account these results and those of anatomical studies.$^{117,121,122}$

Are sacral lateral branch blocks prognostic?

SLBBs have face-validity for predicting the outcome of SLBRFA given that the same nerves blocked are those targeted by SLBRFA; however, the current literature suggests that they may not provide meaningful additional predictive value beyond using historical and physical exam criteria (ie, provocative tests) and a prognostic injection of any type (IA SIJ injection, SLBB or even a large-volume injection into the dorsal ligaments)$^{25,121}$ (see Table 8). Pragmatic trials directly comparing selection pathways and protocols are needed to increase confidence in our understanding of the prognostic value of IA, EA, and SLBBs. It will be important that such trials use anatomically-valid SLBRFA techniques$^{121,122}$ as discussed in Question 13 of these guidelines.

Summary:

$\bullet$ IA SIJ injections are diagnostically valid for IA SIJ-mediated pain; Grade B evidence, moderate level of certainty. $\bullet$ IA SIJ injections with local anesthetic have unreliable predictive value for IA injections with steroids; Grade I. $\bullet$ IA SIJ injections have unreliable prognostic value for SLBRFA—Grade C, low level of certainty.

$\bullet$ SLBBs are diagnostic for pain arising from the interosseous ligament, dorsal sacroiliac ligament, and inferior dorsal SIJ joint; Grade B evidence, low-to-moderate level of certainty.

$\bullet$ There is insufficient evidence to assess the relative prognostic validity of SLBBs compared to other selection criteria for SLBRFA; Grade I evidence.

Recommendations:

* Multiple systematic reviews partially support the diagnostic validity of SIJ injections, but further study is needed to address the variability in sensitivity and specificity data.

* Further pragmatic trials directly comparing selection criteria (history, physical examination, and SLBBs) are needed to increase confidence in the relative prognostic values of different means of optimally selecting SLBRFA candidates.

Q5. What volume should be used for intra-articular sacroiliac joint injections? Should diagnostic injections include corticosteroid?

History and physical examination can suggest the SIJ complex as the pain generator in patients with LBP, but with undetermined specificity and sensitivity. Pain relief following IA SIJ injections of local anesthetic (LA), with or without corticosteroids, is considered a more reliable diagnostic indicator of the SIJ being the pain-generator in patients with LBP than history and physical exam alone. Only 15-30% of patients with suspected SIJ complex-related pain based on history and examination experience analgesic benefit with IA injections of LA, which is likely attributable to a combination of incorrect diagnosis, lack of spread to the pain-generating area(s), and other technical failures.$^{8,80,128}$ Intra-articular SIJ injections are characterized by high false-positive rates, with needle position and the volume of injectate playing key roles in specificity.

Capacity of the sacroiliac joint

The SIJ has a surface area of around 17.5 cm$^{2}$, but the synovial portion has small capacity. The capacity of the SIJ has been measured by injecting contrast medium into the joint until either resistance was met or spillage into EA structures was observed. The reported volumes for retained IA injectates range from 0.8 to 2.5 mL in asymptomatic individuals and 1.0-2.7 mL in symptomatic individuals.$^{129,130}$ considering bony overgrowth (eg, osteophytes) in individuals with osteoarthritis, it is possible that healthy joints can accommodate a greater volume of injectate without capsular tears. It has been proposed that injectate volumes exceeding 2.5 mL are likely to be non-specific for diagnosing the articular portion of the SIJ as the pain generator in patients with mechanical LBP due to EA spillage.$^{131}$

The SIJ is covered dorsally by four layers of sacroiliac ligaments and muscles that can also be pain generators. Extra-articular causes of pain from the SIJ are likely as common as IA causes,$^{6,132}$ and the false-positive rate of single diagnostic IA SIJ injections is approximately 20%.$^{80}$ Therefore, target specificity is an essential criterion in assessing the validity of a diagnostic injection and, for IA LA injections into the SIJ, false-positive rates of 19%-22% have been reported in various studies synthesized in a systematic review.$^{8}$

Intra-articular sacroiliac joint injections

Presence of IA spread of a small volume of injected contrast (up to 0.5 mL) without spillage into anatomical structures

Pain Medicine, 2025, Vol. 00, No. 0

|First author, year, study type|Study population|Intervention|Implications for SI joint volume|Comments| |---|---|---|---|---| |Schwarzer, $1995^{43}$ Prospective|43 patients with LBP below the L5-S1 spinal level investigated with SIJ injections|1 mL of contrast and 1 mL of LA|2 mL injected|30% patients reported pain relief with 19 (44%) patients exhibiting tears in the ventral capsule| |Laslett, $2005^{140}$ Prospective|48 patients with suspected SIJ complex pain|1 mL of contrast and 1 mL of LA|2 mL injected|33% patients had pain relief with SIJ injections| |Fortin, $2003^{133}$ Prospective|74 SIJ injections in 43 patients (age 20-48 years)|63.5% subjects had a positive injection provocation test (reproduction of pain)|Mean injected volume: 1.08 mL (range: 0.5-1.8 mL)|Direct relationship between volume injected and pain reproduction, but not between reproduction of pain with injection and CT or x-ray findings of pathology.| |Waisbrod, $1987^{148}$ Retrospective|21 patients|Injections to prognosticate outcome of SIJ fusion|1-2 mL of saline to provoke pain and 1-2 mL of LA to relieve it|11 patients had 50% pain relief at a mean follow-up of 2.5 years after SIJ fusion| |Schütz, $2006^{149}$ Retrospective|14 patients|Injections to prognosticate outcome of bilateral SIJ fusion|5 mL of LA|Only 18% patients were satisfied with the outcome at a mean follow-up of 39 months| |Fortin, $1999^{42}$ Prospective|76 SIJ injections in 43 patients with LBP (age range 24-48 years)|Injections with contrast to evaluate the pattern of spread|Mean injected volume: 1.08 mL (range: 0.5-1.8 mL)|61% of injections resulted in EA spread - posterior spread into the dorsal sacral foramina, superior spread to the fifth lumbar nerve root, and ventral spread to the lumbosacral plexus| |Stelzer, $2019^{150}$ Prospective|40 SIJ injections in 20 cadavers|Injections with contrast to evaluate the pattern of spread|0.5 mL of contrast injected followed by 2 mL of colored latex|Significantly higher success rate of IA injection with fluoroscopy compared to US| |Luukkainen, $1999^{9}$ RCT|20 patients|Therapeutic EA injections of LA and corticosteroid vs LA only|3 mL of injectate|Better clinical outcomes in LA and corticosteroid group| |Luukkainen, $2002^{10}$ RCT|24 patients|Therapeutic EA injections of LA and corticosteroid vs LA only|3 mL of injectate|Better clinical outcomes in LA and corticosteroid group| |Cohen, $2019^{6}$ RCT|125 patients|Therapeutic fluoroscopically-guided IA SIJ and palpation-guided injections in proximity to the SIJ|3 mL of injectate|No difference at 1 month after the injections but better relief with fluoroscopically-guided injections at 3 months. In the fluoroscopy group, 44% of injections were only IA, 47% were IA and EA and 9% were EA.|

(continued)

24

|First author, year, study type|Study population|Intervention|Implications for SI joint volume|Comments| |---|---|---|---|---| |Maigne, $1996^{80}$ Prospective|19 patients|Dual diagnostic injections into the SIJ in patients with LBP—first injection with lidocaine and second injection with bupivacaine Diagnostic injection into SIJ in patients with pain over the SIJ following lumbar spinal fusion|1 mL of contrast injected followed by 2 mL of lidocaine 2% for first injec- tion and bupivacaine 0.5% for second injection 1 mL of contrast injected followed by 2 mL of lidocaine 2%|35% patients after $1^{st}$ injection and 18.5% after $2^{nd}$ injection had their VAS pain score reduced by ≥75%| |Maigne, $2005^{56}$ Prospective|45 patients|2 separate EA SIJ (deep interosseous ligamentous) injections with LA and corticosteroids to target pain arising from the SIJ and the posterior ligaments covering the joint|5 mL of 0.5% bupivacaine containing 4 mg/mL of triamcinolone|40 patients had IA injections -35% (95% CI: 20.2-49.8%) of these had >75% reduction in pain 15 min after injection| |Yin, $2003^{25}$ CS-R|14 patients|2 separate EA SIJ (deep interosseous ligamentous) injections with LA and corticosteroids to target pain arising from the SIJ and the posterior ligaments covering the joint|5 mL of 0.5% bupivacaine containing 4 mg/mL of triamcinolone|Injections used to select patients for RFA, of which 60% experienced ≥60% subjective relief and ≥50% pain score reduction maintained for at least 6 months.| |Murakami, $2007^{137}$ RCT|50 patients|Diagnostic injection with LA: IA vs EA|2 mL of 2% lidocaine|EA injections effective in all 25 patients but IA injections effective in only 9 of 25 patients. EA injections administered in one or more sections of the posterior EA area of the SIJ and to another section in the extracranial portion.| |Murakami, $2018^{136}$ Prospective|85 patients|Diagnostic injection with LA close to the posterior superior iliac spine: EA first; IA if EA was unsuccessful|2 mL of 2% lidocaine|58 out of 85 patients had pain relief with EA injection; 14 out of remain- der had pain relief with IA injection|

CI, confidence intervals; EA, extra-articular; IA, intra-articular; LA, local anesthetic; LBP, low back pain; PA, periarticular; RCT, randomized controlled trial; RFA, radiofrequency ablation; SIJ, sacroiliac joint; US, ultrasound; VAS, visual analog scale.

Pain Medicine, 2025, VOL. 00, No. 0

|First author, year, study type|Study population|Intervention|Implications for use of corticosteroids in injectates|Comments| |---|---|---|---|---| |Dreyfuss, $1996^{67}$ Prospective|85 patients with suspected SIJ complex pain who received IA injections to determine the validity of historical and physical exam signs for diagnosis|0.2-0.5 mL of contrast, 1.5 mL of LA, 0.5 mL of corticosteroid injected into SIJ|Increase in volume of the injectate by 25% due to use of corticosteroids|52 patients (61%) reported ≥50% pain relief 20 min after the injection| |Borowsky, $2008^{139}$ Retrospective|120 patients with suspected SIJ complex pain (n = 40 who received IA and 80 who received IA + EA injection)|IA-only injection consisted of 1.5 mL of 0.5% bupivacaine + 80 mg MP; Combination injection consisted of 2 mL 0.5% bupivacaine + 40 mg MP injected IA, and 2 mL bupivacaine and 40 mg MP injected into dorsal SIJ ligaments bathing lateral branches 4 mL of injectate (3 mL of LA (bupi- vacaine 0.5%) and 1 mL of MP) as a diagnostic and therapeutic intervention|Increase in volume of the injectate by 100% due to deliberate EA injection|At 3 months after injection, 12.5% patients in IA group and 31.25% patients in IA+EA group reported ≥50% reduction in pain score| |Soneji, $2016^{151}$ RCT|40 patients with suspected SIJ complex pain based on physical exam who underwent ultrasound-guided or fluoroscopically-guided IA SIJ injections|0.4 mL of injectate (3 mL of LA (bupi- vacaine 0.5%) and 1 mL of MP) as a diagnostic and therapeutic intervention|Increase in volume of the injectate by 33% due to use of corticosteroids|50% of ultrasound vs 65% of fluoroscopically-guided injections were IA. Similar outcomes in both groups.| |Liliang, $2011^{53}$ Prospective|52 patients (out of 130) who had lumbar/Lumbosacral spinal fusion and had ≥3 positive provocation tests for SIJ complex pain|Dual diagnostic IA SIJ injections with 0.5 mL contrast, 1 mL of LA (lido- caine 2% or bupivacaine 0.5%) with 1 mL of corticosteroid (40 mg triamcinolone)|Increase in volume of the injectate by 100% due to use of corticosteroids|40% patients had ≥75% reduction in pain with both diagnostic injections| |Chou, $2004^{64}$ Prospective|54 patients with (+) provocation tests who experienced ≥80% pain relief with diagnostic IA SIJ injection using only LA. 2 additional therapeutic injections with LA and corticosteroid were scheduled but cancelled if the first injection conferred ≥80% pain relief 2 weeks post-injection. Those who failed to obtain relief from therapeutic injections were scheduled for controlled dual injections (EA injections with saline and IA injections with LA).|0.5 mL contrast, 0.5 mL of LA (lido- caine 2%) with 2 mL of corticosteroid (12 mg betamethasone)|All 27 patients who did not report pain relief with IA SIJ injection with LA and corticosteroid also did not report pain relief with subsequent dual diagnostic injections with saline and LA.|A mean of 2.5 (range: 1-4) therapeutic SIJ injections were administered to study participants|

(continued)

26

|First author, year, study type|Study population|Intervention|Implications for use of corticosteroids in injectates|Comments| |---|---|---|---|---| |Maugars, $1992^{143}$ Prospective|42 SIJ injections in 22 patients with sacroiliitis|0.5 mL contrast followed by 1.5 mL of Cortivazol (equivalent to 62.5 mg of prednisolone) injected into SIJ without any LA|LA not injected because authors concerned about "transient anesthesia of the pelvis"|79% of patients (81% of joints) reported ≥70% pain reduction for at least 1 month| |Maugars, $1996^{11}$ RCT|13 SIJ injections in 10 patients with sacroiliitis|1-3 mL contrast followed by 1.5 mL of Cortivazol (equivalent to 62.5 mg of prednisolone) or saline injected into SIJ without any LA|Enrollment stopped at 1 year when only $1/3^{rd}$ of required number of participants enrolled|At 1 month, 5 of 6 SIJs in patients injected with corticosteroid reported >70% pain relief compared to 0 out of 7 in the placebo group| |Karabacakoglu, $2002^{142}$ Prospective|22 SIJ injections in 17 patients with ankylosing spondylitis and sacroiliitis|0.2-0.5 mL contrast followed by 1 mL (14 mg) of betamethasone injected into SIJ without LA|Mean IA injectate volume: 1.1 mL|At 1 month, 15 out of 17 patients reported good improvement and 2 of 17 reported fair improvement| |Visser, $2013^{134}$ RCT|51 patients with SIJ and leg pain randomized to physiotherapy (n = 15), manual therapy (n = 18), and IA injection (n = 18)|0.6-2 mL of a mixture of 30 mg of lidocaine and 20 mg triamcinolone injected|Mean IA injectate volume: 1.1 mL|20% of patients in physiotherapy group, 72% of patients in manual therapy group, and 50% of patients in IA injection group reported some reduction in pain 6-12 weeks after injections|

EA, extra-articular; IA, intra-articular; LA, local anesthetic; MP, depo-methylprednisolone; RCT, randomized controlled trial; SIJ, sacroiliac joint; VAS, visual analog scale.

Pain Medicine, 2025, Vol. 00, No. 0

adjacent to the SIJ or evidence of intravascular uptake confirms target specificity. In a systematic review by Kennedy et al., the synthesis of literature suggested that 0.2-0.5 mL of contrast was needed to confirm IA placement of the needle tip, followed by the injection of 0.5-2 mL of local anesthetics (depending on the dose and concentration), and 1-2 mL of corticosteroids (depending on the dose and concentration) into the SIJ for therapeutic effect.$^{8}$ The total volume of the injectate for IA injection in the SIJ has varied from 1.5 to 4 mL in clinical studies (Table 9). The mean volume of retained IA contrast injectate for the SIJ, prior to EA spread being observed, was reported to be 1.1 mL in two studies.$^{133,134}$ The median volume of injectate when only LA is used is approximately 2 mL but increases to 3 mL when a combination of LA and corticosteroids is injected into the SIJ (Tables 9 and 10).

Extra-articular sacroiliac joint injections

Pain arising from the SIJ can be due to stimulation of the nerve endings that exist not only in the articular cartilage but also in the SIJ capsule, synovial lining and the ligaments covering the posterior (and anterior) aspect of the joint.$^{3}$,$^{13}$S An EA injection of LA into the posterior ligaments can often provide effective relief of SIJ-related pain.$^{13}$,$^{13}$ This may be more relevant when soft tissue dysfunction of the SIJ is presumed to be the main cause of pain, such as in the non-elderly population with LBP.$^{13}$B 189 other studies have reported analgesic benefit when a combination of LA and corticosteroid was injected in the EA tissue in close proximity to the SIJ.$^{9}$,$^{10}$,$^{25}$,$^{139}$ However, these studies have several limitations: (1) none pre-screened patients with diagnostic injections; (2) follow-ups were short-term; and (3) functional capacity was not sufficiently evaluated.$^{1}$ Irrespective of the composition of the injectate, the volume for injection into the EA space around the SIJ is usually 2–3 mL.

Should diagnostic sacroiliac joint injections include corticosteroids?

Corticosteroids are usually injected into the SIJ to confer therapeutic benefit, especially if a preceding injection with LA results in substantial (usually ≥50%) but short-lasting pain relief.$^{68,140}$ Ideally, the injection of a combination of LA and corticosteroids would confer a biphasic response with the LA providing the initial analgesic benefit and the corticosteroid conferring delayed but sustained pain relief.

A sustained analgesic response after an SIJ injection with LA and corticosteroids may indicate the SIJ is indeed responsible for a patient's pain but could also reflect a discordant (prolonged) response to injection of the LA, which may be a placebo effect. Further, the addition of corticosteroids often increases the volume of the injectate (Table 10) and the utility of including corticosteroids in diagnostic injections into the SIJ is unclear. In a systematic review, when local anesthetic alone was injected, 35% (95% CI: 29-41%) of 246 patients experienced ≥75% relief, but when corticosteroid was added to the injectate, 49% (95% CI: 47%-51%) of 685 patients had ≥75% relief, suggesting corticosteroids may undermine specificity.$^{8}$ The timing of evaluation is important when determining whether a diagnostic injection is positive or negative, as systemic effects from injectable corticosteroids may be apparent within 6 h depending on intravascular uptake and the corticosteroid injected, which is often the timeframe for evaluating pain diaries.

Although a systematic review found that corticosteroids provide added relief to local anesthetic in only a small percentage of patients undergoing IA injections and nerve blocks,$^{141}$ in some cases, it may be appropriate to inject corticosteroids in the IA space in patients with suspected sacroiliitis because inflammation of the articular surfaces is a key driver of pain in these patients. Injection of corticosteroids in the IA space in these patients is often associated with pain relief lasting for one to six months.$^{11,142,143}$

Goals of the injections

SIJ injections serve multiple purposes beyond diagnosing or confirming SIJ complex pain in patients with LBP. These injections can also predict responses to more definitive interventions, such as surgical joint fusion or SLBRFA. There are no studies evaluating the predictive value of SIJ injections before fusion, but for lumbar fusion, the results are decidedly mixed, with discography rarely used nowadays to select patients.$^{144}$ Logically, when using IA SIJ injections as prognostic indicators for SIJ fusion outcomes, it would be crucial to ensure the injectate remains within the IA space. However, studies evaluating SIJ fusion typically do not report the volume of injectate used.$^{145,146}$

RFA of the nerves supplying the SIJ complex targets the innervation to the posterior sacroiliac ligaments and posterior aspect of the SIJ. Hence, inadvertent EA spread during an intended IA injection that results in analgesic benefit may be more predictive of the analgesic benefit following RFA of the posterior SIJ complex innervation, though it would be impossible to separate the contributions of IA and EA spread.

Another important issue to consider when treating patients with SIJ complex pain is whether the pain is purely IA, purely EA, or from a combination of IA and EA pathology. Even in patients with spondyloarthropathy, controlled trials evaluating EA corticosteroid injections have shown efficacy.$^{9}$ Suspecting the latter would justify an injection approach that involves intentional extravasation of LA into both the IA space as well as the posterior ligaments of the SIJ. In a retrospective study that compared outcomes in 120 patients who received only IA vs a combination of IA and EA injections, Borowsky et al. reported a higher incidence of analgesic benefit in the combination group. At 3 months after the injections, 12.5% patients in IA SIJ injection group and 31.3% patients in the combination group reported 50% or more reduction in their pain intensity scores.$^{139}$

Sacral lateral branch blocks

Prognostic injections of the lateral branches innervating the posterior SIJ have been recommended prior to SLBRFA, with the total cumulative volume for S1-3 ranging between 1.5-2 mL in controlled trials.$^{14,16,22}$ When S4 is targeted, slightly higher volumes may be necessary. For the L5 dorsal ramus, consistent with facet injections, most studies have used volumes ≤0.5 mL. In a pilot study performed in healthy volunteers that found that multi-depth (but not single-depth) SLBBs prevented pain from ligamentous probing but not capsular distension, Dreyfuss et al. performed injections at the L5 dorsal ramus and from S1-2 at the 2:30, 4:00, and 5:00 positions on the face of a clock for right-sided injections (mirror image locations on the left) 8-10 mm from the lateral margin of the posterior foramina, using 0.2 mL per site. At S3, nerves were targeted at the 2:30 and 4:00 positions, while S4 was only injected at the 3:00 position on the right in the single-site,

Pain Medicine, 2025, Vol. 00, No. 0

single-depth study (it was not targeted in the multi-site, multi-depth study). In the multi-depth study, the authors injected first at sacral periosteum, then repeated the injection after pulling back the needles 1 bevel length (approximately 3 mm). This resulted in a volume of 1.2 mL at each peri-foraminal region, injected at two different depths and at three different points at each sacral level.$^{117,118}$ SLBBs are performed purely as predictive tests before SLBRFA and there is no evidence supporting lateral branch nerve entrapment or inflammation as the etiology of SIJ complex pain. There is currently no basis to add corticosteroids to the local anesthetic. Given the variability in anatomy and techniques, more research is required to evaluate the impact of the volume of injectate in predicting outcomes of definitive therapies for relieving SIJ complex pain.

Communication between the sacroiliac joint and proximal neural structures

The lumbosacral plexus runs directly across the anterior joint capsule of the SIJ. Defects in the capsule of the SIJ can result in the spread of the injectate to neural and other pain-generating structures through the following pathways: (1) dorsal spread to the first three dorsal sacral foramina, (2) superior spread to the fifth lumbar nerve root, and (3) ventral spread to the lumbosacral plexus. Various patterns of neural spread were observed in over 60% of joints injected with contrast by Fortin et al.$^{42}$ In a study evaluating landmark-based SIJ injections (no imaging guidance), Rosenberg reported a 44% incidence of spread into the sacral neural foramina (pre-dominantly S1) and a 24% incidence of epidural spread.$^{147}$ In another study evaluating the arthrographic patterns of IA SIJ injections, nine of the 13 patients (69%) with pain relief following the injection exhibited tears in the ventral capsule of the joint.$^{43}$ In this study, the authors found a positive but weak association between capsular tears and analgesic benefit. In a study that enrolled patients with suspected SIJ complex pain after lumbar or lumbosacral spinal fusion surgery, four of the 52 patients (7.7%) patients experienced transient weakness of the lower extremities after IA SIJ injections with LA and corticosteroids.$^{53}$ The authors attributed this to leakage of the injectate from the ventral aspect of the SIJ.

It is difficult to interpret these findings because while capsular tears may raise the likelihood of an SIJ being painful, they can also result in false-positive responses to IA diagnostic injections with LA. Presence of sensorimotor neural blockade in the ipsilateral lower limb following IA SIJ injection of LA suggests spread of LA to the lumbosacral plexus through a tear in the ventral capsule of the SIJ. However, absence of this finding cannot rule out spread to neural structures adjacent to the joint.

Summary:

$\bullet$ IA injections into the SIJ should be performed with a maximum of 2.5 mL of total injectate to prevent capsular rupture and to reduce spread to adjacent structures; if an increase in resistance to the injection is perceived at lower volumes (eg, 2.0 mL), the injection may be halted to avoid tears in the capsule of the joint; Grade B recommendation, moderate level of certainty.

In patients with suspected IA and EA pain arising from the SIJ itself and the ligaments overlying the posterior aspect of the joint, a total injectate volume higher than 2 mL may be required to block pain generators and treat

dysfunction. Up to and sometimes exceeding 3 mL (depending on the extent of pain and pathology) of injectate can be used to achieve this goal; Grade C recommendation, low level of certainty.

* The response to diagnostic IA SIJ injections can be used to select patients for surgical fusion of the SIJ; Grade B recommendation, moderate level of certainty.

* The addition of corticosteroids to LA for IA SIJ injections for diagnosing pain arising from the SIJ does not confer any clear diagnostic benefit while potentially increasing the volume of injectate, thereby increasing the chance of extra-articular spread. Corticosteroids should not be routinely added to LA during purely diagnostic IA injections; Grade D recommendation against the intervention; low level of certainty.

* The volume of LA for SLBBs should range between 0.67 and 1.2 mL per foraminal region to account for variable neuroanatomy and for the possibility of multiple nerves per level, and should be performed at different depths; Grade C recommendation, very low level of certainty.

Q6. How accurate are the use of anatomical landmarks, ultrasound, and CT scan for obtaining intra-articular spread during a sacroiliac joint injection?

Landmark-guided sacroiliac joint injections

Randomized trials

A single randomized multicenter study by Cohen et al.$^{6}$ evaluated the accuracy of landmark-based injections and compared clinical outcomes between landmark-guided and fluoroscopy-guided SIJ injections. The study randomized 125 patients with clinically suspected non-inflammatory SIJ complex pain supported by positive SIJ provocative maneuvers; 61 received landmark-guided injections, and 64 received fluoroscopy-guided injections. Following final needle placement based on randomization (eg, landmark vs fluoroscopy), contrast was injected under fluoroscopy for both groups prior to administration of LA and corticosteroid. In the landmark group, 8% of injections were either IA or IA and EA, whereas in the fluoroscopy group, 44% injections were IA and 47% were IA and EA. In the landmark-guided group, the remaining 92% of injections did not have any IA SIJ contrast spread. For the landmark-guided group, six injections were noted to have contrast spread in the piriformis muscle; one injection had contrast spread over the L5-S1 facet joint; three injections had contrast spread inside a sacral foramen; one injection had contrast spread within 3 cm of the hip; and one injection had contrast spread in the pelvic outlet. Regarding clinical outcomes, no significant difference in pain reduction was observed at 1 month between groups [-2.3 points change in pain numerical rating scale (NRS) in the fluoroscopy group vs -1.7 points in the landmark group; P = .23]. Although the reductions in average pain and worst pain were greater in the fluoroscopy group than in the landmark group at 3 months (-1.8 and -2.2 points in the fluoroscopy group vs -0.9 and -1.4 points in the landmark group; P = .049 and .02, respectively), these clinical differences were modest and accompanied by a higher cost for fluoroscopically guided injections (see Table 11).

contrast was injected under fluoroscopy for both groups prior to administration of LA and corticosteroid. In the landmark group, 8% of injections were either IA or IA and EA, whereas in the fluoroscopy group, 44% injections were IA and 47% were IA and EA. In the landmark-guided group, the remaining 92% of injections did not have any IA SIJ contrast spread. For the landmark-guided group, six injections were noted to have contrast spread in the piriformis muscle; one injection had contrast spread over the L5-S1 facet joint; three injections had contrast spread inside a sacral foramen; one injection had contrast spread within 3 cm of the hip; and one injection had contrast spread in the pelvic outlet. Regarding clinical outcomes, no significant difference in pain reduction was observed at 1 month between groups [-2.3 points change in pain numerical rating scale (NRS) in the fluoroscopy group vs -1.7 points in the landmark group; P = .23]. Although the reductions in average pain and worst pain were greater in the fluoroscopy group than in the landmark group at 3 months (-1.8 and -2.2 points in the fluoroscopy group vs -0.9 and -1.4 points in the landmark group; P = .049 and .02, respectively), these clinical differences were modest and accompanied by a higher cost for fluoroscopically guided injections (see Table 11).

injection had contrast spread within 3 cm of the hip; and one injection had contrast spread in the pelvic outlet. Regarding clinical outcomes, no significant difference in pain reduction was observed at 1 month between groups [-2.3 points change in pain numerical rating scale (NRS) in the fluoroscopy group vs -1.7 points in the landmark group; $P = .23$]. Although the reductions in average pain and worst pain were greater in the fluoroscopy group than in the landmark group at 3 months (-1.8 and -2.2 points in the fluoroscopy group vs -0.9 and -1.4 points in the landmark group; $P = .049$ and .02, respectively), these clinical differences were modest and accompanied by a higher cost for fluoroscopically guided injections (see Table 11).

Non-randomized trials

Three observational studies prospectively assessed the accuracy of landmark-guided SIJ injections. One study by Hansen

Pain Medicine, 2025, Vol. 00, No. 0

|Author, year|Patient population|Design|Results|Comments| |---|---|---|---|---| |Landmark guidance Rosenberg, $2000^{147}$|37 SIJ injections in 33 pts with positive SIJ provocation maneuvers were performed.|A prospective study assessing the accuracy of landmark-guided SIJ injections with confirmation using CT scan.|IA SIJ injection was accomplished in 8 of 37 (22%) patients. Clinically-guided needle placement was based on palpation of the posterior superior iliac spines and identification of the midline.|Injected material was identified within 1 cm of the joint 68% of the time. Epidural contrast was observed 24% of the time and foraminal spread in 44% of injections.| |Hansen, $2003^{152}$|60 pts with positive SIJ provocation maneuvers underwent landmark-guided SIJ injection.|An observational study assessing the accuracy of landmark-guided SIJ injections with confirmation using FL.|Only 5 of 60 pts (12%) had IA SIJ needle placement.|The posterior superior iliac spine as the landmark was found to be a poor indicator of IA SIJ anatomic access.| |Cohen, $2019^{6}$|125 pts with positive SIJ provocation maneuvers were randomized to receive landmark-guided injections or FL-guided injections.|A multicenter randomized comparative-effectiveness study to assess the accuracy of landmark-guided and FL-guided SIJ injections and compare outcomes.|In the landmark group, 8% of injections were either IA or IA and EA. In the FL group, 44% of injections were IA, 47% were IA and EA, and 9% were EA only.|No significant differences in pain reduction were observed between groups at 1 month. At 3 months, the improvement in average pain (-1.8 vs -0.9 points) and worst pain (-2.2 vs -1.4 points) were greater in the FL group than in the landmark group. No differences in clinical outcomes were observed between IA and other types of injections.| |Fluoroscopy guidance Maigne, $1996^{80}$|67 pts with chronic LBP referred for diagnostic IA SIJ injections.|Observational study assessing the diagnostic accuracy of clinical examination in identifying symptomatic SIJ complex pain.|Penetration of joint not possible in 10.4% of pts, sciatic palsy indicating extravasation occurred in 4.5%.|Used double injections| |Stelzer, $2019^{150}$|Cadaveric study with 20 subjects (40 joints).|A prospective study comparing the accuracy of lower US-guided SIJ injection with FL-guided SIJ injection using confirmation with dissection of injected colored latex.|US-guided injections were IA in 1/20 (5%), and EA in 19/20 (95%) with intra-ligamentous spread of latex. FL-guided injections were IA injection in 10/20 (50%) cases, 8/20 (40%) showed intra-ligamentous spread, and 2/20 (10%) revealed vascular spread.|With US, there was good structural visibility in 11 of 20 (55%) cadavers and perception of the needle being IA in 10 of 20 (55%). With FL, good structure visibility occurred in 9 of 20 (45%) cadavers and perception of the needle being IA occurred in 17 of 20 (85%) specimens.| |Ultrasound guidance Pekkafahli, $2003^{157}$|34 consecutive pts with SIJ complex pain diagnosed radiologically or based on physical exam.|A prospective study assessing the accuracy of US-guided SIJ injections with confirmation using FL.|Among 60 US SIJ injections, 46 (76.7%) were IA. The IA injection rate was 60% for the first 30 injections, 93.5% in the last 30 injections, with the last 20 US-guided injections all being IA.|The mean US-guided procedure time was 9 min (range, 4-18 min). Pain outcomes not assessed.|

(continued)

30

|Author, year|Patient population|Design|Results|Comments| |---|---|---|---|---| |Klauser, $2008^{160}$|Cadaver study in 10 subjects (20 joints).|A prospective study comparing the accuracy of upper-level and lower-level SIJ injection with CT confirmation.|US-guided IA needle placement at the upper level occurred in 7 out of 10 (70%) specimens, and at the lower level in 9 of 10 (90%).|The mean procedure time was 3.09 min and was longer in the first cadavers than in the last cadavers. In 4 cases that were not IA, attempts at another level resulted in IA injection.| |Hartung, $2010^{158}$|14 pts (20 joints) with acute sacroiliitis.|A prospective study comparing US-guided SIJ injections with 1 mL corticosteroid in those with IA (40%) vs EA (60%) spread based on MRI.|8 SIJ injections (40%) were IA whereas the other 12 injections (60%) were EA. Similar pain relief was observed in both groups.|Study underpowered to detect a difference in outcomes. 1 mL corticosteroid mixed with contrast, without LA.| |Jee, $2014^{154}$|120 pts with noninflammatory SIJ arthritis.|RCT comparing outcomes with US-guided vs FL-guided SIJ injections up to 12 weeks following injection.|No differences were observed in the ODI and NRS pain scores between groups. The US-guided SIJ was IA in 48 of 55 87.3% of pts and in 54 of 55 (98.2%) pts in the FL-guided group.|Vascularization around the SIJ was seen in 34 of 55 pts in the US group. 3 cases of intravascular injection occurred in the FL group.| |Perry, $2016^{161}$|Cadaver study with nine subjects (17 joints).|A prospective study assessing the accuracy of US-guided SIJ injection with confirmation by dissection of injected paint marker.|Among 17 SI joints, 15 (88.2%) were injected with paint marker. Among the 15 joints with IA placement, 5 joints (33.3%) showed partial EA spread and required redirection of the needle. 3 joints (20%) had EA spread not seen during US.|One of the joints with no IA spread was found to be partially frozen at the time of dissection; the second joint was considered an unsuccessful injection because of osteophytes at the joint line.| |Soneji, $2016^{151}$|40 pts with unilateral SIJ pain diagnosed by physical exam (eg, SIJ provocation maneuvers).|RCT comparing outcomes with US-guided vs FL-guided SIJ injections up to 3 months following injection.|The US-guided SIJ injections were IA in 10 of 20 (50%) vs 13 of 20 (65%) in the FL-guided group.|Reduction from pain scores observed in both groups at 1 month but no difference between groups.| |De Luigi, $2019^{155}$|50 pts with SIJ-mediated LBP diagnosed by physical exam (eg, SIJ provocation maneuvers).|A prospective study assessing the accuracy of US-guided SIJ injections with confirmation using FL.|In 48 out of 50 (96%) pts, IA contrast spread was seen on FL.|The 2 pts w/o IA spread both had a BMI >40. Of the 48 joints with IA spread, one had vascular uptake in addition to an IA arthrogram.| |Fouad, $2021^{156}$|34 pts with acute or chronic SIJ complex pain diagnosed by physical exam (eg, SIJ provocation maneuvers).|A prospective study assessing the accuracy of US-guided SIJ injections with confirmation using FL after final needle position.|33 of 39 (84.6%) injections were IA while 6 of 39 (15.4%) injections were EA.|Both groups improved with no difference between groups.|

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

|Author, year|Patient population|Design|Results|Comments| |---|---|---|---|---| |CT guidance Artner, $2012^{163}$|30 pts with LBP and positive SIJ provocation testing who underwent bilateral SIJ injections.|Prospective study assessing outcomes with low-dose CT-guided injections and retrospective comparison of radiation dose to standard CT- and FL-guided injections.|100% of low-dose CT injections were IA with radiation dose (4.57 mGyc2) lower than FL-guided injec- tions (3.73 mGyc2) and standard CT (76.27 mGyc2m)|The accuracy of historical standard CT- and FL-guided injections was not reported. Clinical outcomes not reported.| |Althoff, $2015^{164}$|29 pts (21 bilateral, 8 unilateral) with inflammatory back pain and active SIJ inflammation confirmed by MRI.|A prospective study evaluating clinical outcomes 6 months after treatment.|23 pts (76%) had IA injection and 7 (24%) pts had EA injection. One patient with bilateral injections had one IA and the other EA.|At 3 and 6 months, greater improvement in the IA group than the EA group.| |Paik, $2016^{166}$|34 pts with presumed spondyloarthropathy or nonspondyloarthropathic SIJ dysfunction.|Retrospective study assessing outcomes after CT- with intermittent FL-guided SIJ injections.|All injections were IA (100%) with median procedural times of 5.28 min for unilateral and 6.72 min for bilateral SIJ injections and a median radiation dose of 0.11 mSv|When compared to reported procedural times in the literature, time was > FL-guided, but <standard CT-guided, US-guided, and MRI- guided SIJ injections| |Bronsard, $2020^{165}$|94 pts with SIJ complex pain with positive Fortin finger sign who underwent prior lumbosacral spinal fusion.|A retrospective study comparing outcomes two days after 1-3 IA and EA injections.|Of 146 SIJ injections, 61% were IA vs 100% of 34 3-dimensional injections.|41% of 2-dimensional injections were effective vs 73% of 3-dimensional injections. Long-term outcomes not assessed.| |Bessar, $2021^{162}$|52 pts with clinically-suspected SIJ complex pain.|RCT comparing outcomes with CT-guided vs FL-guided SIJ injections up to 12 months following injection.|Difficult localization of the SIJ space in 5 pts in the FL group. Improvement in pain score and function from baseline in both groups. Significant difference between groups favoring CT for both pain and function at 6 months, but only pain at 12 months. All injections were IA (100%).|Blinding was not noted. Improvement in both groups with non-significant difference favoring CT-guided group. No difference in radiation dose.| |Oren, $2021^{167}$|17 pts with suspected SIJ-related lower back or buttock pain.|Retrospective study assessing outcomes after CT with fused FL-guided SIJ injections.|All injections were IA (100%).|Of 23 injections, 100% pain relief was achieved 2 h after SIJ injection in 18 of 23 (78%). Long-term outcomes not assessed.|

CT, computerized tomography; EA, extra-articular; FL, fluoroscopy; F/U, follow-up, IA, intra-articular; LA, local anesthetic; LBP, low back pain, NRS, pain numeric rating score, ODI, Oswestry Disability Index; RCT, randomized controlled trial; SIJ, sacroiliac joint; US, ultrasound; VAS, visual analog scale.

Pain Medicine, 2025, Vol. 00, No. 0

et al.$^{152}$ included 60 patients with positive SIJ provocative maneuvers who underwent a SIJ injection using a landmark-guided technique performed by a single experienced physician. The posterior superior iliac spine was used as the landmark, with no imaging to guide injection. Based on fluoroscopic confirmation, only five of 60 patients (12%) had IA needle placement. The authors concluded that because blind placement of the needle infrequently results in IA spread, SIJ injections should be performed under fluoroscopic visualization.

A recent retrospective study by Hasoon et al. compared 2-week outcomes in 10 patients with clinically-suspected SIJ pain, 5 of whom underwent landmark-guided injections by inserting a needle slightly inferior to the PSIS, and 5 of whom underwent fluoroscopically-guided injections. Each group received 3 mL injections containing 2 mL of bupivacaine 0.25% and 40 mg of triaminolone. Two weeks post-injection, the average pain relief was 76% in the image-guided group, which favorably compared to the mean 56% pain relief in the landmark-guided patients.$^{153}$

The inaugural study evaluating the accuracy of blind injections was conducted by Rosenberg et al.$^{147}$ The authors performed 37 SIJ injections in 33 patients who had positive SIJ provocation maneuvers. This study assessed the accuracy of landmark-guided SIJ injections based on palpation of the posterior superior iliac spine and identification of the midline, with position confirmed by CT scan. The final needle location was observed within the SIJ in only eight of 37 (22%) patients, with foraminal and epidural spread noted in 44% and 24% of injections, respectively. The authors concluded that the low accuracy rates for landmark-guided SIJ injections limit their use for injection therapy.

Fluoroscopically-guided sacroiliac joint injections

Randomized trials

As noted above, the accuracy rate in the randomized multi-center study by Cohen et al.$^{6}$ comparing fluoroscopy-guided and landmark-guided SIJ injections found that 91% of fluoroscopically-guided injections were either IA or that the injectate spread within both the joint and to the posterior ligaments.

Non-randomized Studies

In a cadaveric study by Stelzer et al.$^{150}$ the accuracy of fluoroscopically guided SIJ injections was assessed, and joint spread was confirmed with dissection after injection of colored latex. The fluoroscopy guided SIJ injections showed IA spread in 50% of specimens, 40% of injections were within the posterior ligaments, and 10% demonstrated intravascular spread; it is noteworthy that for obvious reasons, readjustment was not performed after contrast injection.

There are several observational studies that assessed analgesic benefit with fluoroscopically-guided SIJ injections.$^{68,80,140}$ These studies prospectively assessed the diagnostic utility of clinical examination in identifying SIJ complex pain based on the outcomes of fluoroscopically-guided IA SIJ injections. In the study by Maigne et al., 1 mL of contrast was injected into the SIJ, followed by 2 mL of lidocaine 2% for the first injection and bupivacaine 0.5% for the second injection. The authors reported that 35% of patients after the first injection and 18.5% after the second injection experienced 2.75% pain reduction.$^{80}$ In the study published in 2005 by Laslett et al., 1 mL of contrast and 1 mL

of LA was injected into the SIJ, and 33% of patients reported pain relief.$^{140}$ In the 2003 publication by Laslett et al., 35% responded positively to the first injection; among these, 18.5% experienced temporarily relief from the confirmatory SIJ injection. However, no attempt was made to interpret the accuracy of fluoroscopic guidance for IA SIJ injections in these studies, and pain relief following injection was used as a surrogate.

Ultrasound-guided sacroiliac joint injections

Randomized trials

Two randomized trials compared clinical outcomes between ultrasound-guided and fluoroscopically-guided SIJ injections. In a study by Jee et al.,$^{154}$ 120 patients with non-inflammatory SIJ arthritis were randomized either to ultrasound-guided SIJ injections or fluoroscopically-guided injections. The spread of contrast was assessed by a blinded radiologist. The ultrasound-guided injections were IA in 87.3% of patients, and the fluoroscopy-guided injections were IA in 98.2% of patients (P < .05). Vascular injection was observed in 34 of 55 (62%) of ultrasound-guided injections while three cases of intravascular injection were found in the fluoroscopy group. No differences were observed in NRS pain and Oswestry Disability Index (ODI) scores up to 12 weeks following injections. Another study by Soneji et al.$^{151}$ randomized 40 patients with unilateral SIJ arthritis to receive either ultrasound-guided or fluoroscopically guided SIJ injections, with participants followed for 3 months. In the ultrasound-guided group, contrast spread was assessed with fluoroscopy. Blinding was not reported. The ultrasound-guided SIJ injections were IA in 50% of patients, and fluoroscopically-guided injections were IA in 65% of patients. No significant differences in NRS pain or ODI scores were observed between groups.

Non-randomized trials

Several studies prospectively assessed the accuracy of ultrasound-guided SIJ injections using fluoroscopy as the reference standard.$^{155-157}$ In the study by Pekkafahli et al.,$^{157}$ 34 consecutive patients with SIJ complex pain underwent 60 ultrasound-guided SIJ injections, with 76.7% found to be IA. The IA injection rate appeared related to proceduralist experience, with an accuracy rate of 60% for the first 30 injections, which increased to 93.5% in the subsequent 30. A study by De Luigi et al.$^{155}$ reported a higher rate (96%) of IA injection in 50 patients with SIJ-mediated LBP who underwent ultrasound-guided SIJ injection. Only one injection resulted in vascular uptake of contrast. The two patients who had undergone ultrasound-guided SIJ injections without IA contrast spread had a body mass index over 40 kg/m$^{2}$. In 34 patients, Fouad et al.$^{156}$ found that 84.6% of injections were IA while 15.4% of injections were EA. Using MRI confirmation after ultrasound-guided SIJ injections in 14 patients (20 injections), Hartung et al.$^{158}$ observed IA spread in eight (40%) injections and EA spread in 60%. Lastly, a large retrospective review of 8,226 patients who underwent musculoskeletal injections by Hynes et al.$^{159}$ observed no difference in the complication rate between ultrasound- and fluoroscopically-guided musculoskeletal injections (SIJ injections were included with other injections).

Three studies assessed the accuracy of ultrasound-guided SIJ injections in cadaver models, with conflicting results. In a study by Klauser et al., $^{160}$ the accuracy of upper-level and

Pain Medicine, 2025, Vol. 00, No. 0

lower-level ultrasound-guided SIJ injections was confirmed using CT. In a total of 20 joint injections, the authors observed IA needle placement at the upper level in 70% vs 90% at the lower level. The overall success rate for ultrasound-guided IA needle placement was 80% (16 of 20 SIJs). Another cadaver study by Perry et al.$^{161}$ assessed the accuracy of ultrasound-guided SIJ injections through the dissection of injected paint markers. Out of 17 SIJ injections, 15 (88.2%) were injected in the IA space, but five (33.3%) also showed partial EA spread on ultrasound during initial injection, necessitating redirection of the needle. Three of the 15 IA joint injections exhibited partial EA spread not observed during ultrasound-guided needle placement. Finally, as above, a study by Stelzer et al.$^{150}$ compared the accuracy of ultrasound-guided SIJ injections to fluoroscopically-guided SIJ injections, evaluating joint spread by dissection after injection of colored latex. With ultrasound-guided SIJ injections, the IA spread was observed in 5% of injections, with 95% of injections being EA. In contrast, the fluoroscopy-guided SIJ injections showed IA spread in 50%, 40% showed intra-ligamentous spread, and 10% resulted in intravascular uptake.

CT-guided sacroiliac joint injections

Randomized trial

A single RCT compared clinical outcomes with low-dose CT-guided IA SIJ injections to fluoroscopically-guided IA SIJ injections. In a study by Bessar et al.,$^{162}$ 52 patients with SIJ complex pain were randomized to either low-dose CT-guided IA SIJ injection or fluoroscopically-guided IA SIJ injection and followed for 12 months. In both groups, significant improvements in pain and function were observed through 12 months, with results generally being more favorable in the CT group. In this study, blinding was not reported. Difficulty in localization of the SIJ space was reported in five patients who underwent fluoroscopically-guided injections. More patients were satisfied in the CT group, with no significant difference in the radiation dose between groups.

Non-randomized studies

Two prospective studies assessed the accuracy and clinical outcomes of CT-guided IA SIJ injections. In a study by Arner et al.,$^{163}$ 30 patients with LBP and positive SIJ provocation testing underwent low-dose CT-guided bilateral IA SIJ injections, with 100% of injections being IA. The authors also found no significant difference in radiation dose when compared to fluoroscopically-guided IA SIJ injections performed the previous month (4.57 mGycm$^{2}$ with low dose CT-guided injections vs 3.73 mGycm$^{2}$ with pulsed fluoroscopically-guided injections). The authors did not delineate how accuracy was determined. When the accuracy was assessed using confirmation by a blinded evaluator in a later study by Althoff et al.,$^{164}$ CT-guided SIJ injections were observed to be IA in 76% of procedures.

Three studies retrospectively assessed the accuracy of CT-guided injections. In a study by Bronsard et al.,$^{165}$ 94 patients with SIJ complex pain after previous lumbosacral spinal fusion surgery underwent up to three CT-guided IA SIJ injections confirmed with arthrography. Out of 146 SIJ injections, 61 were IA, and 41% were effective (>70% pain improvement) two days after injection. It should be noted that patients received repeat injections only if prior injections were ineffective (up to three injections), and the third

injection was performed with 3-dimensional surgical navigation. In retrospective studies by Paik et al.$^{166}$ and Oren et al.,$^{167}$ combined CT-guidance with fluoroscopy was used, with both studies reporting 100% IA spread confirmed by arthrography. In the study by Paik et al., the authors found that the time required for combined CT-fluoroscopy procedures was greater than the reported times for fluoroscopically-guided IA SIJ injections, but less than for standard CT-guided, ultrasound-guided, and MRI-guided IA SIJ injections.

Summary:

$\bullet$ In those with suspected IA SIJ pathology, accuracy for intra-articular injectate spread is greater with fluoroscopically-guided SIJ injections compared to landmark-guided SIJ injections; Grade B evidence, moderate level of certainty.

$\bullet$ IA SIJ injections without imaging guidance (such as landmark or palpation-guided injections) should not be used when performing IA SIJ injections; Grade A recommendation, high level of certainty.

* The use of image guidance should be strongly considered when performing SIJ complex injections when extra-articular involvement is suspected to delineate needle position; Grade B recommendation, moderate-to-high level of certainty.

$\bullet$ The use of fluoroscopic image guidance should be coupled with contrast injection (if no contraindication) to confirm intra-articular placement; Grade C recommendation, low level of certainty.

* Ultrasound guidance should be considered in patients where radiation exposure (eg, pregnancy, children) and/or contrast (eg, anaphylactic reaction) is contraindicated; Grade C-B recommendation, moderate level of certainty.

$\bullet$ The evidence is insufficient to determine the non-inferiority of ultrasound-guided compared to fluoroscopically-guided SIJ injections for accuracy of IA placement; Grade I.

$\bullet$ The accuracy of ultrasound-guided injections depends on the experience of the proceduralist and patient characteristics (eg, body mass index); Grade C evidence, low level of certainty.

* The evidence is insufficient to compare the accuracy of CT-guided to fluoroscopically-guided SIJ injections; Grade I evidence.

$\bullet$ In patients with advanced degenerative changes of the SIJ noted on preprocedural imaging or who have had prior unsuccessful fluoroscopically-guided SIJ injections, low-dose CT guidance with or without fluoroscopy can be considered; Grade C recommendation, low level of certainty.

Q7. Which is more effective for the treatment of sacroiliac joint complex pain, intra- or extra-articular sacroiliac joint complex injections?

The SIJ complex is a sophisticated structure composed of the sacral and iliac bones, which form the joint space enclosed by a fibrous capsule and synovial lining. This complex is supported by an extensive network of posterior and anterior ligaments, as well as a diverse group of muscles that contribute to its stability and function. These muscles include the erector spinae, multifidus, gluteal group, psoas, quadratus lumbo rum, piriformis, obliques, hamstrings, and pelvic floor muscles. The intricate interplay of these osseous, articular,

Pain Medicine, 2025, Vol. 00, No. 0

ligamentous, and muscular components ensures the structural integrity and proper biomechanical function of the SIJ complex. Nociceptors are abundantly present throughout these structures, and researchers have used pain provocation techniques and anesthetic studies on both patients and healthy volunteers to develop comprehensive pain referral maps for IA and EA SIJ complex pain.$^{3,4,59-61,79,129}$

The probability of EA or IA sources being the primary SIJ complex pain generator depends on many factors. Extra-articular pathology is more likely to be predominant in a young individual with unilateral pain who reports a specific inciting event, and injections into the dorsal ligaments typically require a higher volume than for IA injections.$^{1}$ Extra-articular SIJ pain is often associated with exquisite tenderness, and normal or near-normal imaging, although certain MRI sequences such as short tau inversion recovery (STIR) can detect acute inflammatory processes. In contrast, IA pathology is more likely to occur insidiously in an older person with bilateral pain, and is generally associated with radiological abnormalities.$^{1}$ Some patients also have IA and EA pathology occurring simultaneously, as suggested by controlled trials demonstrating efficacy for EA injections in patients with spondyloarthropathies,$^{9}$ and a trend toward a higher responder rate and greater pain relief when a combination of IA and EA spread was observed (60%) compared to when only IA (48%) or EA (54%) spread occurred (P = .25) in injection-based outcome studies.$^{6}$ In a retrospective study conducted in 120 patients, Borowsky and Fagen reported a higher proportion of responders following combination IA/EA injections than for IA injections alone.$^{139}$ In a recent randomized trial by Tantawy and Nazim in which the same dose and volume of corticosteroid and local anesthetic was injected either into or both into (IA) and outside (EA) the joint, the authors found no difference in pain scores at 1 week, but superiority for the combination group at 1- and 6-months (see Table 12).$^{168}$

Clinical studies

Randomized trials

Direct comparisons

In a double-blind study, Cohen et al. randomized 125 patients with clinically-diagnosed SIJ complex pain to landmark-guided injections with sham fluoroscopy or fluoroscopically-guided injections in which the target was IA.$^{6}$ Only 8% of the landmark-guided injections were IA, which is consistent with the low rate of IA spread without image guidance found in other studies.$^{147,152}$ Although greater immediate pain relief and a higher short-term responder rate (69% for combination spread, 62% for IA-only spread, 57% for only EA) was found for combination injections, the differences between groups were not significant. Whereas no significant differences were observed between groups at the 1-month primary endpoint, at 3-month follow-up some outcome measures favored IA over EA spread, which is consistent with the short duration of benefit for soft tissue injections.

Khalil et al. performed a quasi-randomized (based on laterality) trial in which 96 patients with severe clinically-identified SIJ pain were allocated based on their most painful side to receive fluoroscopically-guided right-sided IA injections with a 3 mL mixture of methylprednisolone and 2 mL of 2% lidocaine injected into the center of the joint without contrast to confirm placement (n = 46), or left-sided EA injections administered in the upper, middle and lower regions

containing the same injectate (n = 50).$^{7}$ The authors found statistically significant differences favoring EA injections immediately after the injection (3.0 vs 4.3 and 0.4 vs 1.3 at each of the 2 hospitals), at 1-week (2.2 vs 4.3 and 1.3 vs 2.3), 1-month (1.9 vs 4.6 and 0.4 vs 2.2) and 3-month (1.7 vs 4.7 and 0.3 vs 2.8) follow-ups. This study included only individuals with baseline scores of at least 8 on a 0-10 NRS, did not provide combined data, and incorrectly stated that the blind injections outside of the joint were into different parts of the joint capsule.

Indirect comparisons from randomized trials

There are three placebo-controlled trials that have evaluated outcomes following SIJ injections. In the first one by Maugars et al., 10 patients (13 joints) with spondyloarthropathy and active sacroiliitis were randomized to receive IA corticosteroids or saline following joint arthrography in a double-blind fashion.$^{11}$ At 1-month follow-up, five of the six joints injected with corticosteroid were deemed to have ≥50% pain relief vs one of seven in the saline group. At 1 month, six of the seven control joints were reinjected with corticosteroid. At 3- and 6-month follow-ups, 62% and 58% of participants continued to report pain reduction, respectively.

In the two randomized, double-blind placebo-controlled trials by the same group of investigators that evaluated EA injections, the authors compared depo-corticosteroid and local anesthetic injections to saline and local anesthetic injections in 20 patients with seronegative spondyloarthropathy in their initial study,$^{9}$ and in 24 patients without spondyloarthropathy in a follow-up study.$^{10}$ At both 1- and 2-month follow-up, the treatment group experienced a greater reduction in pain than the control group. In addition to the short follow-up periods, another flaw in these two studies is that the local anesthetic used in both treatment groups is not a true placebo.$^{141}$

Non-randomized comparative-effectiveness studies

In the first study to compare IA to EA SIJ injections, Murakami et al.$^{137}$ performed a non-randomized, evaluator-blinded, prospective study in 50 patients comparing immediate-term (5-min post-injection) outcomes for lidocaine IA injections and EA injections performed in four locations along the SIJ. The EA injections were performed with 0.5-1 mL aliquots of lidocaine in areas where hypertonic saline provoked pain. The authors reported significant improvement in six activities of daily living in 100% of the EA injections vs 36% of the IA injections. They then performed EA injections on the IA non-responders, again reporting a 100% response rate. In addition to the unrealistic 100% response rate, other flaws include selection bias and the outcome measure, which was immediate and did not involve pain relief.

In an observational study by the same group, the authors performed EA injections in 85 patients with pain around the posterior superior iliac spine. The injectate consisted of 0.5 mL aliquots of 2% lidocaine mixed with contrast in individuals whose pain was reproduced by needle irritation.$^{136}$ In those who failed to experience a positive outcome, defined as 270% pain relief 15- to 30-min post-injection, a fluoroscopically-guided IA injection was performed with a 50:50 mixture containing betamethasone and bupivacaine 0.5%. Seventy-two (85%) patients experienced a positive response,

Pain Medicine, 2025, Vol. 00, No. 0

|Author, year|Patients|Design|Results|Comments| |---|---|---|---|---| |Murakami, $2007^{137}$|50 pts with axial LBP and >1 (+) SIJ provocation test|Prospective study comparing IA SIJ injections with 1-2.5 mL LA or 0.5-1 mL EA injections in pts who experienced pain with hypertonic saline provocation in multiple areas|EA injection effective (attainment of predefined functional endpoints) in 100% of pts vs 36% for IA after 5 min.|Non-randomized, blinded evaluation. All non-responders to IA injections responded to EA injections, with 100% response rate raising questions of validity.| |Borowsky & Fagen, $2008^{139}$|120 pts with LBP below the L4 spinal level and radiation into buttock, thigh, groin or calf|Retrospective study comparing fluoroscopically guided IA to IA + EA SIJ injections medial to joint line from S1-3|Combination > IA-only injections through 3 mos.|Unclear why the authors only analyzed data on 120 of 164 eligible pts. Equivalent total corticosteroid dose, but higher volume of LA in the combination group.| |Hartung, $2010^{158}$|14 pts (20 joints) in pts with acute sacroiliitis|Prospective study comparing ultra-sound-guided SIJ injections with 1 mL corticosteroid in those with IA (40%) vs EA (60%) spread based on MRI|No significant differences between IA & EA group at 24 hrs. At 4 wts, there was a non-significant trend toward IA > EA (3.3 vs 2.5 points on a 0-10 scale)|Study underpowered to detect difference. 1 mL corticosteroid mixed with contrast, without LA.| |Nacey, $2016^{169}$|99 pts (113 joints) who underwent SIJ injections by radiologists (unclear selection criteria)|Retrospective study comparing outcomes with IA (49%) vs EA (51%) SIJ injections with 1 mL corticosteroid + LA|No significant differences immediately or at 1 week, though non-significantly higher responder rates in IA group at 1 week.|Evaluators blinded to outcome. 13% of cases in which there was disagreement regarding spread pattern resolved by consensus discussion.| |Murakami, $2018^{136}$|85 pts with pain around the posterior superior iliac spine|Non-randomized, observational study evaluating EA injections, and in non-responders, IA injections|Among the 72 pts with a positive response (≥70% pain relief 15-30-min post-injection), 81% were in the EA group vs 19% in the IA group.|Higher proportion of placebo responders were likely included in the EA group. Total volumes administered not described.| |Cohen, $2019^{6}$|125 pts with clinically-diagnosed SIJ pain|Randomized, double-blind study evaluating fluoroscopically-guided vs landmark-guided injections|<10% of landmark-guided injections IA. No differences between fluoroscopically-guided or landmark-guided, or IA vs EA injections at 1 month. At 3 mo, some outcome measures favored fluoroscopically-guided injections (mean reduction in avg. pain 1.8 vs 0.9).|Very small, non-significant difference in mean change in avg. pain score 1 month stratified by injection location. At 3 months, mean reduction in avg. pain score in IA, EA and combination group 1.9, 1.0 and 1.4, respectively (P = .18). For worst pain, mean reductions 2.6, 1.5, 1.7, respectively (P = 0.12).| |Khalil, $2022^{7}$|96 pts with severe clinically diagnosed SIJ complex pain|Quasi-randomized, non-blinded study evaluating fluoroscopically-guided IA injections (right-sided pain predominance) vs landmark-guided left-side EA injections|Statistically significant differences favoring the EA group at all time points through 3 months.|Incorrectly asserted that the EA injections were in the joint capsule. Did not combine data for both hospitals, and the statistical analysis was flawed.|

EA, extra-articular; IA, Intra-articular; LA, local anesthetic; SIJ, sacroiliac joint.

Pain Medicine, 2025, Vol. 00, No. 0

with the successful group comprised of 81% (n = 58) EA and 19% (n = 14) IA injections. In addition to the non-randomized design, a disproportionate percentage of placebo responders were likely included in the positive EA group.

As alluded to above, in a retrospective study in which the authors compared IA corticosteroid and local anesthetic injections (n = 40) to IA and EA injections (n = 80), the proportion of patients who obtained ≥50% pain relief at 3 weeks (42.5% vs 27.5%) and 3 months (31.3% vs 12.5%) was higher in the combination group.$^{139}$ Although the total dose of corticosteroid was the same in the two groups, the combination group received a higher volume of bupivacaine (4 mL vs 1.5 mL). The authors also did not provide a basis as to why data were only analyzed on 120 of the 164 eligible patients. Hartung et al.$^{158}$ performed a prospective study in 14 patients (20 joints) to determine the accuracy of ultrasound-guided IA SJ injections, finding only a 40% success rate using MRI as the reference standard. Although there was no difference in pain reduction at 24 h between those patients with IA spread and those with only EA spread, at 4 weeks there was a trend toward greater pain relief in the IA group (3.3 vs 2.5 points on a 0-10 numerical rating scale).

Nacey et al.$^{169}$ retrospectively reviewed the records of 113 fluoroscopically-guided SIJ injections performed with 1 mL of corticosteroid and local anesthetic in 99 patients. The injection location was stratified based on a review of images by two musculoskeletal radiologists. For the 55 (49%) of injections deemed to be IA, the mean immediate reduction in pain was 4.4 vs 4.2 for the 58 patients whose injections were EA. One-week post-injection, both groups reported an average pain reduction of 1.9 points on a 0-10 scale. The responder rates immediately post-injection (prevalence rate) and at 1 week in the IA and EA groups were 43% vs 42%, and 22% vs 17%, respectively using ≥50% pain relief as the threshold, and 35% vs 29% and 12% vs 6%, respectively, using ≥75% relief. One limitation in studies evaluating the effectiveness of EA injections is that there is little standardization on selection, location of injections and the injectate composition.

Summary:

$\bullet$ There is slightly stronger evidence for the efficacy of extra-articular (periarticular) than intra-articular SIJ injections with corticosteroid for at least one month in patients clinically diagnosed with SIJ complex pain; Grade C evidence, low level of certainty.

$\bullet$ There is conflicting evidence that extra-articular injections may be more effective at three months; Grade C evidence, low level of certainty.

Q8. Is there an optimal technique for performing intra-articular sacroiliac joint injections and sacral lateral branch blocks?

Anatomical considerations

When determining the optimal technique for performing IA and EA SIJ injections as well as SLBBs, it is important to consider the anatomy relevant to each procedure. The SIJ, a combination diarthrodial and amphiarthrodial joint, typically consists of the sacral segments S1 (largest component), S2, S3, and the ilum.$^{132}$ The SIJ consists of both IA and EA portions. The IA and EA portions consist of a synovial joint

(infero-anteriorly) and syndesmosis (ie, fibrous linkage), respectively.

The synovial portion of the SIJ typically involves the inferior half of the joint with the articular cartilage on the sacral side being thicker in comparison to that of the iliac side (~3-5mm vs ~1-2mm).$^{170}$ Multiple ligaments support the SIJ including ventral, dorsal, transverse and interosseous ligaments. The interosseous ligaments are also supported by the short and long dorsal sacroiliac ligaments.$^{132,171}$ The volume of the joint has been estimated to range from 0.5-2.5 mL.$^{41,117,118,130,133}$

The innervation to the SIJ and surrounding structures (posterior EA elements) is complex, and our understanding continues to evolve.$^{113,132,171,172}$ The IA portion of the SIJ is mainly innervated by anterior sources with a small contribution from posterior sources.$^{171}$ Both myelinated and unmyelinated fibers exist within the articulation of the SIJ.$^{3,109}$ The anterior aspect of the joint is innervated by branches from the lumbosacral trunk, as well as direct branches from the anterior ramus of L4 and the anterior ramus of L5.$^{171}$ Dissection and histology has not confirmed innervation from the anterior rami of S1-53 and sacral plexus.$^{109,171}$ The posterior aspect of the SIJ capsule and posterior EA ligaments are innervated by the posterior sacral network (PSN), which predominantly consists of lateral branches of the posterior rami from S1 to S3.$^{171}$ Figure 1 demonstrates the location of the posterior interosseous ligaments and sacral lateral branches. The L4, L5, and S4 lateral branches inconsistently innervate the posterior SIJ complex. One study demonstrated contributions from S1 and S2 in 100%, S3 in 88%, L5 in 8% and S4 in 4% of specimens, though other studies have reported greater contributions from L5 and S4.$^{108,112,113,115,152}$ The lateral branches are small with mean diameters <1 mm (range 0.21-1.51 mm).$^{113}$ When the L5 lateral branch is involved in the innervation it may anastomose with the lateral branch of S1 or innervate the posterior aspect of the joint directly.$^{108,113}$ The lateral branches emerge from the foramina, with the number of lateral branches from each foramen ranging from one to four and most levels having one to two.$^{108,171}$

Defining optimal technique of each injection procedure depends on its purpose (diagnostic, prognostic, and/or therapeutic) and the relevant structures to be targeted. The following text will focus on diagnostic and/or therapeutic IA SIJ injection and diagnostic/prognostic SLBBs. Optimal technique for therapeutic dorsal ligament complex injection will not be discussed, as its therapeutic value has not yet been well defined (addressed in other sections of this article). Similarly, diagnostic injections of the innervation of the anterior aspect of the SIJ complex will not be discussed here, as viable technique(s) have yet to be validated.

Intra-articular sacroillac joint injections

Multiple techniques have been described for IA SIJ injections. For all IA SIJ injections, image guidance is mandatory to assist with accurate needle placement, prevent deposition of the diagnostic and/or therapeutic injectate into vasculature, and to determine if there is IA and/or EA spread. Placement without image guidance results in a high technical failure rate ranging from 78%-100%.$^{41}$ Although ultrasound guidance can be used for placement, the use of this imaging modality may be associated with lower accuracy rates in some contexts and long learning curves for IA placement.$^{154,157,158,173}$

Pain Medicine, 2025, Vol. 00, No. 0

Therefore, this section will focus on fluoroscopy-guided injections.

For fluoroscopy-guided injections, multiple technical aspects need to be considered including the location of SIJ access, use of contrast medium, and volume of therapeutic and/or therapeutic injectate. First, injection of contrast medium is recommended to delineate the eventual spread of the diagnostic and/or therapeutic injectate and to assist with target specificity. Contrast injection allows for the determination of intravascular spread, EA spread into and onto surrounding structures and the existence of ventral and dorsal capsule tears.$^{8}$ SIJ contrast flow patterns may have predictive value for short-term outcomes, although their utility for long-term prognosis appears limited. Scholten et al.$^{174}$ conducted a retrospective review of 54 subjects who underwent therapeutic SIJ corticosteroid injections, with 49 patients included in the final analysis. Their findings revealed that arthrographic contrast patterns demonstrating injectate delivery to the superior portion of the SIJ were associated with greater pain reduction at 2 weeks post-injection. However, this differential effect was not observed at the 8-week follow-up when compared to subjects whose contrast did not reach the superior portion of the SIJ.

A second variable to consider when performing IA SIJ injections is the volume of injectate. When performing SIJ injections into the inferior portion of the joint, the capacity is <2.7 mL with capsular distention occurring at volumes ≤1.6 mL^8,130,133 Therefore, it is recommended that volume of injections be <2.5 mL including the volume of contrast deposited prior to any diagnostic and/or therapeutic injectate. This is addressed in additional detail within Question 5 (What volume should be used for SIJ injections? Should diagnostic injections include corticosteroid?) of this article.

Traditionally, IA SIJ injections are performed by accessing the lower portion of the joint since the synovial portion of the SIJ involves the inferior half of the joint. However, successfully entering the joint at times can be challenging.$^{8}$ Successfully gaining IA SIJ access may be more difficult in male patients.$^{177}$ Injections may also be more difficult in individuals who demonstrate osteophytes and the presence of substantial vacuum phenomena on preprocedural CT scans.$^{177}$ In general, review of pre-procedural axial imaging may provide insight regarding the most feasible angle of approach for obtaining access to the SIJ. In individuals for whom access is challenging (eg, obesity or severe joint narrowing), the use of CT-guidance may improve technical success and possibly outcomes.$^{162}$ If entering the lower portion of the joint is challenging with fluoroscopy and alternatives are not available, other points of access in the middle and superior portions of the SIJ have been suggested.$^{175,176,178}$ In an observational study of 69 consecutive patients in whom the middle portion of 100 separate SIJs was targeted, the success rate was 80%.$^{176}$ Notably, in three of the 20 cases of failed SIJ access using the middle part of the SIJ, the traditional inferior SIJ access technique was successful. Conversely, in four patients, attempts to access the inferior aspect of the joint were unsuccessful, with successful injections achieved by accessing the middle portion of the joint. This suggests the need for adaptability. In a prospective randomized noninferiority study examining 181 patients who were randomized to undergo injection using a point of SIJ access at either the upper or lower one-third of the joint, the IA injection success rates were 93.3% and 95.6%,

respectively.$^{175}$ The authors concluded that accessing the upper one-third of the SIJ resulted in a similar success rate compared to the traditional point of access at the lower one-third of the SIJ.$^{175}$

Sacral lateral branch blocks

SLBBs are used to select patients for RFA techniques that target the posterior sacral network (PSN) associated with EA SIJ complex pain including the short posterior sacroiliac ligament (SPSL), long posterior sacroiliac ligament (LPSL), interosseous sacroiliac ligament (ISL), and superficial lamina of the sacrotu-berous ligament (STL).$^{171}$ Both single-site, single-depth SLBBs and multi-site, multi-depth SLBBs have been examined for their ability to effectively ashetestize the posterior SIJ complex.$^{117}$,$^{118}$In a randomized, controlled, double-blinded study, single-site, single depth SLBBs were found to be physiologically ineffective on a consistent basis with 60% of volunteers in the active group maintaining sensation with both repeated SIJ capsular disten- tion as well as probing of the dorsal ligaments of the SIJ com-plex.$^{118}$ Furthermore, in a parallel anatomic study examining S1 and S2 lateral branch injections with green dye, only 36% of lateral branches were successfully captured using the single-site, single-depth technique.$^{118}$ Based on this study, it was deter- mined that single-site, single-depth SLBBs are associated with a high false-negative rate and may fail to identify patients with pain arising from portions of the posterior SIJ complex inner- vated by the PSN. In a follow-up, double-blind, placebo-controlled study performed in 20 asymptomatic volunteers, multi-site, multi-depth SLBBs produced a much lower false- negative rate in comparison to the single-site, single-depth SLBBs.$^{117}$ For the multi-site, multi-depth SLBB technique described by Dreyfuss et al.,$^{117}$ needles were placed 8-10 mm lateral to the posterior sacral foramina at multiple locations based on clock face positions around the S1 and S2 foramina (Right: 2:30, 4:00, and 5:30; Left: 9:30, 8:00, and 6:30) and S3 foramina (Right: 2:30, 4:00; Left: 9:30 and 8:00). After ruling out vascular needle tip position with injection of contrast medium at each target area, 0.2 mL of local anesthetic was injected. A second injection at each target area was then performed after withdrawing the needle by one bevel length, for a total volume of 1.2 mL at the S1 and S2 foramina, and 0.8 at S3.

foramina (Right: 2:30, 4:00; Left: 9:30 and 8:00). After ruling out vascular needle tip position with injection of contrast medium at each target area, 0.2 mL of local anesthetic was injected. A second injection at each target area was then performed after withdrawing the needle by one bevel length, for a total volume of 1.2 mL at the S1 and S2 foramina, and 0.8 at S3.

When using this multi-site, multi-depth SLBB protocol, 70% of the active group did not respond to posterior ligamentous probing after receiving injections compared to 0-10% of the sham group. Furthermore, a parallel cadaver study revealed a 91% accuracy rate for multi-site, multi-depth SLBBs to capture the S1 to S3 lateral branch nerves. Notably, in the human subject portion of the study, 86% of participants retained the ability to sense capsular distention; thus, the ventral articular portion of the SIJ is not effectively anesthetized by blocking the PSN.$^{117}$ One study has validated an ultrasound-guided technique for SLBBs using the Dreyfuss et al. multi-site, multi-depth SLBB protocol.$^{179}$ In this randomized-controlled trial, the investigators probed the dorsal aspect of the sacroiliac ligament, the dorsal aspect of the SIJ, and the SIJ interosseous ligament both before and after implementing two different SLBB protocols in order to establish the ability of participants to perceive pain in these regions and then to evaluate the degree to which pain to probing was reduced or completely eliminated by each of the two SLBB protocols. The investigators reported a similar degree of ability to block nociception caused by probing these three SIJ

Pain Medicine, 2025, Vol. 00, No. 0

complex structures when using the previously validated fluoroscopically-guided multi-site, multi-depth SLBB protocol compared to their ultrasound-guided “triple injection” technique. This technique consisted of injecting 0.5 mL of 2% lidocaine at target points along the sacral lateral crest lateral to the S1 foramen, between the S1 and S2 foramina, and between the S2 and S3 foramina.

In conclusion, fluoroscopically-guided multi-site, multi-depth SLBBs (using the Dreyfuss et al. protocol) effectively target posterior EA SIJ structures with lower false-negative rates than single-site, single-depth SLBBs. In addition, ultrasound-guided multisite blocks along the sacral lateral crest (according to the Finlayson et al. protocol) effectively target posterior EA SIJ structures to a comparable degree according to one randomized controlled trial. Therefore, if one is contemplating an RFA procedure to interrupt the PSN, multi-site, multi-depth SLBBs or ultrasound-guided multisite blocks along the sacral lateral crest (according to the Finlayson et al. protocol) should be favored as a prognostic tool. However, it should be noted that sham-controlled studies that did not to use prognostic SLBBs have yielded positive RFA outcomes,$^{21,27}$ while controlled studies that used SLBBs have yielded negative or equival-ocal results.$^{16}$ In addition, if a patient has relief with an IA SIJ injection, it may not predict the ability of an RFA technique that targets the PSN to effectively mitigate pain since these RFA techniques do not target IA sources of pain as demonstrated in the study by Dreyfuss et al.$^{117}$

Summary:

$\bullet$ When performing IA SIJ injections, the preferred method includes fluoroscopic guidance, access to the inferior (synovial) portion of the SIJ, use of contrast medium to confirm appropriate spread of subsequent diagnostic and/or therapeutic injectate (if no allergy/contraindication), and a low injectate volume (<2.5 mL); Grade B recommendation, moderate level of certainty.

* If IA access is not possible using the inferior aspect of the joint, consideration can be given to accessing the middle or superior aspects of the joint or, if available, using CT guidance; Grade C recommendation, low-to-moderate level of certainty.

$\bullet$ When considering radiofrequency ablation (RFA) to interrupt the sacral lateral branch nerves (posterior sacral network) for treating posterior sacroiliac joint complex pain, fluoroscopic guidance should be used to perform multi-site, multi-depth sacral lateral branch blocks (SLBBs) targeting the lateral branches of S1 through S3; Grade C-B recommendation, moderate level of certainty.

$\bullet$ Ultrasound-guided multisite blocks along the sacral lateral crest (according to the Finlayson et al. protocol) targets posterior extra-articular SIJ structures and may be considered for prognostic SLBBs; Grade C recommendation, low level of certainty.

Q9. Should sedation be used for intra-articular sacroilac joint injections? Should sedation be used for sacral lateral branch blocks? Should sedation be used for sacral lateral branch radiofrequency ablation?

The use of sedation for interventional pain procedures is controversial. The debate arises secondary to whether providing

analgesic/anxiolytic medications can alter the patient's "true" response to the intervention. Accurate or "true" responses from the patient are critical to render a diagnosis or predict (prognosticate) outcome for diagnostic injections. In the context of this article, "sedation" will be referred to as the provision of a sedative and/or analgesic medication (eg, benzodiazepine and/or opioid, respectively) as opposed to the patient's physiologic response.

The use of sedation during interventional pain procedures is not uncommon. The use of sedation has been found to vary depending on the procedure being performed. A survey study by Kohan et al. found that sedation was used in 80% of RFA procedures, 66% of sympathetic nerve blocks, 54% of epidural corticosteroid injections, 50% of medial branch nerve blocks, and 42% of the time for SIJ injections. Benzodiazepines and opioids were the most commonly used sedatives.$^{180}$ An international survey study found that only 29% of physicians "never" use sedation for SIJ injections.$^{181}$

Whereas the use of sedation can help decrease patient-related discomfort, decrease patient movement, possibly improve patient satisfaction, decrease the incidence of vaso- vagal reactions,$^{182,183}$ and possibly decrease the risk of false- negative responses,$^{184-186}$ it may increase the risk of false- positives responses, thereby negatively impacting the diagnost- tic and/or prognostic nature of the procedure.$^{44,187}$

Improved patient satisfaction with the use of sedation has not been firmly established. Kubulus et al. found improved patient satisfaction with the use of sedation during regional anesthetic injections; however, this study was limited by its retrospective nature and confounding anxiety that may have been experienced secondary to the surgical procedure.$^{188}$ Other studies report conflicting results. Cucuzella et al. found that 17% of patients requested intravenous diazepam prior to an initial epidural corticosteroid injection, with 28% requesting it again for subsequent procedures. Higher pain scores, increased anxiety, and depression were associated with requests for repeat sedation, which may explain the paradoxical observation that the use of sedation was associated with increased procedure-related pain.$^{184}$ A follow-up study by the same group of authors found that even after an extensive discussion regarding sedation, 58% of patients chose either anxiolysis (oral diazepam) or sedation (IV diazepam) prior to an injection and 93% of those who chose to receive no anxiolysis or sedation were satisfied with their decision.$^{189}$ In a retrospective study performed in 4432 patients who underwent 6878 transaminal epidural corticosteroid injections, Diehn et al. found high levels of satisfaction among non-sedated patients, with 96% reporting good, very good, or excellent satisfaction levels.$^{190}$ Similar to Cucuzella et al.$^{184}$ who reported higher pain scores in sedated patients, the authors reported lower satisfaction scores in patients who received sedation, though this observation may represent selection bias (ie, patients with increased anxiety and pain sensitivity may be more likely to elect to receive sedation) and the small number of patients who received sedation prevented a meaningful statistical comparison.$^{190}$ Analgesia from opioids may decrease pain, making it difficult for the clinician to determine if post-procedural pain relief was related to opioid administration or the intervention itself. Benzodiazepines cause skeletal muscle relaxation and thus may decrease back pain in patients with a myofascial component.$^{191}$⁻$^{193}$ Additionally, benzodiazepines and opioids may alleviate the affective (anxiety) component of pain and alter

Pain Medicine, 2025, Vol. 00, No. 0

post-procedural activity, thereby undermining the accuracy of the post-procedural analgesic response.

Effect of sedation during diagnostic or prognostic procedures on treatment outcomes: Indirect evidence

Several studies have evaluated the effect sedation during prognostic blocks has on neurolytic or RFA outcomes, albeit none for sacral lateral branch RFA. Although a small retrospective study found poorer neurolysis outcomes when sedation was used during prognostic celiac plexus blocks, a larger study (n = 265) evaluating the effect of sedation during genicular nerve blocks on RFA results reported no effect on treatment outcomes.$^{194,195}$ A randomized trial evaluating methoxyflurane on pain during and after genicular nerve blocks reported rapid recovery for the short-acting inhaled medication used worldwide for out-of-hospital posttraumatic analgesia. The authors recommended consideration of its use for other prognostic procedures based on the tenet that a block's predictive value will be higher when the local anesthetic effect outlasts the duration of sedation.$^{196}$

Guidelines

Societies, including the American Society of Anesthesiologists (ASA)$^{197}$ and the International Pain and Spine Intervention Society (IPSIS),$^{198}$ recommend against the routine use of sedation for most interventional pain procedures except for advanced procedures (vertebral augmentation, spinal cord stimulation, etc.). Studies report an increased risk of overall complications and serious neurological injuries with the use of moderate or heavy sedation.$^{199-201}$ A recently proposed local coverage determination (LCD) asserts that the use of moderate or deep sedation, general anesthesia, or monitored anesthesia care (MAC) is not indicated for SIJ injections and is not considered medically reasonable or necessary. The LCD does note that use of anxiolytics may be indicated in patients with anxiety disorders and needle phobias.$^{202}$ Additionally, the LCD panel noted that sedation not only increases risk but may also compromise the validity of the diagnostic procedure.$^{120}$

Diagnostic sacroiliac joint injections

Direct Evidence

Cohen et al. examined the impact of sedation with midazolam and fentanyl on the diagnostic validity of SIJ injections and sympathetic blocks in a randomized, controlled crossover trial. The use of sedation was found to result in lower pain diary scores and a higher proportion of positive diagnostic injections than for injections performed without sedation, suggesting a possible increase in false-positive injections. Although procedure-related pain scores were lower in injections performed with sedation, sedation did not improve patient satisfaction or 1-month treatment outcomes.$^{44}$

Therapeutic sacroiliac joint injections

Direct evidence

No studies were identified.

Indirect evidence: sedation for pain-relieving procedures

Dreyfuss et al. found no difference in post procedure pain scores assessed immediately after a therapeutic epidural corticosteroid injection without local anesthetic between patients who had received sedation with midazolam and/or fentanyl and those who did not.$^{203}$

Sacral lateral branch blocks

Direct evidence

No studies were identified.

Indirect evidence: sedation prior to medial branch nerve injections

Manchikanti et al. published several studies investigating the impact of sedation on the diagnostic validity of medial branch blocks. The authors suggested that although saline, midazolam, or fentanyl can all create false positives, the likelihood was low when 80% pain relief was used as the cutoff threshold for a positive block rather than 50%. However, these studies had many methodological limitations, including only assessing patients pre- and post-medication administration but before the actual procedure was performed (ie, failure to assess post-injection pain relief), that these procedures were being done therapeutically and not diagnostically, and a high rate of chronic opioid therapy in subjects.$^{204}$

Sacral lateral branch radiofrequency ablation

Although the use of sedation is rarely indicated with SIJ injections, it can be indicated during SLBRFA according to the subject matter expert (SME) panel in the recent LCD. In addition to multiple lesions at multiple levels, "finder" needles (if used) are often placed in 3-4 sacral foramina as landmarks for electrode placement, and can elicit radicular pain.$^{120}$ Clinical trials are mixed regarding whether they specifically noted that patients routinely received sedation, though no trial reporting prohibiting it.

Summary:

$\bullet$ For diagnostic SIJ injections and prognostic sacral lateral branch injections, the routine use of sedation is not recommended as it may result in a higher proportion of false-positive diagnostic injections; Grade B recommendation, low-to-moderate level of certainty.

* For therapeutic SIJ injections, the routine use of sedation is not recommended; occasionally, light or moderate sedation by properly trained personnel, in extenuating circumstances such as with severe needle phobias, other psychiatric indications, a history of severe vasovagal reactions, or movement disorders that preclude a safe and/or accurate injection, may be necessary; Grade C recommendation, low level of certainty.

* For SLBRFA, the use of light or occasionally moderate sedation with patients remaining responsive may be indicated in certain cases and should depend on patient and provider preferences; Grade C recommendation, low level of certainty.

* If sedation is used, the lowest dose of short-acting medications should be administered; Grade A recommendation, high level of certainty.

Non-surgical treatment

Q10. How effective are non-invasive treatments for sacroiliac joint complex pain?

Non-operative (conservative) treatments for sacroiliac joint pain

Non-operative or conservative management of SIJ complex pain commonly includes a trial of analgesic and anti-inflammatory medications, physical therapy (PT) modalities,

Pain Medicine, 2025, Vol. 00, No. 0

bracing, psychological therapies, acupuncture, and various other treatments. Although supported by little evidence, these treatments are frequently recommended before consideration of operative or procedural treatments. Despite the lack of medium or high-quality evidence, conservative treatments are recommended for the treatment of SIJ complex pain by the Centers for Medicare and Medicaid Services$^{202}$ and commercial payors in the United States.$^{205}$ The Centers for Medicare and Medicaid Services require 4 weeks of conservative care prior to performing procedures on the SIJ complex. There is disagreement among professional guidelines, with some expert medical consensus recommending conservative care based on low-quality evidence$^{206}$,$^{207}$ and others recommending against a requirement$^{208}$ due to the lack of moderate or strong evidence of efficacy. Many clinical studies evaluating IA SIJ injections, SLBRFA, and SIJ fusion have required a course of conservative treatment,$^{11,16,21,22,24}$ while others have not.$^{27,146}$ The rationale behind the de facto use of conservative management of SIJ complex pain is primarily based on the presumption that SIJ complex pain, like many musculoskeletal conditions, will improve on its own with time, and low-risk conservative measures may facilitate this. However, the typical clinical course of SIJ complex pain in the absence of formal treatment has not been reported in the literature. The rationale for a mandatory trial with conservative therapy may be a more reasonable assumption for acute than chronic SIJ complex pain in which symptoms have persisted despite time and self-care, and conservative treatments can pose a financial burden. In an RCT by Cohen et al.,$^{14}$ the control group who received a cluster of conservative measures had a poor response rate, with only 11% having a positive outcome with three months of therapy.

It must also be acknowledged that the majority of studies evaluating the effectiveness of various non-invasive treatments for sacroiliac joint complex pain did not use SIJ injections and/or SLBBs to confirm the diagnosis of SIJ complex pain. As such, a proportion of the individuals in these studies likely did not have SIJ pain given the known limitations of history, physical examination, and imaging for identifying a painful SIJ complex.

Pharmacological management

Direct evidence

The chronic inflammatory rheumatic musculoskeletal disease axial spondylarthritis, which usually involves the SIJ, is responsive to NSAIDS and disease-modifying anti-inflammary medications including tumor necrosis factor inhibitors, interleukin-17 inhibitors, and janus kinase inhibitors.$^{209}$ There are no RCTs examining the efficacy of oral or topical non-steroidal anti-inflammatory drugs (NSAIDS), opioids, or SNRs1s such as duloxetine in SIJ complex pain of non-rheumatic etiology.

Indirect evidence

There are numerous randomized trials that demonstrate the efficacy of muscle relaxants, NSAIDS, and antidepressants for LBP. Oral NSAIDs have been shown to improve pain and disability in patients with back pain in the short-term, but they are associated with significant adverse effects on the gastrointestinal, renal, and cardiovascular systems, and their unmonitored long-term use is not recommended.$^{210,211}$ COX-21 inhibitors have been shown to be effective in patients with chronic LBP and are associated with fewer GI side

effects; however, they are often more expensive than traditional NSAIDs and are still associated with adverse cardiovascular and renal effects with chronic use. A Cochrane review found no significant difference in effectiveness between non-selective NSAIDs and COX-2 inhibitors for LBP.$^{210}$ Acetaminophen has not been found to provide clinically meaningful relief for back pain$^{212}$ and is inferior to COX-2 inhibitors in head-to-head trials.$^{213}$ There is strong evidence for serotonin-norepinephrine reuptake inhibitors for back pain,$^{211,212}$ and guidelines by the American College of Physicians recommend duloxetine as a second-line treatment.$^{207}$ In a recent systematic review evaluating antidepressants for back pain and osteoarthritis, tricyclic antidepressants (3 trials, 145 patients) were not found to provide meaningful relief at 2 weeks or less based on low-quality evidence, and neither were other non-SNRI antidepressants.$^{214}$ In another systematic review, gabapentinoids were not found to be effective for the treatment of LBP.$^{215}$ Topical capsaicin has been shown to be beneficial in patients with LBP though penetration into deep tissues is limited,$^{216}$ and topical diclofenac is effective but less so than oral NSAIDs.$^{217}$ Opioid analgesics produce a decrease in the perception of pain, increased pain tolerance, and decreased discomfort after acute post-surgical injury. However, despite the lack of supporting evidence for opioid efficacy in the treatment of clinical acute musculoskeletal pain of the neck or low back,$^{218}$ these remain commonly prescribed medications.$^{219,220}$ Additionally, opioids are associated with significant short and long-term adverse effects such as feelings of intoxication, cognitive/motor dysfunction, constipation, hormone dysregulation, physiological and psychological dependence, and persistent use can lead to hyperalgesia, opioid misuse and/or opioid use disorder.$^{218,221}$

Sacroiliac joint belts

The purpose of an SIJ belt is to provide compression and stability of the pelvis across the SIJ complex, thereby reducing pain and disability. Vleeming et al.$^{222}$ tested this hypothesis by applying SIJ belts in cadavers, revealing a 30% reduction in rotation of the joint. In a prospective survey study, Hammer et al.$^{223}$ compared outcomes in 17 patients with injection-confirmed SIJ complex pain and 17 healthy control patients who received SIJ belts. After 6 weeks of use, the authors reported significant improvements in quality of life in the SIJ belt patients, a small reduction in pain, improved gait and cadence, but no significant changes in muscle activity except for decreased tone in the rectus femoris muscle.

Chiropractic treatment and physical therapy (including structured exercise therapy)

Early uncontrolled and low-quality studies suggested that PT, and osteopathic or chiropractic manipulation reduce pain and improve mobility modestly in patients with SIJ complex pain resulting from altered gait mechanics and spine mala-lignment.$^{224,225}$ A later systematic review by Al-Subahi et al.$^{226}$ investigated how effective PT techniques were for the reduction of pain in patients with SIJ complex pain. Only one of the nine identified studies included a placebo control group. The systematic review examined three types of PT interventions for reducing pain and improving disability: Exercise, manipulation or manual therapy (MT), and kinesio-taping (KT). The authors found the included studies were of low-quality, but that MT was nevertheless most

Pain Medicine, 2025, Vol. 00, No. 0

effective in terms of improving pain and functional capacity. In addition to reducing pain and improving disability, KT was effective in restoring normal pelvic symmetry and position in an uncontrolled study, but ineffective when compared to placebo. A more recent systematic review and meta-analysis by Trager et al.$^{227}$ examined the effectiveness of MT for SIJ complex pain. This review included 16 RCTs with a total of 421 adult patients. The results did not show a reduction in pain but did show a small, significant effect in the reduction of disability. The meta-analysis showed that when non-manual physiotherapy (exercise therapy), regardless of whether it included passive modalities, was compared to sham in patients with SIJ dysfunction, there was no reduction in pain. A systematic review of RCTs by Macedo et al.$^{228}$ found that a motor control exercise program was more effective than spinal manipulative therapy and home exercise for patients with chronic LBP. A more recent narrative review by Prather et al.$^{229}$ concluded that therapeutic exercise alone or combined with manipulation was successful in decreasing SIJ complex pain and accompanying dysfunction. The indirect evidence for psychologically-informed PT was examined in a very large clinical trial (n = 230, 77 sites) that included patients with acute LBP at medium and high risk for developing chronic LBP. The authors found that intensive psychologically-informed PT was no better than usual care.$^{230}$

Acupuncture

No placebo-controlled clinical trials examining the efficacy of acupuncture on SIJ complex-related pain have been published to date. A systematic review and meta-analysis by Liu et al.$^{231}$ examined acupuncture for SIJ complex pain attributed to malposition in 10 studies (n = 1019). The quality of the RCTs was deemed to be extremely low with little to no follow-up and "random" intervention techniques. However, indirect evidence from studies of LBP supports the use of needling, including traditional acupuncture and placebo acupuncture, for short-term relief of pain. One study examined 2065 patients with chronic LBP (3 to 12 months) and found that after 3 weeks of twice weekly and 1 week of once weekly treatment, any form of acupuncture or placebo acupuncture was more effective than usual care at 8 weeks, with persistence to one year.$^{232}$ A similar study performed in patients with acute LBP found that acupuncture and sham acupuncture produced modestly more pain relief than no treatment 1 week after the 2-week treatment.$^{233}$ A recent Cochrane review of 33 trials found that acupuncture was no better than sham and may be no better than usual care for chronic LBP.$^{234}$

Psychological interventions (cognitive behavioral therapy, physical therapy-delivered cognitive behavioral therapy)

No studies have evaluated psychological interventions for SIJ complex pain, so results must be extrapolated from LBP studies. A systematic review found no effect for cognitive behavioral therapy (CBT) compared to no treatment but a substantial effect when CBT was combined with structured exercise.$^{235}$ However, a very large RCT found no benefit for CBT delivered by a physical therapist in combination with manual intensive PT for transition from acute to chronic LBP.$^{230}$ For behavioral therapy and counseling, no significant benefits were found either as stand-alone treatments or combined with structured exercise.$^{235}$ In a narrative review, Cohen et al.$^{65}$ found indirect evidence to support the

influence of psychosocial pathology on the development of SIJ complex pain and concluded CBT may be beneficial in those with comorbid psychological conditions.

Summary:

$\bullet$ There is indirect evidence that physical therapy (PT) and structured exercise may provide intermediate-term benefit for SIJ complex pain; Grade C evidence, low level of certainty.

$\bullet$ There is indirect evidence that cognitive behavioral therapy (CBT) may provide benefit for patients with comor- bid psychological conditions; Grade C evidence, low level of certainty.

There is indirect evidence that PT combined with CBT does not reduce the transition from acute to chronic low back pain; Grade B evidence, low-to-moderate level of certainty.

* There is evidence that SIJ belts may provide benefit in some patients; Grade C evidence, low level of certainty.

* There is insufficient evidence that chiropractic treatment is effective for SIJ complex pain; Grade I evidence.

* There is indirect evidence supporting the use of NSAIDs and SNRIs, but not other antidepressants, for SIJ complex pain; Grade C evidence, low level of certainty.

* There is insufficient evidence supporting acute use of opioids for SIJ complex pain; Grade I evidence.

* There is indirect evidence suggesting no benefit for the use of long-term opioid therapy; Grade D evidence, low level of certainty.

* There is indirect evidence that acetaminophen is ineffic-tive for SIJ complex pain; Grade D evidence, low level of certainty.

Q11. What is the evidence for non-corticosteroid injections in the treatment of sacroiliac joint complex pain?

Corticosteroids are the most used injectable treatment for SIJ complex pain. Whereas corticosteroid injections are an effective treatment for SIJ complex pain, relief may be short-lived for some individuals.$^{9,10,139}$ Unwanted systemic and local side effects of corticosteroid use are well described.$^{41,236-239}$ As such, physicians and scientists have proposed alternative treatments to overcome the relative limitations of corticosteroid use. Numerous other potentially therapeutic injectable agents have been proposed, including autologous orthobiologics, allogeneic orthobiologics, hyaluronic acid, protherapy, chemoneuloysis, botulinum toxin, and saline. These proposed therapeutic agents can be classified according to the tissue being targeted for injection (articular, ligamentous, and neuronal).

Regenerative agents may contain properties that promote repair of dense fibrous tissues such as joint capsules, ligaments, and fascia. These injectates fall into one of two categories: (1) naturally occurring bioactive agents [eg, platelet rich plasma (PRP)] and/or extracellular structural components (eg, hyaluronic acid); and (2) irritating or debriding solutions that initiate endogenous repair mechanisms (eg, protherapy, saline). IA injections are theorized to relieve pain through mitigation or reversal of degenerative changes to the articular cartilage of the SIJ. Literature reports of IA injectates that have been investigated as potential treatments for

Pain Medicine, 2025, Vol. 00, No. 0

SIJ complex pain include autologous and allogenic orthobiologics (specifically platelet-rich plasma,$^{240-242}$ platelet-rich fibrin,$^{243}$ Wharton's jelly placental allograft$^{244}$; and hyaluronic acid$^{245,246}$). Dextrose-based protherapy and saline injections into and surrounding the SIJ have also been reported.$^{247-249}$ Whereas vigorous lavage with saline may provide a "washout" of inflammatory and pain-mediating molecules from the interior joint space, this is unlikely to occur with a single IA injection; however, introduction of fluid within the joint may indirectly stimulate an anti-inflammatory cascade if proteoglycans are removed from the superficial extracellular matrix of articular cartilage.$^{250}$ The rationale for EA injection of regenerative agents into structures of the posterior SIJ complex is to relieve pain by strengthening or "tightening" damaged or lax ligaments in order to improve stability. Hypertonic dextrose protherapy purportedly induces cell proliferation at ligamentous attachment sites.$^{251}$ There is also an existing report of EA injection of PRP into the posterior SIJ complex.$^{252}$

The second class of therapeutic injectates target neural tissues with the goal of relieving pain by interrupting or altering nociceptive signal transmission from the SIJ complex. Published literature includes two types of injectable therapies to potentially treat SIJ complex pain: Phenol-based chemoneurolysis and botulinum toxin A. Phenol is a non-specific protein coagulant; aqueous solutions have previously been injected in both the IA and EA spaces for the purpose of neuronal coagulation.$^{253-255}$ Currently, there is only one publication describing the use of botulinum toxin, which was injected into the EA structures of the posterior SIJ complex.$^{256}$ Although best known as a paralytic agent that interferes with synaptic signaling at the neuromuscular junction, there is evidence that botulinum toxin A acts on sensory neurons as well.$^{257,258}$ Proposed mechanisms include reduction of neurogenic inflammation, alterations of efferent and afferent signals in spinal nerves leading to decreased central sensitization and inhibition of the release of sensory neuropeptides, as well as inhibiting protein translocation from the endosomal compartment to the cell plasma membrane of transient receptor potential vanilloid 1 (TRPV1) and Transient Receptor Potential Ankyrin 1 (TRPA1) receptors.$^{256}$

Herein, we describe the published evidence related to effectiveness within each category of potentially therapeutic injectable agents. Notably, no sham-controlled (explanatory) RCTs were identified addressing the efficacy of any non-steroid injectables to treat pain specifically arising from the SIJ complex.

Clinical studies

Autologous orthobiologics

Pragmatic RCTs

In 2022, Chen et al. published results from a randomized study in which 26 participants were treated with a single IA injection of PRP or corticosteroid.$^{240}$ Among other criteria (Table 13), patients were selected based on ≥80% relief of index pain following a single IA SIJ injection of local anesthetic. Fifteen and 11 patients were then allocated to PRP and corticosteroid treatment, respectively. Pain and function improved significantly from baseline for both groups at 1-, 3-, and 6-months post-injection. However, patients in the corticosteroid group reported significantly lower pain scores than

those in the PRP group at all follow-up timepoints, as well as significantly lower ODI scores at 1 and 3 months. Additionally, responder rates for ≥50% pain reduction were significantly higher for the corticosteroid group compared to the PRP group at 1 and 3 months (80% vs 21% and 70% vs 21%, respectively), though the intergroup difference did not reach statistical significance at 6 months (67% vs 25%).

A similar randomized comparative trial was reported on by Singla et al. in 2017.$^{242}$ The investigators randomized 40 patients with three or more positive SIJ provocation tests (no diagnostic injection for selection) to treatment with a single IA injection of either PRP ($n=20$) or corticosteroid ($n=20$). Contrary to the findings of Chen et al., PRP was associated with superior short-term outcomes compared to corticosteroid. Median pain scores were significantly lower at 6- and 12-weeks post-injection in patients treated with PRP vs corticosteroid, and responder rates for ≥50% pain improvement were significantly higher in the PRP group compared to the corticosteroid group at 12 weeks (90% vs 25%).

Observational studies

These studies are presented in Table 13.

Allogeneic orthobiologics

Pragmatic RCTs

No pragmatic RCTs were identified.

Observational studies

These studies are presented in Table 13.

Hyaluronic acid

Pragmatic RCTs

In 2022, Abdelkreem et al. published the results of a randomized study comparing short-term outcomes following ultrasound-guided IA injections of either low molecular weight hyaluronic acid (HA) or corticosteroid in 40 participants with sacroiliitis.$^{245}$ Patients were classified as having degenerative (DSI; $n=20$) or inflammatory (ISI; $n=20$) sacroiliitis based on modified New York criteria. No diagnostic SIJ injections were performed as a part of the selection process. Ten patients from each group were allocated to the two treatment arms, producing a total of four patient phenotype/treatment combinations: DSI+HA, DSI+corticosteroid, ISI+HA, and ISI+corticosteroid. Pain was significantly improved from baseline in all four subgroups post-injection, with no between-group differences detected. However, respective comparisons of patient outcomes between treatment modalities for the DSI and ISI groups revealed IA injections of corticosteroid were more effective than HA at reducing SIJ complex pain secondary to sacroiliitis, regardless of etiology. For patients with DSI, average VAS pain scores were significantly lower among those treated with corticosteroids vs HA at all follow-up timepoints (5.6 ± 3.3 vs 8.0 ± 1.3 at 1-week, 4.4 ± 3.0 vs 7.5 ± 1.2 at 2-weeks, and 3.4 ± 2.3 vs 6.1 ± 1.4 at 4-weeks, respectively). Although 1-week VAS pain score comparisons did not reach statistical significance for ISI patients, similar outcomes favoring corticosteroid over HA were observed at 2-weeks (5.3 ± 2.6 vs 7.9 ± 1.3) and 4-weeks (4.4 ± 2.1 vs 6.8 ± 1.5) post-injection.

Observational studies

These studies are presented in Table 13.

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Population|RCT|Initectate|Injectate|Occupational maximum of 3 IA injec-tions at 2.5 mL of mixed-buvipacine|Primary outcome: 2.50% improvement on NRS 6.6% in protherapy group $v$ < 6.3% in protherapy group $v$ < 2.7% in $CS group $v$ < 10.0%|On average, patients in pro proheracy group greater number of injec group ($2.7 \pm 1.5 \pm 0.8$) $P < \langle 0.01 \rangle$)| |---|---|---|---|---|---|---|---| |Dextrose protherapy|48 patients with SJ complex pain lasting 2 50% pain relief following single IA injection with 2.5 0.2% patients required to fail 1 month of conservative nreatment after positive|RCT|(r = 23): 2.5 mL of 25% dextrose solution mixed with levupivacaine|Fluorocopic guidance; maximum of 3 IA injec-tions at 2.5 mL of mixed-buvipacine|6.6% in protherapy group $v$ < 27.2% in $CS group $v$ < 10.8% group $v$ < 10.0% $P < \langle 0.01 \rangle$|On average, patients in pro proheracy group greater number of injec group ($2.7 \pm 1.5 \pm 0.8$) $P < \langle 0.01 \rangle$|Only abstract, tables, and figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English figures are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English images are in English imag|

Table 13. (continued)

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Hoffman & Agnish, 2018$^{247}$|103 US veterans referred for LBP with suspected SIJ involvement based on local tenderness and limited mobility on Gillet test; unspecified number of patients received either a fluoro-guided confirmatory diagnostic injection with LA + CS or were diagnosed with SIJ complex pain based on unspecified response to the first therapeutic protherapy injection|Retrospective cohort study|10 mL of 15% dextrose solution mixed with lidocaine|3 IA injections at approximately 1-month intervals; initially performed with fluoroscopic guidance but switched to no imaging after difficulty accessing fluoroscope|• 24/103 (23%) patients met responder definition of ≥15% ODI improvement • Average decrease in ODI scores from baseline was significant (P < 0.05) but small (mean 4±17 points)|Follow-up was 3-4 months after final injection (median 110 days); authors split patients into “improved” and “not improved” groups based on ≥15% ODI improvement (MCID; study objective was to look for predictors of treatment success)| |Chemical neurolysis Weksler, 2007$^{255}$|42 patients with severe chronic nonmalignant pain lasting ≥6 months who failed conservative treatment; n = 5 SIJ patients; dual diagnostic injection protocol consisting of one LA (injectate and volume not specified) injection requiring 70% pain reduction and one sham injection with 0.9% saline (no pain relief)|Prospective observational study|4% phenol solution|Fluoroscopically guided IA injection|• Treatment was successful (defined as an absolute VAS pain score ≤3) in 4/5 SIJ patients, who received an average of 2.6 injections each (range 2-4) • Mean follow-up was 13.6±5.7 months (range 6-26)|Results presented as aggregate with no breakdown by diagnosis; no info about injection frequency, injectate volume| |Frederico, 2022$^{253}$|2 patients with low back/ hip pain (NRS = 7) and 3 positive clinical tests for SIJ stress; single US-guided EA diagnostic injection (1 mL bupivacaine 0.25%) requiring >75% relief|Case report|0.75 mL of 7% glycerinated phenol|Single series of 3 US-guided injections into posterior sacral network (area of posterior SIJ complex innervated by L4/5, S1-S3) at levels of S1-S3|• Both patients reported ≥50% NRS pain reduction and decreased daily analgesic use through 6 months|| |Ward, 2002$^{254}$|10 patients with sacroiliitis based on unspecified pain relief lasting 2-4 weeks following fluoroscopically-guided IA injection of unspecified volume of bupivacaine 0.5% + 80 mg Depo-Medrol|Retrospective case series|2.5 mL of 6% phenol in saline|Multiple (no quantification provided) fluoroscopically-guided IA injections|• 2/10 patients (20%) reported >70% improvement in symptoms with average duration of 24 weeks • 6/10 patients (60%) had 50-70% improvement for an average of 20 weeks • 1 patient (10%) reported 20-49% improvement lasting 12.5 weeks • 1 patient (10%) reported <20% improvement|Symptom improvement was determined by an unspecified combination of patient interviews and responses on the McGill short-form pain questionnaire • 1 patient (10%) reported 20-49% improvement lasting 12.5 weeks • 1 patient (10%) reported <20% improvement|

44

Pain Medicine, 2025, Vol. 00, No. 0

(continued)

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Platelet-rich plasma (PRP) Chen, $2022^{240}$|26 patients with chronic unilateral LBP without radiculopathy; ≥5 NRS pain score, ≥3 positive SIJ provocation tests, and no improvement of symptoms after ≥4 weeks of physical therapy; SIJ confirmed as pain generator with single fluoroscopically-guided IA diagnostic injection (2 mL lidocaine 2%) requiring >80% pain relief|RCT|PRP group (n = 15): 2 mL PRP CS group (n = 11): 1 mL betamethasone sodium phosphate suspended in acetate (6 mg/mL) + 1 mL lidocaine 2%|Single fluoroscopically-guided IA injection|Responder rates: ≥50% NRS pain reduction • 1 month: 3/14 patients (21.4%) in PRP group vs 8/10 patients (80.0%) in CS group; P = .011 • 3 months: 3/14 patients (21.4%) in PRP group vs 7/10 patients (70.0%) in CS group; P = .035 • 6 months: 3/12 patients (25.0%) in PRP group vs 4/6 patients (66.7%) in CS group; P = 0.141 • Both groups experienced significant NRS pain reduction and disability improvement across follow-up time points; mean NRS pain and disability scores for CS group were significantly lower than those for PRP group at 1, 3, and 6 months|| |Singla, $2017^{242}$|40 patients with LBP lasting ≥3 months; diagnosis based on SIJ pathology on imaging and ≥3 positive provocation tests; no diagnostic injections performed|RCT|PRP group (n = 20): 3 mL leukocyte-free PRP + 0.5 mL $CaCl_{2}$ CS group (n = 20): 1.5 mL methylprednisolone (40 mg/mL) + 1.5 mL lidocaine 2% + 0.5 mL saline|Single US-guided IA injection|• A significantly greater proportion of patients in the PRP group reported ≥50% VAS pain improvement compared to the CS group at 3 months (90% [95% CI: 77-10%] vs 25% [95% CI: 6-44%]) • Pain scores significantly lower in PRP vs CS group at 6 weeks (median [IQR] = 1[1-1] vs 3.5 [2-5]) and 3 months (median [IQR] = 1[1-3] vs 5 [3-5])||

46

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Ko, $2017^{2.52}$|4 patients with SIJ complex pain unresponsive to conservative treatment who received at least one type of therapeutic IA injection (n = 3 sodium morhuate and dextrose prolotherapy; n = 1 botulinum toxin A); no diagnostic injections performed; SIJ complex pain confirmed based on a combination of pt history and clinical examination, unspecified # of positive provocation tests, and imaging|Retrospective case series|10 mL PRP prepared with a citric acid dextrose solution|Series of 2 US-guided EA injections into the posterior SIJ ligaments performed at unspecified time intervals; each session consisted of injecting 3 x 0.5 mL doses into Hackett's trigger points A, B, and C|Pooled data revealed pain was significantly reduced at 1-year post-treatment, with Short form McGill Questionnaire, NRS, and ODI scores reduced by 93%, 88%, and 75%, respectively.Pain improvement remained significant at 4 years (no statistics provided)|No raw data provided| |Navani, $2015^{241}$|10 patients with LBP lasting >6 months unresponsive to conservative treatment and at least 1 positive provocative test; diagnosis based on >50% pain relief following "one or more diagnostic or therapeutic SIJ injections with LA ± corticosteroids"|Retrospective case series|4 mL PRP injection without additives or activating agents|Single, fluoroscopically-guided IA injection|All patients experienced ≥50% pain improvement on VAS pain scale through 12 months|Diagnostic injection details (location, imaging guidance, injectate, volume, etc.) not reported. No raw data or statistical interpretation of findings provided.| |Eldin, $2019^{243}$|186 patients with LBP >6 months unresponsive to conservative treatment; diagnosis of SIJ complex pain based on ≥3 positive provocation tests and a single diagnostic injection (no details were provided regarding location, injectate, pain relief requirements, etc.)|Prospective case control study|PRP group (n = 62): PRP prepared with citrate dextrose and calcium gluconate solutions (injected volume not specified)|Single, fluoroscopically-guided IA injection|No significant between-group differences in average VAS pain scores at baseline or 1-month follow-upAverage VAS pain scores were significantly lower for PRF vs PRP group at 6 months (4.61±2.00 vs 5.19 ±1.49; P = .045)||

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Hyaluronic acid Abdelkreem, $2022^{245}$|40 patients previously diagnosed with either inflammatory (ISI; n = 20) or degenerative (DSI; n = 20) sacroilitis based on modified New York criteria or radiographic findings, respectively. No diagnostic injections.|RCT|CS subgroups (n = 10 ISI; n = 10 DSI): 2 mL betamethasone dipropionate + 2 mL lidocaine 2% Hyaluronic acid subgroups (n = 10 ISI; n = 10 DSI): 2 mL low molecular weight hyaluronic acid + 2 mL lidocaine 2%|US-guided IA injection|• No significant differences in VAS pain scores between ISI vs DSI groups at any time point ISI group: • VAS pain scores were sig- nificantly different before and after injection (P < .001) • Significantly lower VAS pain scores for patients who received CS vs HA at 2 weeks (5.3±2.6 vs 7.9± 1.3; P = .012) and 4 weeks post-injection (4.4±2.1 vs 6.8±1.5, P = .008) DSI group: • VAS pain scores were sig- nificantly different before and after injection (P < .001) • Significantly lower VAS scores for patients who received CS vs HA at 1 week (5.6±3.3 vs 8.0±1.3, P = .044), 2 weeks (4.4± 3.0 vs 7.5±1.2, P = .006), and 4 weeks (3.4±2.3 vs 6.1±1.4, P = .005)|Repeated measures ANOVA used to compare VAS pain scores between different time points; no post-hoc tests performed to ascertain which paired comparisons (baseline vs follow-up) were significant| |Srejic, $1999^{246}$|4 patients with chronic LBP refractory to conservative treatment; SIJ complex pain confirmed by dual fluoroscopically guided IA injections with 2 mL bupivacaine 0.75% (no pain relief threshold established)|Case report|1 mL (8 mg) hyaluronic acid (Hylan GF 20) per joint|3 fluoroscopically- guided IA injections administered at 2-week intervals|3 fluoroscopically- guided IA injections administered at 2-week intervals|• All patients reported pain improvements of 40%-67% 3 months after the last injection|

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Bentulium toxin Lee, $2010^{-26}$|39 patients with back or pelvis pain, 23 positive sign of inflammatory|Prospective case-control study|Botulium toxin (BT) group (n = 20); 100 units of botulinum tox units of botulinum tox-type A (volume not noted)|Single fluorosopically-guided EA injection into the middle portion of the posterior ligaments|• NR5 pain scores were significantly lower at 2- and 3-months|| |Bentulium toxin Lee, $2010^{-26}$|||CS group (n = 19); 40 mg trancinone in 2 mL lidocaine 0.5% in relief||BT group (1.8±1.7 and 1.9±2.0) compared to the 1.9±2.0) compared to the 4.9±1.7, respectively|| |Bentulium toxin Lee, $2010^{-26}$|||||CDI scores were 2. significantly lower at post-treatment for the BT 11.9±9.1) compared to the 11.9±9.1) compared to the 24.0e-10.6), respectively|| |Bentulium toxin|||CS group (n = 19); 40 mg trancinone in 2 mL lidocaine 0.5% in 2 mL lidocaine 0.5% in�c - 2.40% OD1 reduction 2-40% OD1 reduction 3-40% trancinone in 2 mL lidocaine 0.5% in 2 mL lidocaine 0.5% in 2.40% OD1 reduction 4-40% OD1 reduction 5-40% OD1 reduction 6-40% D1 reduction 7-40% OD1 reduction 8-40% OD1 reduction 9-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 reduction 10-40% OD1 redu||||

|Study|Population|Design|Injectate Details|Injection Protocol|Outcomes|Comments| |---|---|---|---|---|---|---| |Placental (Wharton's)|Jelly Allograft|||||| |Lai, $2023^{244}$|38 patients with degenerative structural defects of SIJ articular cartilage based on radiologic imaging who failed >8 weeks of conservative treatment; no diagnostic injections performed (unspecified number of patients had prior diagnostic SIJ injections)|Case series|2 mL (100 mg) Wharton's jelly|Single, fluoroscopically-guided IA injection into inferior third of SIJ|• 32/38 patients (84%) reported a reduction in NRS pain score at 90 days; average NRS pain score reduced from 6.7±2.1 at baseline to 3.6±2.3 at 90 days post-injection (42% decrease)|No statistical comparisons/significance levels reported for NRS pain outcomes| |Saline Maugars, $1996^{11}$|10 patients (13 joints total) diagnosed with painful inflammatory sacroilitis (based on CT imaging) who failed >1 month of treatment with anti-inflammatory medications (NSAIDS or corticosteroids); no diagnostic injections performed|RCT|Saline group (n = 7 SIJ joints): 1.5 mL isotonic saline Corticosteroid (CS) group (n = 6 SIJ joints): 1.5 mL cortivazol (equivalent to 62.5 mg of prednisone)|Single, fluoroscopically-guided IA injection; patients in saline control group offered crossover to CS at 1 month|• 1-month treatment success rates, defined as >70% SIJ complex pain improvement, were significantly higher for CS vs saline group (5/6 vs 0/7 cases, respectively; P < .05) • 11-point dolorimetry scores were significantly reduced from baseline in the saline group at 1 month (7.0 ± 0.6 vs 5.2 ± 0.5; P < .005) • 6/7 cases originally allocated to saline treatment crossed over to CS at 1 month with 12/14 (86%) joints experiencing a positive 1-month outcome, which declined to 62% and 58% at 3 and 6 months, respectively; the one patient who opted not to cross over maintained 50-70% pain improvement through final follow-up at 6 months||

ANOVA, analysis of variance; CS, corticosteroid; EA, extraarticular; IA, intraarticular; LA, local anesthetic; LBP, low back pain; MCID, minimal clinically important difference; NRS, numeric rating scale; ODI, Oswestry disability index; PRF, platelet-rich fibrin; PRP, platelet-rich plasma; RMDQ, Roland-Morris disability questionnaire; SIJ, sacroiliac joint; VAS, visual analog scale.

50

Pain Medicine, 2025, Vol. 00, No. 0

Protherapy

Pragmatic RCTs

In a 2010 RCT, Kim et al. compared long-term treatment outcomes in 48 patients randomized to IA injections with corticosteroid or 25% dextrose protherapy solution.$^{248}$ Among other criteria, patients were selected by single diagnostic IA injections with local anesthetic that provided ≥50% relief of index pain. Participants allocated to both treatment arms received up to three injections at 2-week intervals. Of note, the average number of study injections was significantly higher among the 23 protherapy patients compared to the 25 patients treated with corticosteroid. Responder rates for ≥50% pain improvement were significantly higher among patients treated with dextrose protherapy vs corticosteroid at all follow-up timepoints: 64% vs 27% at 6-months and 59% vs 10% at both 10- and 15-months post-treatment.

Observational studies

These studies are presented in Table 13.

Chemoneurolysis

Pragmatic RCTs

No pragmatic RCTs were identified.

Observational studies

These studies are presented in Table 13.

Botulinum toxin

Pragmatic RCTs

No pragmatic RCTs were identified.

Observational studies

These studies are presented in Table 13.

Saline

Pragmatic RCTs

In 1996, Maugars et al. published the results of a small RCT in which 10 patients with SIJ complex pain secondary to inflammatory sacroiliitis were randomized to receive a single IA injection with either corticosteroid (n=6 joints) or saline (n=7 joints).$^{11}$ No diagnostic SIJ injections were performed as a part of the selection process. Although pain was improved significantly from baseline for the saline group at 1-month follow-up, none of the participants who received saline injections reported pain relief exceeding 70%, which was the pre-defined criterion for success. In contrast, five of six (83%) participants who received corticosteroid injections reported a successful outcome at 1 month. Between-group comparisons were not possible at subsequent follow-up timepoints, as all but one patient in the saline arm crossed over to corticosteroid treatment at 1 month. The individual who did not cross over reported 50%-70% pain improvement at 3 months and the final 6-month follow-up assessment.

#

These studies are presented in Table 13.

Summary:

* There is insufficient evidence demonstrating efficacy or effectiveness of autologous orthobiologics, allogeneic orthobiologics, hyaluronic acid, non-dextrose-based

prolotherapy, chemoneurolysis, botulinum toxin, and saline for the treatment of SIJ complex pain; Grade I evidence.

$\bullet$ Dextrose-based prolotherapy may be effective for SIJ complex pain 6 months post-procedure; Grade C evidence, low level of certainty.

$\bullet$ Platelet-rich plasma may be effective for the treatment of SIJ complex pain at 3 months post-procedure; Grade C evidence, low level of certainty.

Q12. What is the evidence for cryoablation in the treatment of sacroiliac joint complex pain?

Cryoablation of peripheral nerves leads to cellular destruction and temporary or permanent interruption of pain signal transmission along the affected nerve pathway by applying extreme cold to the target nerve, causing ice crystal formation within the nerve cells and surrounding tissue. Cryoablation may cause less extensive damage to surrounding tissues and may allow for better preservation of the nerve's connective tissue structures (endoneurium, perineurium, and epineurium) compared to RFA. This relative preservation of the nerve's scaffold can potentially facilitate remyelination and nerve regeneration in some cases, depending on the intensity and duration of the freezing process.$^{259}$ Other benefits of cryoablation may include: Creation of large volume lesions,$^{260}$ less procedural pain, and lower cost compared to some RFA technologies, though this depend in part on the system used and how costs are attributed.$^{261}$

Evidence-base for treatment of sacroiliac joint complex pain

The literature search revealed very limited published evidence addressing the clinical outcomes of cryoablation for SIJ complex pain. Two case series and one retrospective study were identified. One case series included five cases in which patients were selected for cryoablation used a diagnostic IA SIJ injection with 3 mL of 2% lidocaine.$^{261}$ A strip-lesion technique of cryoablation was used to capture the S1, S2, and S3 lateral branch nerves. The authors reported that patients experienced “more than 50%” pain relief for up to 6 months after SLB-cryoablation using combined ultrasound and fluoroscopic-guidance. Another case series included two patients who were treated with cryoablation (without a prior prognostic injection). The authors did not report which nerves were included (L5 dorsal ramus, sacral lateral branch nerves, or both). The authors reported that both patients experienced 50% pain reduction up to the end of the 4-month follow-up.$^{262}$

In a retrospective case-control study, outcomes in 39 patients treated with cryoablation for SIJ complex pain were compared to 44 patients who underwent IA SIJ corticosteroid injection. In the cryoablation group, participants were selected using a single IA SIJ injection with 1% lidocaine that resulted in ≥50% pain reduction. Both groups demonstrated statistically significant pain relief at 1-, 3-, and 6-months compared to their baseline pain scores. However, the cryoablation group demonstrated a larger magnitude of pain relief and a greater proportion of responders based on using both a ≥50% pain reduction and complete pain relief thresholds as the definitions of “treatment success”.$^{263}$

Extrapolation of evidence for treatment of similar conditions

A randomized, placebo-controlled, single-blind study compared RFA, cryoablation and sham cryoablation in 120 patients who

Pain Medicine, 2025, Vol. 00, No. 0

reported >50% pain relief following high-volume (2 mL) peri-articular lumbar facet injections. Ablations were performed with small radiofrequency electrodes (22-gauge) and large cryoablation (14-gauge) probes. The authors reported no difference between groups with regard to any outcome measure except for improvement on the Patient Global Impression of Change scale in the cryoablation group, but not the RFA group at 6 months post-procedure.$^{264}$ In a narrative review evaluating cryoneurolysis for knee osteoarthritis, non-specific knee pain and postarthroplasty knee pain (14 studies, 3 RCTs), Diep et al. found evidence for improvements in pain, quality of life, function and opioid consumption.$^{265}$ However, the low quality of the trials (10 targeted only the infrapatellar branch of the saphenous nerve, while 4 did not report the neural targets) led the authors to the conclusion that while promising, cryoablation for knee pain remains investigational.

Summary:

$\bullet$ There is insufficient evidence to assess the efficacy and effectiveness of cryoablation for SIJ complex pain; Grade I evidence.

Recommendation:

$\bullet$ Further studies, including RCTs, are warranted. Future studies should include investigation of anatomically valid techniques for sacral lateral branch nerve capture, like what has been done in clinical trials and other outcome studies evaluating SLBRFA.

Q13. Is sacral lateral branch radiofrequency ablation effective for the treatment of sacroiliac joint complex pain?

SLBRFA is the process of coagulating the neural tissues of the posterior sacral network (PSN), which is comprised of sen-sory fibers from the L5 dorsal ramus and sacral lateral branches of the S1-S4 dorsal rami, in order to disrupt noci-ceptive signals from the EA structures and posterior bony

cortex of the posterior SIJ complex.$^{108,113,266}$ Variations of this procedure have been applied for over two decades by interventional pain physicians. In a 2001 retrospective case series, Ferrante et al. first attempted RFA by lesioning the posterior inferior part of the posterior joint by sequentially walking bipolar radiofrequency cannulas cephalad in the joint space as far superior as possible. The authors reported 35% of patients experienced ≥50% pain relief through 6 months.$^{19}$ Although these results established the feasibility of RFA, future investigations that targeted the nerves themselves would go on to yield more robust findings as subsequent advances in our understanding of PSN neuroanatomy has led to the development of increasingly comprehensive and anatomically-valid lesioning protocols.

Numerous reports have now been published that address the efficacy and effectiveness of SLBRFA. Some of these studies have used anatomically validated techniques for capturing the sacral lateral branch nerves (Figure 2), while others failed to do so.$^{122}$ In addressing the question as to whether SLBRFA is effective, it is necessary to specifically evaluate anatomically-valid procedural techniques, as other techniques potentially represent modifications that are not likely to capture the sacral lateral branch nerves and therefore result in predictable technical failure (ie, small-gauge electrodes placed perpendicularly near the foramina, or a single longitudinal electrode placement technique do not provide precision or flexibility for anatomic differences). Here, we catalogue the RCTs within the published literature and describe in detail those that used an anatomically-valid SLBRFA technique.

Clinical studies

Randomized trials Explanatory RCTs using anatomically valid SLBRFA techniques

In 2008, Cohen et al. reported results of a double-blind, sham-controlled study performed in 28 patients selected by a single IA SIJ injection with local anesthetic and corticosteroid resulting in ≥75% relief of index pain.$^{21}$ Study participants were randomized in a 1:1 ratio to receive either true or placebo denervation. RFA of the L4 medial branch and L5

 Figure 2. Schematic drawing demonstrating (A) Expected peri-foraminal lesion locations, commonly used with internally-cooled radiofrequency ablation; (B) Ablation pattern for a longitudinal strip lesion extending from S1 through S3, often performed using a bipolar leapfrog technique, with a separate lesion site targeting the L5 dorsal ramus; (C) Optimal strip lesion pattern from the site of the L5 dorsal ramus, extending beyond the inferior aspect of the S3 foramen.

Pain Medicine, 2025, Vol. 00, No. 0

dorsal ramus was performed with conventional electrodes and RFA of the S1-S3/4 lateral branches used an internally cooled probe with monopolar lesioning technique. For sham procedures in the control group, electrodes were similarly situated, electrostimulation was performed in an identical manner, and lidocaine was injected, but no RFA current was applied. Control patients were given the option to cross over to active RFA treatment with a conventional (non-internally cooled) monopolar technique. Average NRS pain scores at 1-month follow-up (the primary outcome) were significantly lower for the treatment group compared to the sham group (2.4±2.0 vs 6.3±2.4; P<.001). Significantly greater improvements in function and analgesic consumption were also observed at 1 month for patients who received active vs sham RFA procedures. Similarly, 1-month treatment success rates were higher among those who received RFA vs sham procedures: 79% of participants in the RFA group met all criteria for the predefined composite variable "successful outcome" compared to only 14% in the sham group. Eleven sham arm participants crossed over to active RFA treatment by 3 months, precluding between-group comparisons at subsequent timepoints. The treatment effect of RFA remained durable through 6 months for the original active treatment group: 3- and 6-month success rates were 64% and 57%, respectively, before declining to 14% by 12 months post-RFA. For the 11 participants who crossed over from SLBRFA with conventional monopolar technology at either 1 month (n=9) or 3 months (n=2) after the sham procedure, RFA success rates were slightly (but not significantly) lower than those for the original treatment group at 1-, 3- and 6-month follow-up (44%, 67%, and 52%, respectively).

Patel et al.$^{22}$ later reported on the results of a single-blind RCT comparing outcomes of 51 participants who received either internally-cooled monopolar SLBRFA procedure or a sham procedure. SIJ complex-mediated pain was confirmed by ≥75% relief of index pain with dual fluoroscopically guided injections to anesthetize the L5 dorsal ramus and S1-S3 lateral branches. A 2:1 randomization strategy resulted in 34 and 17 patients allocated to the treatment and sham groups, respectively. Treatment success was defined as ≥50% pain reduction accompanied by a 10-point improvement on either SF-36BP or ODI score. Success rates at 3-month follow-up were significantly greater for participants in the active treatment group compared to those who received a sham procedure ($^{47}$% vs 12%, respectively). No subsequent comparisons between the treatment and sham groups were made, as all sham group patients either crossed over to SLBRFA (n = 16) or dropped out without crossover (n = 1). Among the active treatment group, successful outcomes were observed for 38% and 59% of patients at 6-and 9-months, respectively. Outcomes for patients who opted to cross over from sham to active treatment were comparable to those for the original treatment group, with 44% reporting treatment success at 3- and 6-months post-RFA.

Explanatory RCTs using non-anatomically valid SLBRFA techniques

These studies are shown in Table 14.

Pragmatic RCTs using anatomically valid SLBRFA techniques

In 2014, Zheng et al. reported the results of a randomized study conducted in 155 participants with ankylosing spondylitis who were treated with either CT-guided RFA of the L5 dorsal ramus and a conventional bipolar technique for RFA

of the S1-S4 lateral branches (n=82) or daily doses of celecoxib (n=73).$^{267}$ Patients were selected based on ≥50% pain relief following a single IA SIJ injection with local anesthetic and corticosteroid. Pain intensity scores at follow-up were lower on average for participants treated with RFA compared to those taking celecoxib at both 3- and 6-month follow-up time points (2.5 vs 4.4 at 3-months; 2.8 vs 5.0 at 6-months; P < .001). Categorical responder data were not reported.

Cánovas Martínez et al. published the results of a randomized study in 2016 comparing treatment outcomes following conventional bipolar RFA technique for coagulation of the sacral lateral branches to IA injections of local anesthetic and corticosteroid in 60 participants.$^{268}$ Participants were selected through a single IA SIJ injection with local anesthetic and corticosteroid resulting in ≥50% relief of index pain.$^{268}$ The investigators also evaluated whether a modified version of the “palisade” SLBRFA technique (12 mm electrode separation) was more effective than the original protocol (10 mm electrode separation).$^{268}$ Twenty patients were allocated to each of three treatment groups: Conventional bipolar RFA from S1 to S3, modified conventional bipolar RFA, and ultrasound-guided IA SIJ corticosteroid injection. Whereas index pain was substantially reduced from baseline at 1-month follow-up in all three groups, only the two groups treated with SLBRFA maintained statistically significant levels of pain relief at 3- and 12-month post-procedure. Between-group comparisons, which were performed for the 12-month time point only, revealed that patients belonging to the modified bipolar RFA group had significantly lower follow-up VAS pain scores than those treated with the original bipolar RFA technique or IA SIJ corticosteroid injection (4.2±0.8 vs 6.1±1.1 and 6.2±0.8, respectively; P < .001). However, it is unclear why a greater inter-electrode separation distance would result in a superior outcome given that there is a greater chance of incomplete bipolar lesion development (“hourglass” lesion effect) compared to a reduced inter-electrode separation distance, particularly with the use of small-gauge electrodes (22 ga).$^{269-271}$ This finding may be due to the small sample size.

A recent study by Cohen et al.$^{14}$ compared intermediate-term outcomes following an anatomically valid monopolar icRFA technique coagulating the L5 dorsal ramus and S1-S3/4 lateral branches to standard medical management (SMM), which included conservative therapies (eg, PT, exercise, analgesic medications, corticosteroid injections, and nerve blocks). Two hundred and ten patients were randomized in a 1:1 fashion. Patients were selected if they experienced ≥50% relief of index pain from a previous therapeutic or diagnostic IA SIJ injection, in addition to ≥50% relief of index pain following a single set of SLBBs with local anesthetic. Treatment effectiveness was evaluated based on a primary outcome of average NRS pain reduction at 3 months, as well as predefined responder rates considering both NRS pain intensity score and Patient Global Impression of Change scores. At 1-month follow-up, both average pain improvement and the proportion of patients meeting the responder definition were greater for the SLBRFA group compared to the SMM group (39% vs 14% for NRS pain improvement ≥50% and 42% vs 15% for responder rates, respectively). The differences favor-ing SLBRFA persisted at 3 months for both substantial (≥50%) pain improvement from baseline (42% vs 7%) and responder rate (52% vs 4%). The 12-month follow-up study

|Study|Population|Procedure Details|Procedure Details|Outcomes|Outcomes| |---|---|---|---|---|---| |Study|Population|SLBRFA|Comparator|Responder Rates|Pain Relief| |Explanatory RCTs $Mehta^{27}$|17 pts with LBP >6 months and 1-week average NRS pain score ≥5; positive FABER provocation test|N = 11 Monopolar RFA Cannula: 22G, 10mm active tip Target(s): L5DR Temp: not listed Duration: not listed Multipolar Strip Lesion Simplicity III Target(s): S1-S3 Temp: not listed Duration: not listed|N = 6 Sham procedure with crossover to RFA at 3 months (4/6 pts)|Responder definition not specified|3 months, RFA arm: mean improvement on NRS $58^{b,c}$ 3 months, sham arm: mean improvement on NRS $11^{b}$ 3 months, crossover arm (n = 4): mean improvement on NRS 45% 6 months, RFA arm: mean improvement on NRS $47^{b}$| |Van Tilburg$^{24}$|60 pts with LBP >3 months who did not improve with conservative care|N = 30 Monopolar RFA Cannula: 10cm Sluijter-Mehta Kit needle Target(s): L5DR Temp: $85^{C}$ Duration: 90s|N = 30 Sham procedure with 3-month crossover (19/30)|Responder definition: ≥2-point decrease in pain on NRS 1 month, RFA arm: 13/30 pts (43%) 1 month, crossover arm: 8/19 pts (42%) 1 month, sham arm: not reported|1 month, RFA arm: mean improvement on NRS 25% 1 month, crossover arm: not reported 1 month, sham arm: mean improvement on NRS 28%| |Patel$^{22,c}$|51 pts aged 18-88 years with chronic axial LBP >6 months and 3-day avg NRS pain score between 4 and 8|N = 34 icRFA$^{c}$ Cannula: 17G, 75mm electrode, 4mm active tip Target(s): L5DR, S1-S3 Temp: $60^{C}$ Duration: 150s|N = 17 Sham procedure with 3-month crossover to RFA (16/17)|Responder definition: ≥50% improvement in pain on NRS corroborated by either: 1) 10-point improvement on SF-36BP; or 2) 10-point improvement on ODI|1 month, RFA arm: mean improvement on NRS 44% 1 month, sham arm: mean improvement on NRS 29% 3 months, RFA arm: mean improvement on NRS $39^{c}$ 3 months, sham arm: mean improvement on NRS 14%| |||||||

Pain Medicine, 2025, Vol. 00, No. 0

54

|Study|Population|Procedure Details|Procedure Details|Outcomes|Outcomes|| |---|---|---|---|---|---|---| |Study|Population|SLBRFA|Comparator|Responder Rates|Pain Relief|| |$Cohen^{2,1,c}$|28 pts >18 years of age with axial low back or backtock pain lasting over 6 months (plus tenderness over SI joint) who failed to respond to conservative care; pain relief <2 months with single IA local anesthetic and corticosteroid injection under fluoroscopic guidance|N = 14 Monopolar RFA • Cannula: 22G, 5 mm active tip • Target(s): L4MB, L5DR • Temp: 80°C • Duration: 90s ic$RFA^{c}$ • Cannula: 17G, 75 mm electrode, 4 mm active tip • Target(s): S1-S3 (S4 included in 10pts) • Temp: 75°C • Duration: 150s|N = 14 Placebo with 1- (n = 9) and 3-month (n = 2) crossover to conventional RFA (11/14)|Responder definition: ≥50% improvement on NRS, a positive global perceived effect, and either 1) 10-point improvement on ODI; or 2) 4-point improvement on ODI accompanied by reduction in medication usage 1 month, RFA arm: 11/14 pts (79%)|1 month, RFA arm: mean NRS improvement 60%$^{b,d}$ 1 month, placebo arm: mean NRS improvement 3% 1 month, crossover arm: mean NRS improvement 44%$^{b,d}$ 3 months, RFA arm: mean NRS improvement 60%$^{b}$ 3 months, placebo arm (n = 2): mean NRS improvement 8% 3 months, RFA arm: 9/14 pts (64%) 3 months, placebo arm: 0/14 pts (0%) 6 months, RFA arm: 8/14 pts (57%) 12 months, RFA arm: 2/14 pts (14%)|1 month, RFA arm: mean NRS improvement 60%$^{b,d}$ 1 month, placebo arm: mean NRS improvement 3% 1 month, crossover arm: mean NRS improvement 44%$^{b,d}$ 3 months, RFA arm: mean NRS improvement 60%$^{b}$ 3 months, placebo arm (n = 2): mean NRS improvement 8% 3 months, RFA arm: 9/14 pts (64%) 3 months, placebo arm: 0/14 pts (0%) 6 months, RFA arm: 8/14 pts (57%) 12 months, RFA arm: 2/14 pts (14%)| |Pragmatic RCTs $Cohen^{14,c}$|210 pts aged >21 years with SIJ complex pain lasting ≥3 months and 7-day average NRS score ≥4; at least one positive clinical provocation test; ≥50% pain relief from prior therapeutic or diagnostic injection|N = 105 ic$RFA^{c}$ • Cannula: 17G, 4 mm active tip • Target(s): L5DR, S1-S3 (S4 included at provider discretion) • Temp: >80°C • Duration: 150 s|N = 105 Standard medical management (SMM): • Included corticosteroid injections, nerve blocks, physical therapy, analgesics, discretion) • Temp: >80°C • Duration: 150 s|Responder definition: ≥30% or 2-pt decrease in 7-day average NRS and ≥5 on PGIC at 3 months|1 month, RFA arm: mean NRS improvement 37%$^{c,d}$ 1 month, SMM arm: mean NRS improvement 17% 3 months, RFA arm: mean NRS improvement 39%$^{c,d}$ 3 months, SMM arm: mean NRS improvement 5%|| ||Single SLBBs with ≤2 mL long-acting LA requiring ≥50% pain relief under fluoroscopic guidance||||||

(continued)

54

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Population|Procedure Details|Procedure Details|Outcomes|Outcomes||| |---|---|---|---|---|---|---|---| |Study|Population|SLBRFA|Comparator|Responder Rates|Pain Relief||| |$Eissa^{273}$|30 pts with LBP lasting >3 months and unresponsive to conservative care; required to have SIJ complex discomfort with a positive response to at least 3 provocative clinical movement tests|N = 15 Monopolar RFA • Cannula: 22G, 5 mm active tip • Target(s): L4, L5DR, S1-S3 • Temp: 80°C • Duration: 90 s|N = 15 IA corticosteroid injection (2 mL methylprednisolone + 1 mL lidocaine 2%; n = 15)|Responder definition: >50% improvement on NRS 1 month, RFA arm: 9/15 pts (60%) 1 month, corticosteroid arm: 7/15 pts (46%) 3 months, RFA arm: 9/15 pts (60%) 3 months, corticosteroid arm: 6/15 pts (40%)|1 month, RFA arm: mean NRS improvement 53%$^{b}$ 1 month, corticosteroid arm: mean NRS improvement 53%$^{b}$ 3 months, RFA arm: mean NRS improvement 55%$^{b,d}$ 3 months, corticosteroid arm: mean NRS improvement 37%$^{b}$||| |Cánovas $Martínez^{268,c}$|Single IA injection with LA (2 mL lidocaine 2%) under fluoroscopic guidance requiring ≥50% pain relief|N = 40 Conventional Bipolar RFA (1 cm distance between electrodes, n = 20)$^{c}$ • Cannula: 22 G, 10mm active tip • Target(s): S1-S3 • Temp: 90°C • Duration: 180 s|N = 20 Two ultrasound-guided IA injections of 3 mL 5% levoubupivacaine + 12 mg betamethasone sodium with a 7-day interval|Responder definition not specified|1 month, corticosteroid arm: mean VAS improvement 53%$^{b}$ 1 month, conventional RFA arm: mean VAS improvement 41%$^{b}$ 3 months, corticosteroid arm: mean VAS improvement 15% 3 months, conventional RFA arm: mean VAS improvement 41%$^{b}$ 3 months, modified RFA arm: mean VAS improvement 52%$^{b}$ 12 months, corticosteroid arm: mean VAS improvement 9% 12 months, conventional RFA arm: mean VAS improvement 25% (P = .01)$^{d}$ 12 months, modified RFA arm: mean VAS improvement 48%$^{d}$|1 month, corticosteroid arm: mean VAS improvement 53%$^{b}$ 1 month, conventional RFA arm: mean VAS improvement 41%$^{b}$ 3 months, corticosteroid arm: mean VAS improvement 41%$^{b}$ 3 months, modified RFA arm: mean VAS improvement 52%$^{b}$ 12 months, corticosteroid arm: mean VAS improvement 9% 12 months, conventional RFA arm: mean VAS improvement 25% (P = .01)$^{d}$ 12 months, modified RFA arm: mean VAS improvement 48%$^{d}$||

(continued)

55

|Study|Population|Procedure Details|Procedure Details|Outcomes|Outcomes| |---|---|---|---|---|---| |Study|Population|SLBRFA|Comparator|Responder Rates|Pain Relief| |$Salman^{18}$|30 pts aged ≥18 years with LBP lasting ≥6 months|N = 15|N = 15|Responder definition: ≥50% VAS improvement|1 month, RFA arm: mean VAS improvement 69%$^{b,c}$ 1 month, corticosteroid arm: mean VAS improvement 3%| |$Salman^{18}$|Single IA injection confirmed by SIJ arthrogram with LA (2 mL lidocaine 2% + 1 mL bupivacaine 0.5%) requiring ≥75% pain relief|Monopolar RFA • Cannula: 22 G, 10mm active tip (L4-L5); 22 G, 5 mm active tip (S1-S3)• Target(s): L4MB, L5DR, S1-S3• Temp: 80°C• Duration: 90 s|IA corticosteroid (1 mL depo-methylprednisolone 40 mg/mL)|1 month, RFA arm: 11/15 pts (73%)$^{c}$ 1 month, corticosteroid arm: 3/15 pts (20%) 3 months, RFA arm: 9/15 pts (60%)$^{c}$ 3 months, corticosteroid arm: 0% 6 months, RFA arm: 8/15 pts (53%)$^{c}$ 6 months, corticosteroid arm: 0%|1 month, RFA arm: mean VAS improvement 69%$^{b,c}$ 1 month, corticosteroid arm: mean VAS improvement 3% 3 months, RFA arm: mean VAS improvement 63%$^{b,c}$ 3 months, corticosteroid arm: mean VAS improvement 0% 6 months, RFA arm: mean VAS improvement 58%$^{b}$ 6 months, corticosteroid arm: no pts remaining| |$Zheng^{267,c}$|155 pts diagnosed with ankylosing spondylitis with LBP lasting >3 months and pain exacerbation upon SIJ palpation|N = 82 Bipolar $RFA^{c}$ • Cannula: 20 G, 5 mm active tip • Target(s): L5DR, S1-S4• Temp: 90°C• Duration: 180 s|N = 73 Celecoxib (NSAID): 200 mg twice daily for 24 weeks|Responder definition not specified|3 months, RFA arm: mean VAS improvement 65%$^{b,d}$ 3 months, NSAID arm: mean VAS improvement 36%$^{b}$ 6 months, RFA arm: mean VAS improvement 61%$^{b,d}$ 6 months, NSAID arm: mean VAS improvement 28%$^{b}$| |$Zheng^{267,c}$|Single IA injection with 3 mL triamcinolone 40 mg + bupivacaine 0.5% under fluoroscopic guidance requiring ≥50% pain relief|||||

CI, confidence interval; GPE, Global Perceived Effect; IA, intraarticular; icRFA- internally-cooled radiofrequency ablation; L4MB, L4 medial branch; L5DR, L5 primary dorsal ramus; LA, local anesthetic; LBP, low back pain; NRS, Numerical Rating Scale; NSAID, non-steroidal anti-inflammatory drug; ODI, Oswestry Disability Index; PGIC, Patient Global Impression of Change; SF-36BP, 36-Item Short Form Survey, Bodily Pain subscale; SIJ, sacroiliac joint; SLBRFA, sacral lateral branch radiofrequency ablation; VAS, Visual Analog Scale.

a Anatomically valid technique.

b P < .05 vs baseline of the respective group.

c P < .05 vs comparator group (responder rate/mean improvement from baseline).

d P < .05 vs comparator group (mean follow-up score).

Pain Medicine, 2025, Vol. 00, No. 0

revealed sustained relief (56% combined responder rate) in both the original icRFA group and icRFA patients who crossed over from SMM.$^{272}$

Pragmatic RCTs using non-anatomically valid SLBRFA techniques

These studies are shown in Table 14.

Summary:

$\bullet$ SLBRFA is effective in the treatment of SIJ complex pain for at least 3 months post-procedure using the criteria of ≥50% pain reduction to define a robust improvement threshold in individuals with chronic low back pain; Grade B evidence, moderate-to-high level of certainty. $\bullet$ SLBRFA is effective in the treatment of SIJ complex pain for at least 6 months post-procedure when non-anatomically valid RFA techniques that fail to account for neuroanatomy are excluded; Grade B evidence, moderate level of certainty.

* SLBRFA is effective in the treatment of SIJ complex pain for at least 6 months post-procedure when non-anatomically valid RFA techniques that fail to account for neuroanatomy are excluded; Grade B evidence, moderate level of certainty.

Q14. What nerves should be targeted (dorsal innervation) during sacral lateral branch radiofrequency ablation?

There are several (sometimes contradictory) reports on inner- vation of the SIJ complex leading to ongoing debate on this topic, with the most inconsistent reports pertaining to the ventral part of the SIJ complex. One of the earliest reports on SIJ complex innervation is from Horwitz,$^{110}$ who during the dissection of 114 human cadaveric pelvises, traced numerous branches from the first and second sacral nerves to the long and short posterior sacroiliac ligaments. In a subsequent monograph by Solonen, the author described extensive inner- vation of the SIJ complex emanating from the S1 and S2 spal nial nerves dorsally, and ventrally from L3-S2 spinal nerves.$^{114}$ Ikeda reported posterior SIJ complex innervation arising from the L5 and sacral dorsal rami and anterior inner- vation arising from the S1 and S2 ventral rami.$^{111}$ Bradley asserted that the nerves innervating the posterior portion of the SIJ complex are closely associated with the interosseous and sacrotuberous ligaments.$^{106}$ The main nerve trunks in this region vary significantly in diameter, with the fifth lum- bar nerve descending vertically near the S1 superior articular process, then dividing into lateral and medial branches. The lateral branch joins up with the first sacral nerve, while the medial branch curves beneath the lumbosacral facet joint and enters the multifidus muscle. Bradley described a complex network of nerves, including branches from the fifth lumbar and first to fourth sacral nerves, which contribute to the innervation of the posterior sacral region and sacroiliac ligaments, which was named the "Posterior Sacral Network" (PSN) by Roberts et al.$^{113}$ Figure 1 demonstrates the location of the posterior interosseous ligaments and sacral lateral branches.

In the 1990s, two studies evaluating the innervation of the SIJ complex were published, one by a Japanese author and the other by German authors. In the earlier study by Ikeda$^{111}$ in which 13 specimens (26 SIJ complexes) were examined anteriorly, and 7 specimens (14 SIJ complexes) were examined posteriorly, the authors reported dorsal innervation, which was consistent with earlier findings, but also identified extensive ventral innervation (see Table 16). According to macroscopic examination, the author found the upper ventral portion of the SIJ complex to be innervated by the ventral

ramus of the fifth lumbar nerve in seven cadavers bilaterally, and in two unilaterally, for an innervation rate of 61.5%. The lower ventral portion of the joint was supplied by the ventral ramus of S2 or branches from the sacral plexus in eight dissections unilaterally, and in the remaining five dissectations bilaterally, for an innervation rate of 69.2%. In the dorsal upper part of the SIJ complex, the lateral branch of the L5 dorsal ramus penetrated the posterior and interosseous ligaments of the SIJ on seven of the 14 hemipelves (50%). Innervation arising from S1 dorsal rami was observed in three hemipelves from three specimens (21.4%). In the lower portion of the dorsal SIJ, 100% of the branches were mainly directed to the SIJ after a network of communication between the S2 and S3 lateral branches, which the authors referred to as the "dorsal plexus." Additionally, in 21.4% of specimens, SIJ complex innervation was derived directly from the dorsal ramus of S3.

In an anatomical dissection and histochemistry study by Grob et al. using four male and four female pelvises,$^{109}$ the authors reported that the posterior innervation was exclusively derived from the S1-S4 dorsal rami. They found that after the S1-4 dorsal rami exit the posterior sacral foramina, they emerge in the loose connective tissue of the dorsal SIJ complex ligaments where they anastomose in a longitudinal fashion. Fine branches innervated the dorsal ligaments. Branches from S3 and S4 extended into the sacrotuberous and sacrospinous ligaments. Based primarily on the results of histological and immunocytochemical analyses conducted on fetal SIJs, which demonstrated the presence of neurofilament-positive nerve fibers in their dorsal portions, the authors could not identify significant ventral joint innervation, finding no neural contribution from the sacral plexus or obturator nerve. In their histological examination of dissected nerves, they found a multitude of thick, myelinated fibers, as well as thin myelinated and unmyelinated fibers, indicating a diverse array of sensory receptors, including encapsulated mechanoreceptors.

In 2002, Vilensky et al.$^{116}$ examined histological staining patterns of ligaments tissue adjacent to the medial aspect of the posterior superior iliac spine and sacrum obtained from six patients scheduled for posterior SIJ arthrodesis. Using the gold chloride staining technique, the authors reported both myelinated and unmyelinated nerve fibers, as well as mechanoreceptors. However, when they tried to confirm their findings using immunohistochemical staining for protein gene product 9.5, the presence of axons, nerve fascicles, or mechanoreceptors could not be conclusively identified. When immunohistochemical staining for substance P (a neuropeptide associated with pain signaling) was performed, reactive components were identified, but due to background staining, the authors could not definitively conclude these components represented neural tissue. In a 2008 immunohistochemical study by Szadek et al.,$^{3}$ the authors confirmed the presence of Substance P- and CGRP-immunoreactive fibers in tissue obtained from anterior and posterior interesseous ligaments surrounding the SIJ. During dissection, the authors observed small fibers from the lumbosacral plexus traveling toward the anterior SIJ ligaments.

Since RFA treatments for SIJ complex pain target the posterior portion of the joint and associated dorsal ligaments, the most recent anatomical studies have focused on this part of the joint complex. Only one study by Cox et al.$^{107}$ focused on the ventral portion of the joint, finding small branches

||Lateral branch|Contribution to posterior SIJ innervation (%, N)|Number of branches|Exit quadrant relative to the sacral tubercle (N)| |---|---|---|---|---| |Roberts et al. $^{113}$|L5|8% (2)|1 branch (n = 25)|IL (2)| |Roberts et al. $^{113}$|S1|100% (25)|1 branch (n = 15) 2 branches (n = 10)|Superior to posterior sacral ligament (23) IL (15)| |Roberts et al. $^{113}$|S2|100% (25)|2 branches (n = 19)|Emergence of superior branch Emergence of inferior branch Emergence of superior branch Emergence of inferior branch| |Roberts et al. $^{113}$|S3|88% (22)|1 branch (n = 18) 2 branches (n = 3) 3 branches (n = 1)|IL (5) SL (1) IL (21)| |Cox and $^{108}$|S4|4% (1)|1 branch (n = 1) 1 branch (n = 9)|2 branches originated from IL quadrant and furnished SIJ innervation directly SL (1)| |Cox and $^{108}$|L5|75% (9)|1 branch (n = 9)|SL (1)| |Cox and $^{108}$|S1|100% (12)|2 branches (n = 1) 3 branches (n = 8)|foramen were from 1:00 to 5:00 o'clock on the right and 8:00 to 10:00 on the left| |Cox and $^{108}$|S2|92% (11)|1 branch (n = 5) 2 branches (n = 4) 3 branches (n = 2) 1 branch (n = 2) 2 branches (n = 6)|SL (2) SL & IL (9:00 (3) SL & IL (4) SL & IL (2) IL (1) SL (1) SL & IL (8:00, 2:00-5:00 (6)) SL & IL (3) SL & IL (1) SL & IL (1)| |Cox and $^{108}$|S3|67% (8)|1 branch (n = 1) 4 branches (n = 1) 2 branches (n = 7) 3 branches (n = 1)|IL (1) IL (1) SL & IL (9:00 (6)) SL & IL (1) SL & IL (1) IL (1) SL & IL (6) IL (1)|

IM, inferomedial; IL, inferolateral; SL, superolateral; SM, superomedial.

Pain Medicine, 2025, Vol. 00, No. 0

|Author|L5|S1|S2|S3|S4|Superior gluteal|Number of specimens (Hemipelvic dorsal dissections)| |---|---|---|---|---|---|---|---| |Horwitz $^{110}$|||33.3% (20)|13.3% (8)|||30 (60)| |Solonen $^{114}$|(+)|(+)|5% (3)|6.7% (4)|||| |Bradley $^{106}$|100%|100%|100%|100%|100%|(+)|x| |Ikeda $^{111}$|50% directly to SI joint|21.4% directly to SI joint|21.4% directly to SI joint|21.4% directly to SI joint|||7 (14)| |Grob $^{109}$|||Dorsal plexus 100% indirectly to SI 100% Dorsal Rami of S1-S4|100% Dorsal Rami of S1-S4|||8 (x)| |McGrath & Zhang $^{112}$||96% (21) passed superiorly away from LPSL. 4% (1/22) penetrated the LPSL|96% (21) penetrated the LPSL obliquely. 4% penetrated near superior end|100% (22) penetrated the LPSL obliquely. 4% penetrated near superior end|59% (13) 60% passed superficially above or emerged from the superficial portion of the LPSL. 40% passed inferiorly away from LPSL. 4% (1)||16 (22)| |Roberts $^{113}$|8% (2)|100% (25)||88% (22)|4% (1)||25 (x)| |Cox & Fortin $^{108}$|75% (9)|100% (12)|92% (11)|92% (11)|67% (8)|42% (5)|9 (12)| |||Average range of the lateral branches of the dorsal sacral foramina were from 1:00 to 5:00 O'clock on the right and 8:00 to 10:00 on the left|||||| |Steinke $^{115}$|||Communicating branches of the PSN did not reach the SIJ, L5-S1 branches pierced the i-SIL and S1 to S4 pierced the o-SIL||||43 (6)|

iSIL, inner sacroiliac ligaments; LPSL, long posterior sacroiliac ligament; oSIL, outer sacroiliac ligaments; PSL, posterior sacroiliac ligament; PSN, posterior sacral network.

Pain Medicine, 2025, Vol. 00, No. 0

originating from either the L4 ventral ramus (10%), the L5 ventral ramus (80%), or from both the L4 and L5 ventral rami (10%). A combined cadaveric and clinical study by Yin et al.$^{25}$ reported results consistent with previous studies, finding at least one sacral lateral branch nerve from S1 to S3 innervating the SIJ (ranging from 2:00 to 6:00 O'clock on the right or 6:00 to 10:00 O'clock on the left), with an S4 branch identified in one specimen. These sacral lateral branches frequently anastomosed with the sacral arcade and occasionally sacral medial branches. A study by Roberts et al. further defined the dorsal SIJ complex innervation in order to inform optimal SLBRFA technique.$^{113}$ The authors found that the PSN emerges from the superolateral and inferolateral quadrants of the posterior sacral foramina relative to the sacral tubercle, located directly on the periostum or on the overlying interesseus sacroiliac ligament. More laterally, the nerves were covered and present within the long posterior sacral ligament. The contribution of the lateral branches to the innervation of the SIJ complex is shown in Table 15. Although Roberts and colleagues reported that the S4 lateral branch innervated the SIJ complex only in one specimen (4%), this finding is clinically important for SLBRFA techniques. In this specimen, the inferior border of the SIJ was adjacent to the S4 neuroforamen. This is consistent with previous anatomical studies that reported an S4 lateral branch contribution to SIJ complex innervation in between 13% and 100% of specimens,$^{25,106,109,110,112}$ more commonly when the foramen was level to or above the inferior margin of the SIJ.$^{113,274}$ As such, paraforaminal S4 lateral branch ablation should be considered in these cases. The authors concluded that RFA targets should be located along the medial portion of the PSN, just as they exit the foramina.$^{113}$

The most recent study by Steinke et al. reported the results of both dorsal and ventral dissections in 51 cadavers.$^{115}$ Although the authors described an extensive network of the nerves ranging from L5 to S5 penetrating the dorsal ligaments of the SIJ complex, they emphasized that most of these nerves did not reach the SIJ itself. However, similar anatomical dissection studies have confirmed an extensive dorsal innervation network extending to the SIJ (see Tables 15 and 16).$^{108,112}$,$^{113}$

Clinical studies

The results of clinical trials are consistent with cadaveric studies in terms of levels targeted. All sham-controlled trials have targeted innervation from the L5 to S3 spinal nerve levels. In the first placebo-controlled trial by Cohen et al. in 2008, the authors also lesioned the L4 medial branch nerve and S4 lateral branch nerve in the 36% of individuals in whom the foramen was situated at or above the level of the lower margin of the SIJ.$^{21}$ In another study comparing icRFA to sham ablation, Patel et al. targeted L5 dorsal ramus and S1-3 lateral branches.$^{22}$ In the only definitively negative study, van Tilburg et al. targeted the L5-S4 spinal nerve levels in all patients;$^{24}$ however, it is notable that these investigators used RFA technology and a technique that is not anatomically valid and likely did not effectively capture the sacral lateral branches.$^{121,122}$ Finally, a recent multi-center comparative-effectiveness study by Cohen et al. targeted L5-S3 in every participant, and S4 at the discretion of the physician when the foramen was located at or above the inferior margin of the SIJ.$^{14,272}$

Transitional anatomy

Transitional anatomy, including partial or complete sacralization of L5, is associated with back pain in general and SI joint pain in particular. One study found a prevalence of transitional anatomy in 26% in patients with low back pain. Among these patients with low back pain, the prevalence of SIJ complex pain was 28.5% in those with transitional anatomy vs 10.8% in those with normal lumbosacral anatomy, with sacralization of L5 being particularly common.$^{275}$ In these individuals, because the SIJ extends more cephalad, there are biomechanical alterations that increase the likelihood of pain, especially when the altered anatomy is unilateral, and theoretically a greater likelihood of L5, or even L4 (which often assumes the innervation pattern of L5), innervating the superior joint.$^{276,277}$

Summary:

$\bullet$ The dorsal innervation to the SIJ is complex and variable between individuals, with most high-quality cadaveric studies demonstrating innervation from S1 to S3 in over 90% of specimens. For the L5 dorsal ramus and S4 lateral branch nerve, most studies have found that over 50% of cadaveric specimens contain branches innervating the dorsal ligaments. Grade B evidence, moderate level of certainty.

$\bullet$ S1 to S3 should be targeted close to the lateral margin of the neural foramina in all patients undergoing SLBRFA; Grade A recommendation, high level of certainty.

* Given variability between individuals, physicians should consider also targeting the L5 dorsal ramus (in those with transitional anatomy or upper SIJ complex pain) and the S4 lateral branch when the foramen is located at or above the inferior margin of the SIJ, to ensure the most complete posterior SIJ complex denervation possible; Grade B recommendation, moderate level of certainty.

Q15. What should the pain relief cutoff be for selecting patients for treatment based on prognostic sacroiliac joint injections or sacral lateral branch blocks and for designating an outcome as positive after therapeutic sacroiliac joint injection, sacral lateral branch radiofrequency ablation, and sacroiliac joint fusion surgery?

The question of what percent pain relief constitutes a positive injection depends on the purpose of the injection and what type of injection is being considered, with SLBBs being purely prognostic before RFA.

Therapeutic sacroiliac joint complex procedures

SI injections can be IA, EA, diagnostic, prognostic, and therapeutic. For therapeutic injections, randomized trials evaluating outcomes have employed different cutoffs for designating a binary result as “positive.” For therapeutic injections, most randomized trials that contained a categorical outcome used ≥50% pain relief as the cutoff for designating that outcome as positive,$^{242,248,256}$ though the largest randomized trial comparing IA to EA injections used a ≥ 2-point pain reduction coupled with a positive satisfaction score on a 5-point Likert scale.$^{6}$ This is consistent with the IMPMPACT guidelines and randomized studies specifically evaluating LBP treatments, which assert that a ≥ 2-point or 30% reduction in pain is considered to constitute clinically meaningful benefit,

Pain Medicine, 2025, Vol. 00, No. 0

|First author, year|Design and pain relief threshold|Treatment groups|Designation of positive $outcome^{a}$|Results| |---|---|---|---|---| |Maugars, $1996^{11}$|Randomized, double-blind, placebo-controlled trial in 10 pts (13 joints) with spondyloarthropathy and clinically-diagnosed SI joint pain|IA injection of corticosteroid or saline|Good improvement >70% relief at 1-month, fair improvement 50%-70% relief|Corticosteroids > placebo| |Kim, $2010^{248}$|Randomized, comparative-effectiveness trial in 50 pts who obtained ≥50% pain relief from diagnostic injection|IA protherapy or corticosteroid + LA up to 3 times|≥50% pain relief at 2 weeks|Equivalent in short term, protherapy > corticosteroids in long term| |Lee, $2010^{256}$|Randomized, comparative-effectiveness trial in 39 pts who obtained ≥50% pain relief from diagnostic injection|EA botulinum toxin or corticosteroid + LA|≥50% pain relief at 1 month|Equivalent in short-term, botulinum toxin > corticosteroids in intermediate term| |Visser, $2013^{134}$|Randomized comparative-effectiveness study in 51 pts with clinically-diagnosed SI joint pain|Physical therapy or manual therapy or IA corticosteroid + LA|Any reduction in pain score at 12 wks or complete reduction at 6 wks|Manual therapy > IA injection > physical therapy| |Jee, $2014^{154}$|Randomized, single-blind trial in 120 pts who obtained >80% pain relief from IA diagnostic injection|Fluoroscopically- vs. ultrasound-guided IA injection with corticosteroid + LA|Verbal pain scale 2 and 12 weeks after injection (no categorical outcome)|No difference in outcomes, fluoroscopy more accurate| |Singla, $2017^{242}$|Randomized, comparative-effectiveness trial in 40 pts with clinically-diagnosed SI joint pain|IA platelet-rich plasma or corticosteroid + LA|≥50% pain relief at 2-week to 3-month follow-ups.|Platelet rich plasma > corticosteroid| |Cohen, $2019^{6}$|Randomized, double-blind comparative-effectiveness trial in 125 pts with clinically-diagnosed SI joint pain|Fluoroscopically-guided (usually IA) or landmark-guided SI joint injections|≥2-point reduction in pain + positive satisfaction score at 1 month|Equivalent at 1-month, fluoroscopically-guided and IA > landmark-guided and EA injections for some outcome measures at 3 months.|

EA, extra-articular; IA, intra-articular; LA, local anesthetic.

ª Does not necessarily denote the primary outcome measure.

Pain Medicine, 2025, Vol. 00, No. 0

and is the most commonly used cutoff in randomized drug trials seeking regulatory approval.$^{278,279}$ One study from the 1990s used a 70% cutoff to constitute “good improvement,” with 50% indicating “fair improvement,” while another considered any sustained pain relief from baseline to be a positive outcome (see Table 17).$^{11,134}$

Diagnostic sacroiliac joint injections

The Centers for Medicare and Medicaid Services require ≥75% pain relief for a "diagnostic" injection to be considered positive, but note that subsequent therapeutic injections require ≥50% pain relief to be considered reasonable and necessary.$^{280}$ This inconsistency may be due to conflation of the terms "diagnostic" and "prognostic" in the medical literature, as SIJ injections can be used to predict the response of more invasive procedures such as SLBRFA and SIJ fusion surgery. Compared to more definitive procedures such as SLBRFA, patients generally experience greater relief with prognostic injections (eg, EA or IA) injections because of their lack of specificity (ie, spread to non-targeted pain-generating structures).

In epidemiological prevalence that have used controlled injections, cutoff thresholds have been equally divided between those that used ≥50% pain relief to identify the SIJ as the primary pain generator^44,281,282 and those that employed higher cutoff values (≥75%, including two studies by the same group of investigators).^53,128,283 In a recent systematic review by an international group of experts that sought to determine the diagnostic utility of physical exam signs with diagnostic injections as the reference standard, the authors used a ≥50% pain relief cutoff as the criterion for diagnosis.⁷¹ In an earlier systematic review, Kennedy et al. found only a small difference in prevalence rates stratifying epidemiological studies based on 50% vs ≥50% cutoff thresholds but found that using a second set of injections significantly reduced the “positive” rate. The authors also noted that the addition of corticosteroids, which is commonly used even in purely diagnostic injections, increases the rate of “positive” injections (see Table 18).^8,284

Extra-articular injections as stand-alone prognostic blocks

There are no randomized trials that have used only response to EA injections to select patients for RFA and the few that have used this approach are small, observational, and older studies that used non-validated RFA techniques. In one of the earliest observational studies evaluating SLBRFA, Yin et al. required at least 70% pain relief after two high-volume EA injections with corticosteroid before proceeding to RFA$^{25}$; the success rate in this small (n = 14) retrospective study, in which sacral lateral branch targets were selected based on sensory stimulation before lesioning (L5 and S1 in all patients, S2 and S3 in select patients), was 64%. In a 2002 prospective observational study, Gevargez et al. performed CT-guided L5 dorsal ramus and posterior interosseous ligament RFA in 38 (of 43) patients who obtained only short-term (<2 weeks) but “definite” improvement following “intra-ligamentous neural blockade.” Three months post-procedure, 65.8% of patients reported at least substantial pain relief.$^{20}$

Diagnostic-prognostic sacroiliac joint injections and sacral lateral branch blocks before radiofrequency ablation and categorical outcome thresholds

For randomized trials evaluating RFA, studies have been roughly equally divided in using 50% (n=3) or higher (n=4) cutoffs for both IA SIJ injections and SLBBs in selecting patients for RFA, with no differences in outcomes.$^{14,16,21,22,24,27}$ In the only negative study, a cut-off of ≥2-points was used, but the investigators inexplicably included patients with radicular pain and also used an unvali-dated technique for SLBRFA.$^{24,122}$ No study used a higher cutoff value than 50% to designate a categorical outcome as “positive,” with most of the more recent studies,$^{14,16,24,27}$ including the two largest comparative-effectiveness studies$^{14,16}$ using ≥2-points as the designation standard for a clinically important outcome, consistent with the IMPMPACT guidelines.$^{278}$ In two studies comparing SLBRFA outcomes stratified by percent pain relief following SIJ injections or SLBBs, both Cohen et al. and Amatto et al. found no difference between 50% and 80% thresholds.$^{119,285}$

Diagnostic-prognostic joint injections before surgery and categorical outcome thresholds

In a randomized trial by Whang et al. comparing minimally-invasive SIJ fusion to nonsurgical management, the authors used ≥50% pain relief from an IA SIJ local anesthetic injection to select patients for enrollment, and a ≥20 mm reduction on a 0-100 VAS pain scale (equivalent to ≥2-points on a 0-10 pain scale) as the binary primary outcome measure to assess the effect of the SIJ fusion intervention.$^{286}$ In a subsequent randomized trial that also compared minimally-invasive SIJ fusion using triangular titanium implants to conservative therapy, Sturesson et al. used the same diagnostic injection cutoff (≥ 50% pain relief) and the same criteria for a positive binary outcome (≥20 mm on a 0-100 VAS scale).$^{287}$ A randomized study comparing SIJ fusion to sham fusion also used ≥50% pain relief from an IA SIJ local anesthetic injection to select patients, and similar to other SIJ fusion trials, the authors designated an outcome as positive if a ≥2-point change was reported on a 0-10 NRS pain scale.$^{288}$ In this first-ever sham-controlled trial comparing minimally-invasive fusion to sham placement and removal of a single blunt pin from the surface of the ilium, the authors reported no significant difference in any outcome measure between treatment groups.$^{288}$

Summary:

* $\ge 50\%$ pain relief should be considered the cutoff for considering a diagnostic or prognostic SIJ injection to be "positive" with regard to selecting patients for SLBRFA; Grade B recommendation, moderate level of certainty.

$\bullet$ For EA injections as stand-alone prognostic tools, there is insufficient evidence to designate their use, or a specific cutoff threshold, before RFA given the low-quality studies and invalid RFA techniques; Grade I evidence.

$\bullet$ Evidence is mixed for using >50% pain relief as the prognostic threshold to select patients for SIJ fusion, with two pragmatic RCTs showing effectiveness but a sham-controlled study failing to distinguish SIJ fusion from sham outcomes. Considering the invasiveness of SIJ fusion surgery, in the context of these data, dual prognostic SIJ injections without corticosteroid and with a high

Pain Medicine, 2025, VOL. 00, No. 0

|First author, year|Design and pain relief threshold|Treatment groups|Designation of positive $outcome^{a}$|Results| |---|---|---|---|---| |Cohen, $2008^{21}$|Randomized, double-blind, sham-controlled trial in 28 pts who obtained ≥75% temporary pain relief from IA joint LA and corticosteroid injection (volume 3 mL with EA extravasation)|L5 dorsal ramus and S1 to S3/54 lateral branch icRFA + local anesthetic or local anesthetic and sham RFA|≥50% pain relief coupled with positive global perceived effect requiring no additional therapy from 1-6 months.|RFA > sham RFA| |Patel, $2012^{22}$|Randomized, double-blind, sham-controlled trial in 51 pts who obtained short-lasting pain relief (% not specified) from IA or EA SIJ injection + ≥75% relief from L5 dorsal ramus and S1-3 SLBBs|L5 dorsal ramus and S1 to S3 lateral branch icRFA + LA or LA and sham RFA|≥50% pain relief coupled with either a 10% change in Short Form-36 bodily pain or Oswestry Disability index, from 1-9 months.|RFA > sham RFA at 3-month primary endpoint| |Salman, $2016^{18}$|Randomized, double-blind, comparative-effectiveness study in 30 pts who obtained ≥75% relief from IA injection with LA (volume 3 mL with EA extravasation)|L4 and L5 dorsal rami and S1-3 lateral branch RFA or IA corticosteroid injection|≥50% pain relief at 1, 3 and 6 months|RFA > corticosteroids| |van Tilburg, $2016^{24}$|Randomized, double-blind, sham-controlled trial in 60 pts (included pts with sciatica) who obtained ≥2-point decrease in average pain from IA injections with LA (volume 3 mL with EA extravasation)|L5 dorsal ramus and S1-S4 RFA with single probe containing 3 active electrodes + LA or LA and sham RFA|Pain score on 0-10 NRS at 1 and 3 months; no categorical outcome noted, but the percentage of crossover pts who obtained ≥2-point pain relief at 1 month noted.|No difference in outcomes| |Juch, $2017^{16}$|Randomized, controlled trial in 228 pts who obtained ≥50% pain relief from L5 dorsal ramus and S1-3 SLBBs|L5 dorsal ramus and S1-3 RFA with either internally-cooled, bipolar RFA or RFA with a single probe containing multiple electrodes, with LA before lesioning + exercise therapy, or exercise therapy without RFA. Pts could receive psychotherapy in both groups if needed.|≥2-point decrease in pain on 0-10 NR5, and/or ≥20 percentage points on 0%-100% Oswestry disability index.|Difference in primary outcome measure (pain reduction) favoring RFA at primary endpoint but not afterwards. For categorical outcome, RFA > control group at 3 wks but not later.|

(continued)

|First author, year|Design and pain relief threshold|Treatment groups|Designation of positive $outcome^{a}$|Results| |---|---|---|---|---| |Mehta, $2018^{27}$|Randomized, double-blind study in 17 pts who obtained >80% pain relief from 2 IA injections with LA|RFA with LA using a single probe containing multiple electrodes or LA + sham RFA|Improvement of ≥2-points on 0-10 pain scale, at 3 and 6 months.|RFA > sham RFA at 3-month primary endpoint| |Cohen, $2024^{14}$|Randomized, comparative-effectiveness study in 210 pts who obtained ≥50% relief from IA or EA SIJ and ≥50% relief from L5 dorsal ramus and S1-3 or 4 SLBBs|icRFA with LA at L5 dorsal ramus and S1-3 or 4 lateral branches or standard medical management|Improvement of ≥2-points or 30% on 0-10 pain scale + at least moderately better (≥5/7) on Patient Global Impression of Change scale, followed through 12 months.|RFA > standard medical management at 3-month primary endpoint and after.| |Whang, $^{286}$|Randomized comparative-effectiveness study in 148 patients who obtained ≥50% pain relief from an IA SIJ injection|Minimally-invasive surgical fusion or conservative treatment that included medications, physical therapy, joint injections and RFA in a stepwise fashion. Cognitive behavioral therapy was not permitted.|Reduction in pain score by ≥20mm on a 0-100 VAS scale and absence of neurological or surgical complications.|Surgery > conservative care at 6 months and later follow-ups.| |Sturesson, $^{287}$|Randomized comparative-effectiveness study in 103 patients who obtained ≥50% pain relief from an IA SIJ injection|Minimally-invasive surgical fusion or conservative treatment that included medications, physical therapy, psychological interventions if indicated. Nonsurgical procedures were not permitted.|Pain score on 0-100 VAS at 6 months. No categorical outcome noted, but the percentage of pts who obtained ≥20 mm pain relief was reported.|Surgery > conservative care at 6 months and later follow-ups.| |Randers, $^{288}$|Randomized, double-blind, sham-controlled trial in 63 pts who obtained ≥50% pain relief from an IA SIJ injection|Minimally-invasive surgical fusion or sham surgery (pin inserted to iliac crest & removed). Both groups received an SIJ injection with LA.|Pain score on 0-10 NRS at 6 months; no categorical outcome noted, but the authors reported those who achieved a ≥2-point reduction in pain score and those who reported they were "better or much better".|No significant difference in primary outcome or categorical outcomes between groups.|

Pain Medicine, 2025, Vol. 00, No. 0

pain relief threshold (with documented functional improvement) are recommended to select patients for SIJ fusion; Grade I recommendation for the number of blocks and designating a specific high cutoff percentage.

$\bullet$ For prognostic procedures before lateral branch RFA, since SLBBs have greater face validity to predict outcomes than IA (and possibly EA) SIJ injections, we recommend using a 50% cutoff for SLBBs. The rationale for this is that the pain relief from RFA is generally less than from prognostic injections because of the lower specificity of the latter; Grade B recommendation, moderate level of certainty.

$\bullet$ For designating an outcome as positive when performing a therapeutic SIJ complex procedure, the most recent literature and guidance from the IMPMPACT group and the FDA suggest using thresholds of ≥2 points or 30% improvement provided there are other objective measures of benefit (eg, satisfaction, improved quality of life, decreased resource utilization); Grade B evidence, moderate level of certainty.

Dissent: On this question, three societies—IPSIS, NASS, and the NASS Fellowship Consortium—dissented, recommending more stringent criteria: Dual blocks with ≥75% relief for SLBRFA, and ≥75% relief on two separate diagnostic injections for SIJ fusion.$^{289}$

016. Who are the best candidates for sacral lateral branch radiofrequency ablation?

Identifying the most appropriate patients for any interven- tional procedure is important.$^{290-293}$ Appropriate selection may be even more critical for procedures involving the SIJ complex due to its intricate anatomy and varied presenta- tions.$^{43,79}$ Numerous studies demonstrate marked referral map variability in selection criteria when investigating SIJ complex denervation.$^{19,20,294}$ It is likely that different areas of the SIJ complex contribute to these different referral pat- terns. Thus, given that SLBRFA does not disrupt all the affer- ent input from the SIJ complex, identifying which referral patterns are most conducive to SLBRFA could improve out- comes and promote economic savings.$^{119}$

The first consideration in identifying patients who may benefit from SLBRFA is to appropriately diagnose SIJ complex dysfunction as the source of LBP. Approximately 15%-30% of mechanical LBP predominantly below L5 can be attributed to SIJ dysfunction.$^{295}$ SIJ dysfunction can result from prior spinal fixation, pregnancy, true and apparent leg length discrepancy, trauma, systemic inflammation, natural degeneration, and other factors leading to abnormal SIJ motion, and joint and soft tissue injury.$^{1,2}$ After putatively diagnosing suspected SIJ dysfunction with physical exam maneuvers,$^{2}$ in an attempt to better identify which patients are more likely to achieve improvement following SLBRFA, studies have used both prognostic SLBBs with local anesthetic$^{294,296}$ and/or confirmatory IA SIJ injections. These differences are important because the sacral lateral branches innervate the dorsal ligament complex, posterior aspect of the SIJ capsule, and posterior bone structure of the SIJ, but not the anterior components of the SIJ complex.$^{117}$ In addition, identifying the ideal nerves to target and optimizing the cutoff threshold used to delineate a diagnostic or prognostic injection as “positive” have been the focus of

investigation.$^{119}$ Clinical and psychosocial factors have also been investigated.

Failure of non-invasive therapies

Treatment options for SIJ complex pain include conservative therapies such as PT, yoga, tai chi, chiropractic care, acupuncture, analgesic medications, as well as minimally-invasive IA, EA injections, SLBB, SLBRFA and SIJ fusion surgery.$^{297}$ The noninvasive modalities and joint injections can all be considered first-line therapies under different circumstances$^{2}$ and may be offered to the patient based on a shared decision model. A progression of SLBB followed by SLBRFA is typically offered after failure of noninvasive therapies.$^{298}$

General factors associated with decreased response to invasive procedures

In a multicenter study aiming to identify responders to various LBP procedures, Cohen et al. found that individuals with greater disease burden, depression, sleep dysfunction, com- comitant pain conditions, smoking history, and obesity were more likely to fail interventions, indicating a need to address these comorbid conditions prior to or concurrent with proce- dures.$^{73}$ Additional factors reported to be associated with poorer outcomes to invasive procedures include psychosocial pathology (poorly controlled depression or anxiety, poor coping skills, high levels of catastrophizing, untreated sub- stance use disorders, and sedentary life-style), greater disease burden (opioid use, numerous previous failed treatments, higher pre-procedure pain scores, and reduced physical func- tion), central sensitization, and work-related factors (disabil- ty, secondary gain, poor job satisfaction, performing manual labor).$^{38,293}$

Demographic and clinical factors

In their bi-institutional study evaluating outcome predictors for SLBRFA, Cohen et al. reported that sex, physical examination signs (including SIJ tenderness, positive Patrick's test, positive Gaenslen's test), previous surgery, etiology, concomitant medical conditions, and disability claims, were not predictive of outcome in 77 patients with SIJ complex pain.$^{119}$ One consideration for the lack of association between provocation tests and outcomes may be that provocation tests are more likely to reproduce pain from IA vs EA pathology, as most studies examining correlations between exam maneuvers and interventional procedures have used IA injections. Since SLBRFA reduces pain primarily from EA pathology, commonly used “SIJ provocation maneuvers” on physical examination may be less predictive of SLBRFA outcome. Most prior randomized studies have reported favorable results despite requiring one or no provocative tests as a criterion for enrollment.$^{21,22,27}$

Referral patterns and opioid use may also impact outcomes. Although not statistically significant, Cohen et al. noted a higher percentage of unsuccessful outcomes in patients with pain radiation below the knee and on chronic opioid therapy, which is consistent with studies evaluating RFA outcomes for other regions.$^{119,194,299,300}$ Since <30% of SIJ complex-associated LBP produces pain distal to the knee and there is a high false-positive rate for diagnostic injections, many of these patients may be misdiagnosed.$^{79}$ Baseline opioid consumption has been shown to decrease the likelihood of treatment success in other studies assessing the

Pain Medicine, 2025, Vol. 00, No. 0

effectiveness of RFA interventions for chronic LBP as well as other chronic pain conditions.$^{194,301,302}$

Pre-procedure pain intensity, duration of symptoms, and age 265 years also increased the odds of failure in the study by Cohen et al.$^{119}$ It is not uncommon for duration of symptoms, a hallmark of greater disease burden, to be negatively correlated with outcomes,$^{291}$ which can be a consequence of neuroplasticity or refractory psychosocial issues.$^{303,304}$ In a retrospective study by Amatto et al., the authors also found younger age to be associated with a higher odds for success for SLBRFA.$^{285}$ Whereas age has not traditionally been found to be a negative predictor for outcome,$^{291}$ it may play a larger role in SLBRFA outcomes given that older patients are more likely to have IA as opposed to EA sources SIJ complex pain. These results suggest that while no single demographic or clinical variable strongly influences outcome, selecting patients without risk factors for treatment failure may improve success rates.$^{118}$ One limiting factor is that many studies employed only anterior-posterior fluoroscopic views, raising questions regarding the accuracy of prognostic injections.$^{305}$

In contrast to the finding that older age is associated with SLBRFA treatment failure, a recent multi-center randomized study comparing standard medical management (SMM) to internally cooled SLBRFA reported that age ≥65 years was associated with reduced pain in both the SMM and internally cooled (ic) SLBRFA groups.$^{14}$ The study also reported a trend toward higher BMI being associated greater pain reduction. One possible explanation for this may be that posterior EA structures (innervated by the sacral lateral branches) contribute to a greater amount of total SIJ complex pain in patients with higher BMI. Although duration of symptoms has consistently been associated with reduced effectiveness, the investigators of this multi-center clinical trial found that patients with long-standing (mean duration >115 months) SIJ complex pain can be effectively treated with SLBRFA.$^{14}$

Prognostic/diagnostic procedures

Number and type of injections

The SIJ is complex with numerous potential pain generators. The sacral lateral branches do not run at a consistent depth or location on the sacrum. Therefore, single-site, single-depth SLBBs may fail to consistently anesthetize the nerve supply. The use of multi-site, multi-depth SLBBs may more effectively target the sacral lateral branches. Dreyfuss et al. reported the use of multi-site, multi depth SLBBs resulted in staining of 91% of targeted nerves in cadavers. The authors also reported a 70% clinical success rate, defined as no evoked pain following EA (interosseous and dorsal sacroiliac ligament) and dorsal IA SIJ entry when using this technique. Most patients, however, did report pain upon joint distension. Given these findings, the authors concluded that IA or single-site, single-side SLBBs may not reliably predict ideal candidates for SLBRFA compared to multi-site, multi-depth SLBBs.$^{117}$ Among RCTs, all three studies that used prognostic SLBBs reported positive outcomes at the primary endpoint.$^{14,16,22}$

Dual injections are thought to decrease false-positive rates and increase diagnostic accuracy, though a false-positive rate with dual injections ranging between 12.5% and 26% has been reported.$^{298,306,307}$ Most studies to date have used single injections or IA corticosteroid injections as selection criteria for SLBRFA, possibly leading to inappropriate patient selection.$^{298}$

Yet, the few studies that have been performed do not show that more rigorous selection criteria lead to better SLBRFA outcomes. Cohen et al. reported no correlation between a positive SLBB, multiple positive SIJ injections, or percent pain relief (50%-79% vs ≥80%) from diagnostic injection(s) and SLBRFA outcomes.$^{119}$ Several other studies also suggest no association between the use of multiple prognostic injections and SLBRFA outcomes.$^{294,308,309}$ In the recent study by Amatto and colleagues, the authors found no differences in outcome stratified by single or double blocks (SIJ or lateral branch), 2 vs 1 SLBB, and 50% vs 80% cutoff thresholds, though in unique analysis those who experienced <80% pain relief with only a single SLBB fared slightly worse.$^{285}$

Predictive modeling

Predictive modeling utilizing large sets of data is a growing resource that may help predict responders to pain procedures. The Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) guidelines highlight the association of certain phenotype domains (psychosocial, pain variability and quality, sleep and fatigue, quantitative sensory testing (QST) and sensory profiling, and condition pain modulation) as areas of interest when adapting predictive modeling profiles.$^{310}$

Summary:

* Older age (≥65 years), longer duration of symptoms, higher baseline pain levels, and opioid use are associated with negative outcomes; Grade C evidence, low level of certainty.

$\bullet$ Considering the innervation of the nerves targeted during SLBRFA, there is deductive evidence that anterior intra-articular pathology and, like nearly all pain therapies, poorly controlled psychopathology (eg, depression, catastrophizing) and central sensitization are associated with treatment failure; Grade C evidence, low-to-moderate level of certainty.

$\bullet$ Although studies have reported positive outcomes using both prognostic intra-articular and extra-articular injections, the preponderance of evidence supports the use of prognostic SLBBs to select SLBRFA candidates; Grade C-B evidence, low level of certainty.

Recommendation:

$\bullet$ Currently, there are no single, stand-alone clinical, imaging, or injection-based factors that strongly predict the outcome of SLBRFA; future studies will need to identify patient phenotypes of success and failure

Q17. What is the ideal technique and evidence for performing sacral lateral branch radiofrequency ablation?

Significant variation in the posterior sacral neuroanatomy poses challenges to successful treatment with SLBRFA.$^{113}$ There are numerous techniques described for SLBRFA treatment. The largest fluoroscopic anatomical study to date correlated the location of the sacral lateral branches of 40 hemipelves to the published SLBRFA techniques.$^{122}$ It found that standard monopolar techniques published at the time of the study did not reliably capture the sacral lateral branch nerves; thus, such techniques are not anatomically valid and

Pain Medicine, 2025, Vol. 00, No. 0

should not be used. Several bipolar strip lesion techniques, and one icRFA technique using new targets published in 2018$^{311}$ were considered anatomically valid, achieving expected 100% capture of the lateral branches in 95% or more of the specimens. The Simplicity probe (multi-electrode radiofrequency cannula) technique was not considered anatomically valid, as the described technique can result in the probe being too far from the periosteum where the nerves lie, thus eluding capture.$^{122}$ This is consistent with the fact that the only clearly negative sham-controlled trial of SLBRFA used this technology.$^{24}$

Sensory stimulation used to guide denervation during radiofrequency treatments has been described in clinical studies.$^{25,312}$ However, this technique is not described in fluoroscopic anatomical detail, and thus, cannot be validated anatomically. There is clear evidence of lack of sensitivity of electrical stimulation for identification of electrode contact or close proximity to sensorimotor nerves.$^{313}$ In the lumbar spine where there is usually one medial branch being targeted; two prospective studies provide conflicting evidence for a correlation between sensory stimulation and lumbar facet RFA outcomes.$^{314,315}$ For SLBRFA, where multiple nerves carrying sensory information can converge on a single foramen, obtaining concordant sensory stimulation for a single nerve even at low thresholds cannot guarantee interruption of nociception.

In animal and ex vivo models, the size and type of RFA probe tip used and the lesioning parameters influence the size and shape of the ablation zone formed, which are critical in determining procedural success.$^{269,316,317}$ To date, however, most ablation zone measurements of RFA techniques have been made using ex vivo chicken breast or bovine liver models.$^{269,316,317}$ Ex vivo study of RFA lesion characteristics is intrinsically limited by lack of blood perfusing through the subject tissue. Chicken breast and bovine liver are homogeneous tissues without adjacent osseous structures, which makes them suboptimal surrogates for predicting in vivo response to SLBRFA. In vivo tissue effects of icRFA and standard monopolar RFA procedures in rats have been studied with MRI prior to sacrifice and verification through histological analysis.$^{318}$ Structural changes with icRFA may outlast those lesions created with conventional RFA. This has been shown in retrospective data analyses but has not been formally studied in clinical trials, though one study found that among successful SLBRFA procedures, there was a slower success drop-off rate with icRFA compared to conventional.$^{319,320}$

A small, in vivo human study of icRFA MRI lesion size showed significant lesion morphology variability at the same targets and between different target locations.$^{321}$ Notably, the icRFA coagulation zone volume was significantly smaller at periforaminal sacral locations compared to lumbar medial branch target locations [0.6915 (±1.08) vs. 1.685 (±2.51) cm$^{3}$] despite using the same size probe tip (4 mm) and lesion parameters. It was determined that icRFA probe placement very close to bone negatively altered the shape and reliability of the ablation zone. Although this study was small, it is known that bone disperses radiofrequency energy, with these results providing a potential explanation for inconsistent results of SLBRFA outcomes associated with icRFA and other RFA techniques.

As noted above, icRFA, especially using the targets defined in 2018,$^{122}$ and several bipolar strip lesion techniques appear to be the most anatomically-valid SLBRFA techniques, though these findings are based on ex vivo lesion size data, with limited clinical data to confirm or challenge these assumptions. The theoretically increased capture rate, and ensuing improved effectiveness, are particularly important given the anatomical variability in the number and location of sacral lateral branches. One non-randomized study compared treatment outcomes following SLBRFA using a bipolar strip lesion technique to a historical comparison group who had received icRFA.$^{322}$ The investigators observed a higher proportion of treatment responders (based on at least 50% reduction in NRS pain score) in the bipolar strip lesion group compared to the icRFA group for up to 12 months post-procedure. In two additional studies that compared SLBRFA outcomes stratified by RF electrode type, one study found a trend for superiority of icRFA over conventional SLBRFA, while the other found better outcomes with an 18-gauge quadripolar electrode inserted perpendicularly compared to a 16-gauge longitudinal approach and an 18-gauge bipolar perpendicular approach, which was attributed to an enhanced ability to navigate and accommodate the undulant lateral sacral crest topography.$^{119,285}$ This is consistent with studies for other conditions demonstrating that larger lesions lead to better RFA outcomes.$^{319,323,324}$ For the sacral lateral branches, the proximity of the nerves to the undulating ossous surface of the sacrum and thick dorsal ligaments requires further in vivo study to determine the best RFA technique(s) for successful nerve capture.

Clinical studies use heterogeneous selection criteria, which limit comparison of SLBRFA techniques to a degree. Due to several factors including selection and RFA technique heterogeneity, a recent systematic review downgraded the original “high” quality evidence rating to “moderate” quality evidence that SLBRFA provides clinically-significant reductions in pain and disability for up to one year.$^{121}$ The authors of this review then examined the data for only anatomically valid techniques. There were three sham-controlled RCTs. The quality rating was downgraded one level due to concern for bias and heterogeneous selection criteria but upgraded by one level for large attributable effect sizes. The authors determined there was high-quality evidence that SLBRFA effectively reduces pain and disability for up to one year when using anatomically-valid periforaminal icRFA and bipolar strip lesioning techniques. Figure 2 illustrates three among many different RFA techniques used to target the innervation of the SIJ.

Summary:

$\bullet$ For the anatomically-valid techniques (including specifically icRFA monopolar lesion techniques and bipolar strip lesion techniques), SLBRFA improves pain and disability for at least 6 months post-treatment; Grade B evidence, moderate level of certainty. $\bullet$ Without excluding anatomically-invalid techniques, the evidence for improving pain and disability is Grade C, low level of certainty. $\bullet$ Given differences in outcome likely related to neuroanatomical variability, there is direct and indirect evidence

Pain Medicine, 2025, Vol. 00, No. 0

supporting techniques that create large lesion sizes; Grade B evidence, moderate level of certainty.

Q18. What are the most common complications with sacral lateral branch radiofrequency ablation and how can they be minimized?

Potential and reported complications of sacral lateral branch radiofrequency ablation

SLBRFA has been shown in numerous studies to improve quality of life in well-selected patients if performed safely and in a technically sound manner. Because it is elective, aggressive prevention of complications is paramount. Adequate data collection regarding complications and research on preventing complications is critical to establishing and maintaining patient and clinician confidence in the safety of this procedure.

Data on complications from SLBRFA are usually collected and reported as part of clinical studies. Although common complications are usually self-limiting, rarer but more serious or even catastrophic complications may need to be garnered from sources such as large databases, quality and safety reports (ie, from the Food and Drug Administration) or even from autopsy reports. In theory (never before reported), and based on proximal anatomy, such complications could include rectal perforation or injury to nerves of the sacral plexus innervating the bladder or bowel.

In addition to the theoretical potential for injury to the sacral plexus and rectum, SLBRFA may cause predictable complications such as RFA of other peripheral nerves, skin burns, infection, hematomata, localized neuritis, sensory disturbance at the skin due to denervation of cutaneous nerves branching from sacral lateral branches, and a localized increase in persistent pain following thermal lesioning.

A systematic review by De Salvatore et al. reported on complications in the analyzed studies that included the largest and most rigorous RCTs performed, finding that the majority related to post-procedure local complications such as increased localized pain, transient paresthesias and hematoma.$^{325}$ No serious complications were reported.

Specific complications

Neuritis and other forms of post-procedural pain

As noted above, neuritis is a risk for all ablative procedures and may present as worsening pain of neuropathic quality. In some cases, it can be difficult to distinguish from inflammatory post-procedure pain. In a placebo-controlled trial evaluating icRFA in 28 patients, Cohen et al. reported that a majority of patients experienced temporary worsening of pain lasting between 5 and 10 days attributed to either the procedure itself or temporary neuritis, though only one reported paresthesias.$^{21}$ This study did not use preventative corticosteroids or non-steroidal antiinflammatory drugs before or after lesioning. In a large, multicenter, randomized trial comparing icRFA to standard medical management that did not report on the procedural use of prophylactic corticosteroids, Cohen et al. reported that 11 of 96 patients experienced worsening pain in the lower back or around the SIJ complex, but the presence of neurological symptoms was not reported.$^{14}$ In a small, retrospective study by Hegtary evaluating slBRFA using a 10-cm single mono- and bipolar electrode in 11 women, two patients reported new paresthesias surrounding the SIJ complex, while a third patient reported a

burning sensation 3-months post-procedure that responded to infiltration with local anesthetic and corticosteroid as well as topical lidocaine.$^{326}$ The authors recommended post-procedure bupivacaine and corticosteroid as a possible preventive measure to reduce the incidence of neuritis and post-procedural pain. The prevention of neuritis has been evaluated in several studies, most for lumbar medial branch RFA. Although one small, randomized trial found that post-RFA pentoxifylline or corticosteroids reduced the severity of post-procedure pain compared to saline,$^{327}$ a larger retrospective study failed to find benefit for post-lesioning cortico-steroid injection.$^{328}$ In the only study that included patients who underwent SLBRFA with icRFA technology (n = 2 out of 39 patients and 47 procedures), Fitzpatrick et al. found a 0% incidence of neuritis based on numerical rating scale pain scores and the administration of Douleur Neuropathique 4 Questionnaires (DN4) in both patients who received post-procedure corticosteroids (betamethasone) at ablation sites and those who did not.$^{329}$ Although the DN4 score was lower in the non-corticosteroid group (0.1 vs 0.4), both values fell far below the 4-point cutoff required to detect neuropathic pain, and this difference was no longer statistically significant when accounting for repeat procedures. A 2025 systematic review that included two studies not limited to any region concluded that post-RFA corticosteroid administration has not been demonstrated to reduce the risk of post-procedure neuritis.$^{330}$

The effectiveness of adjuvants and non-steroidal anti-inflammatory drugs to prevent post-RFA neuritis has been reported, also with mixed results. A large retrospective study performed in 215 patients who underwent lumbar and cervical medial branch RFA found a non-significantly lower incidence of neuritis in individuals taking gabapentin for at least 2 weeks prior to the procedure (7.1% vs 13.2%).$^{331}$ However, a later retrospective study by Singh et al. conducted in 164 patients found no difference in the incidence of neuritis post-lumbar medial RFA based on whether patients were receiving pre-procedure membrane stabilizers and antidepressants.$^{328}$ In a placebo-controlled study performed by Ma et al. in 66 patients, the authors found that 3- and 7-day post-treatment regimes with diclofenac significantly reduced post-procedure pain compared to placebo.$^{332}$ Aside from the small, underpowered retrospective study by Fitzpatrick et al.,$^{329}$ a major limitation in most of these studies was the failure of the investigative team to discriminate between neuritis secondary to nerve ablation and inflammatory post-procedure-related pain from radiofrequency cannula insertion.

Impact of radiofrequency ablation on existing implanted devices

Monopolar RFA of innervation to the SIJ complex involves the use of electrical current that emerges from the tip of the radiofrequency cannula and flows through the body before exiting through the dispersive electrode (a.k.a., grounding pad). Electromagnetic fields created from the use of short-wave diathermy can result in interference with the functioning of implanted electrical devices (eg, battery and electrical leads) such as deep brain or spinal cord stimulators.$^{333}$ A retrospective chart review examining interference of implantable devices following lumbar medial branch, peripheral nerve, and SIJ complex monopolar RFA found no alteration in implantable device function.$^{334}$ Although this single study

Pain Medicine, 2025, Vol. 00, No. 0

did not discuss the location of the implantable device, the safe use of monopolar lumbar medical branch RFA has also been reported in a patient with deep brain stimulators, with one of the implanted pulse generators located in the anterior abdominal wall.$^{335}$ Similar to lumbar medial branch nerve RFA, a procedure in which the grounding pad is usually placed on the lower extremity, SLBRFA theoretically carries a lower risk in the case of an implantable device placed in the torso or chest as the dispersive pad can be placed at a distance greater than the recommended 15 cm away from the leads and battery.$^{336}$ It may, however, present a greater risk of damage to implanted device placed in the low back or below the iliac crest as many spinal cord stimulators are implanted in the gluteal region, meaning the current will likely pass near the implantable device before exiting at the grounding pad. If using monopolar RFA, placing the grounding pad close to and caudal to the RFA site will reduce the size of the induced electromagnetic field. This will minimize the risk of heating the neurostimulator battery and electrodes. However, placing the grounding pad too close to the RFA site can increase the risk of tissue burns,$^{337-339}$ particularly when using high current, long activation times, and the use of conductive fluid, since the energy has less tissue through which to dissipate.$^{340}$ There is some evidence that bipolar RFA may be safer than monopolar RFA in patients with implanted devices, with report of its safe use for cervical and lumbar medial branch RFA.$^{341}$ In bipolar RFA, the electromagnetic field is limited to the tissue region between the two electrodes, which is significantly smaller compared to the distance between a monopolar RFA electrode and the grounding pad (when the grounding pad is placed at a safe distance away from the RFA electrode side to prevent a skin burn). In a position paper by IPSIS, the authors recommended consultation with the device manufacturer and cardiologist or electrophysiologist, with consideration of the following: (1) Having on-site support for interrogation of the cardiac device during the procedure in the event that reprogramming of the device is required; (2) Placing a magnet over the device to inhibit triggering the device by RFA; and (3) Removing the magnet or using external defibrillator/pacing electrodes in case cardiac arrhythmias occur during the RFA procedure.$^{342}$

Radiofrequency ablation in proximity to surgical hardware

In the context of lumbosacral or SIJ fusion, the intervention-alist may encounter posterior surgical hardware during a SLBRFA procedure. Indeed, SIJ complex pain is more prevalent in patients with arthrodesis of L5 to the sacrum, as the SIJ complex becomes the “adjacent segment” accepting greater forces post-fusion compared to when the L5-S1 seg-ment (or additional lumbar segments) remains mobile. As such, the interventionist must be aware of safety considerations when SLBRFA is performed in this context, as an RFA electrode may be placed in close proximity to posterior surgical hardware.$^{343}$ Cadaveric and human studies have demonstrated that direct contact of an RFA electrode with pedicle screws results in heating$^{344,345}$ of metal. In a 2015 cadaveric study, investigators observed significant temperature changes in pedicle screws and surrounding soft tissues when a RFA electrode directly contacted a screw, compared to control conditions. The thermal impact was more pronounced with titanium screws than with stainless steel. Specifically,

titanium screw heads exhibited temperature increases of 6.66°C, and screw tip temperatures increased by 16.72°C when the RFA electrode was positioned directly over the screws and heated to 80°C for 90 s. Alternatively, stainless steel screw heads demonstrated temperature increases of 4.4°C at the head and 2.46°C at the tip under similar conditions.$^{344}$ A 2016 human study evaluated temperature changes at posterior surgical hardware during lumbar medial branch RFA procedures.$^{345}$ A temperature probe was inserted via a cannula adjacent to posterior hardware to monitor temperature changes during each RFA procedure. The investigators observed that pedicle screw head temperatures increased in six out of the 10 RFA procedures. Notably, in two cases, temperatures escalated to 42°C, necessitating termination of the procedure. The clinical implications of this depend on hardware location. If hardware is contained completely within corticated bone, it is unlikely that adjacent soft tissues (including neural structures) would be heated. However, in cases of hardware malpositioning (for example, a pedicle screw that breaches the spinal canal), heating of the hardware may be of concern. Separately, in a patient with minimal tissue (either due to intrinsic body habitus or weight-loss following surgery) separating the head of a pedicle screw from the skin, screw heating could cause a skin burn. Notably, this risk remains theoretical (never before reported in the published literature). Regardless, in order to optimize patient safety, and consistent with IPSIS recommendations,$^{346}$ interventionists should avoid direct contact between the RFA electrode and posterior hardware.

Bleeding and anticoagulation

As noted above, the risk of clinically-relevant bleeding is exceedingly low, even with the large cannulas associated with icRFA and single-cannula combination mono- and bipolar lesions. The ASRA guidelines do not directly address anticoagulation and SLBRFA, but do not recommend withholding anticoagulation for low-risk procedures such as SLBBs or IA SIJ injections. However, the ASRA and ASIPP guidelines currently consider lumbosacral RFA and lumbosacral transforaminal injections to be intermediate risk; therefore, if finder needles are used during a SLBRFA to locate the foramina it may be considered an intermediate-risk procedure.$^{347,348}$ For physicians considering use of finder needles, this theoretical increase in risk should be weighed against possible enhancement in procedural accuracy. When performing a SLBRFA without a finder needle, it can be considered the same risk as a peripheral joint or nerve block procedure. The ASRA guidelines generally recommend stopping anticoagulants, including non-steroidal, anti-inflammatory drugs (NSAIDs), before intermediate-risk procedures, while the ASIPP guidelines tend to be slightly more liberal (ie, stopping NSAIDS and aspirin is a shared decision based on weighed risks and benefits). Both sets of guidelines recommend stopping warfarin and platelet aggregation and thrombin inhibitors. The risk of periprocurdural thromboembolic events must be weighed against the risk of epidural hematoma. When anticoagulants are held before pain procedures, the risk of a serious thromboembolic event is estimated at between 0.2% and 0.5%,$^{349,350}$ far exceeding the risk of a serious complication from bleeding.$^{348,350}$ There are very few (<5) reports of hematoma (several with unclear etiology) following lumbar transforaminal

Pain Medicine, 2025, Vol. 00, No. 0

epidural corticosteroid injections, and none during S1 or S2 injections, or that resulted in permanent neurological sequelae.$^{351-355}$ Alternatively, for lumbar and sacral RFA procedures, there are no reported cases of epidural or foraminal hematomas, and no reports of serious sequelae from bleeding complications. Furthermore, guidelines from IPSIS conclude that the risks clearly outweigh the benefits from withholding anticoagulants for lumbosacral RFA and transforaminal procedures based on evidence from large cohort studies.$^{349,350,356-359}$ As such, the risk: benefit ratio favors continuing rather than withholding anticoagulant and anti-platelet agents for SLBRFA procedures when technically sound and safe technique is implemented.

Prevention of complications

Safety precautions for SLBRFA based on synthesis of recommendations from reviews cited above:

Preoperatively

$\bullet$ Use an adequate consent process to advise patients of common, non-serious, but painful complications including transient localized increase in pain, painful paresthesias, or hematomata.

$\bullet$ Manage anticoagulant and antiplatelet risks on an individual patient level following a shared decision-making model, favoring continuation of these agents (ie, exceptions can be considered when multiple finder needles placed into the foramina are needed).

* Review existing imaging preoperatively to identify any anatomical variations which may require accommodation during the procedure.

* RFA can be performed safely in patients with SCS, cardiac pacemakers and surgical metalware if direct contact with the implant is avoided. Use of bipolar rather than monopolar RFA may be considered to limit the size of the electromagnetic field, which might disrupt or heat an implanted device in close proximity.

* Patients with impaired glucose tolerance should be warned of the potential for transient hyperglycemia if corticosteroids are used to prevent neuritis.

Intraoperatively

$\bullet$ Use sound, aseptic technique.

$\bullet$ Optimize fluoroscopic imaging (typically by introducing relative cephalic tilt, and possibly ipsilateral tilt at S1) such that the sacral foramina can be visualized adequately, and thus needle trespass can be avoided.

$\bullet$ Maintain a safe distance between the lateral margin of each sacral foramina and the RFA needle/electrode during the entire procedure.

$\bullet$ Sensory testing has limited safety value or therapeutic value when an aggressive lesioning strategy accounting for multiple lateral branches is employed. Its use is at the discretion of the interventionalist. Motor stimulation testing can be used if there is concern for foraminal proximity that could result in the heating of the neural elements within the foramen during lesioning. If employed, place hand on or clearly observe the lower extremity during motor stimulation to ensure there is no muscular contraction in the lower leg. Stimulation of sacral nerves

innervating the bowel and bladder may be ascertained via patient communication.

$\bullet$ Ensure a grounding pad has been correctly placed in full contact with a large area of the skin remote from the lesion site.

$\bullet$ Routinely check lateral radiographic views to ensure that the RFA probes remain on the dorsal surface of the sacrum, avoiding the sacral plexus and rectum.

$\bullet$ Avoid direct contact of the RFA probe with posterior sur- gical hardware in relevant situations.

Summary:

$\bullet$ There is weak evidence supporting non-steroidal anti-inflammatory drugs to prevent or treat neuritis; Grade C evidence, low level of certainty.

$\bullet$ There is insufficient evidence for the use of corticosteroids and membrane stabilizers, which themselves carry risks, in the prevention or treatment of neuritis; Grade I evidence.

$\bullet$ In patients on anticoagulants, consultation with an anti-coagulation specialist or review with the patient's prescriber should be conducted to determine the risk of stopping anticoagulants. However, in most cases, anticoagulants may be continued during SLBRFA; Grade C-B recommendation, moderate level of certainty.

$\bullet$ SLBRFA can be safely used in patients with implanted devices, with bipolar lesioning conferring an increased measure of safety; Grade B recommendation, moderate level of certainty.

* Radiographs in multiple views should be checked to ensure that RFA electrodes avoid contact with surgical hardware; Grade B recommendation, moderate level of certainty.

$\bullet$ Sensory stimulation confers minimal safety value, and limited therapeutic value when employing an aggressive lesioning approach that accounts for multiple lateral branches; Grade C evidence, low level of certainty.

* Motor stimulation can be used in conjunction with observation of lower extremity movement and patient feedback (eg, to detect bowel or bladder sensations) to reduce the risk of collateral nerve injury; Grade C recommendation, low level of certainty.

Surgical treatment

019. How effective is surgery for the treatment of sacroiliac joint complex pain?

One of the earliest descriptions of surgery for SIJ pathology was by Smith-Peterson and Rogers in 1926.$^{360}$ The authors of that study suspected that the SIJ was an underappreciated anatomic site of pain, and that better understanding was needed. It was not until almost a century later that the number of publications describing surgical treatment of SIJ pathology significantly increased. Open fusion techniques involve either anterior or lateral approaches to the SIJ. More recently, minimally-invasive lateral and posterior techniques have been introduced to achieve arthrodesis (see Figure 3 illustrating 2 locations for minimally invasive rod placement). The following is a summary of available literature evaluating the efficacy and effectiveness of surgery for SIJ complex pain.

Open surgery

Early indications for open surgery included history and physical (typically at least three positive provocative tests), and

Pain Medicine, 2025, Vol. 00, No. 0

 subsequent studies added positive response to at least one local anesthetic SIJ injection.

Sham-controlled randomized controlled trials

We identified no published RCTs comparing open surgery for fusion of the SIJ to sham surgery.

Pragmatic randomized controlled trials

We identified no published RCTs comparing open surgery for fusion of the SIJ to an active treatment.

Observational studies

We identified four cohort studies. In 1926, Smith-Peterson and Rogers described an open, lateral approach to fuse the joint with bone transport from ilium through the SIJ in 26 patients.$^{360}$ Twenty-five of the 26 patients returned to their original occupation. Two were deemed treatment failures due to continued postoperative pain. Buchowski et al. reported on a series of 20 patients who underwent open anterior plating, in which all patients demonstrated statistically significant improvement in all SF-36 categories—except for General Health and Mental Health, and 17 of the 20 had radiographic evidence of joint fusion.$^{361}$ Sixty percent of the cohort stated that they would choose to have the surgery again. Slinkard et al.$^{362}$ and Murakami et al.$^{363}$ published case series of open, anterior plating for fusion of the SIJ. In the former, the average ODI in 19 patients significantly decreased by 17.9 ± 25.4 (P = .007).$^{362}$ The open plating technique described by Murakami et al. resulted in a mean change in VAS pain score in 21 patients from 84.4 to 26.6 (P < .001), and the Roland Morris disability questionnaire score for 17 patients decreased from 21.2 to 7.2 postoperatively (P < .001).$^{363}$

We found two retrospective case-control studies comparing open anterior fusion to lateral minimally-invasive surgical (MIS) fusion. In a study involving 263 patients, Smith et al. found that 82% patients achieved a clinically significant pain score improvement (2 points on a 0-10 NRS) at 24 months in

the lateral MIS group compared to only 50% in the open group.$^{364}$ Ledonio et al. reviewed 63 patients with a minimum 12-month follow-up after SIJ fusion using an open anterior vs an MIS lateral approach. The authors reported statistically significant improvements in ODI in both the open and MIS groups. The average change in VAS pain score at 12 months was -2.7 in the open group and -6.2 in the MIS group.$^{365}$ Both studies reported increased blood loss, operative time and hospitalization length in the open anterior approach group compared to the lateral approach group.$^{364,365}$

Minimally-invasive lateral approach

Criteria for surgery in the following studies included history and physical (typically at least 3 positive provocative tests), and many modern studies added positive response to at least one local anesthetic SIJ injection.

Sham-Controlled randomized controlled trials

We identified one single-blind, sham-controlled study (n = 63) evaluating percutaneous lateral SIJ fusion by Randers et al. that demonstrated no statistically significant improvement in pain scores at 6-months post-procedure.$^{288}$ The sham group was placed under anesthesia, an incision made, and the procedure was “simulated” without implant placement. The average improvement in VAS pain score for the treated group was 2.6 vs 1.7 in the sham group. Patient-reported impression of change revealed that 45% of participants in the operative group felt they were “better” or “much better” after surgery vs 31% in the sham group. Other outcomes reported included average ODI in the surgery group of 47% vs 50% in the sham group, and a “Timed Up and Go” test of 10.6 s in the surgery group vs 11.7 s in the sham group. None of these differences were statistically significant.

Pragmatic randomized controlled trials

We identified four publications stemming from the iMIA study (iFuse Implant System Minimally Invasive Arthrodesis),

Pain Medicine, 2025, Vol. 00, No. 0

a RCT comparing lateral MIS fusion to nonsurgical management for refractory SIJ complex pain involving nine European medical centers. In the first publication, Dengler et al.$^{366}$ reported on improvement in referred leg pain (RLP) related to SIJ dysfunction over 6 months in patients conservatively managed or treated by MIS SIJ fusion. Overall, there was a 76.2% prevalence of RLP in the study population. After 6 months, the VAS pain score (0-100 mm) in the operative group decreased from 58 to 13.5 but was unchanged in the conservatively managed group. The 6-month outcomes for the primary outcome measure, self-reported reduction in LBP, were reported in 2017.$^{287}$ The VAS pain score in the operative group decreased from 77.7 to 34.4, which favorably compared to a reduction from 77 to 67.8 in the CM group (P < .001). The difference in ODI score at 6 months was 19.8% (P < .001), with the SIJ fusion group also reporting less disability. One-year data from the iMIA trial was reported by Dengler et al.$^{367}$ on 52 participants at 12-month follow-up, while the fourth publication reported on 103 participants at 24-month follow-up.$^{368}$ Treatment success, defined as a 20-point or greater improvement in VAS pain score (0-100 scale), occurred in 79% (37 of 47) of the operative patients, but fewer than 25% (11 of 46) of conservatively treated participants at the 24-month endpoint. Statistically significant improvements in mean VAS pain and ODI scores were reported in the SIJ fusion compared to the nonsurgical management group at both 12 and 24 months. Of note, the publications resulting from these studies were sponsored by industry, increasing the risk of bias.

There are two published studies for the INSITE trial, which was a prospective, multicenter, parallel group, unblinded RCT comparing minimally-invasive fusion with triangular titanium implants to conservative management. The 12-month outcome study of 148 subjects (102 with fusion, 46 with CM) revealed a clinically important improvement in ODI (≥15%) in 73.3% of surgical patients vs only 13.6% in the CM group, which was sustained until the end of trial at 12 months.$^{369}$ EuroQol-5D, a validated instrument evaluating generic quality of life, improved by 0.29 points (P < .001) in the SIJ fusion group but only 0.06 points (P < .17) in the NSM group (P < .001) at 6 months. Thirty-five of 46 patients in the CM cohort crossed over at 6 months, and their collective improvement after surgery mirrored that in the operative group. The 24-month outcomes were similar, with 83.1% of operative patients experiencing clinical improvement in their VAS pain scores vs < 10% in the CM group (P < .01).$^{146}$ The ODI scores at 24 months were also much better in the SIJ fusion group vs the CM group (68.2% improved >15% vs 10%). Similar to the iMIA study, the publications resulting from these studies were sponsored by industry.

Observational studies

Numerous published series are available regarding the effectiveness of the lateral MIS fusion approach. Details relevant to each of these studies are provided in Table 19. A retrospective study of 137 patients compared outcomes of conservative medical management vs SIJ complex denervation by SLBRFA vs the lateral MIS fusion approach with 6-year follow-up.$^{370}$ Patients were selected in this study based on at least 3 of 8 positive provocative SIJ tests and a positive response to an IA injection. The authors reported minimal improvement in pain

and function at long-term follow-up with conservative medical management, an intermediate response to denervation by SLBRFA, and significant improvements in VAS pain and ODI scores in the surgical group. The surgical group was more likely to be working (70.4%) than the conservatively managed (19%) and SI denervation (34%) groups (P < .01 for both). Similarly, the percentage of patients taking opioids in the surgery group was initially 63%, and decreased to 7.4% at last follow-up, which favorably compared to the conservative management only group (49% to 82%) and the SI denervation group (55% to 85%), which both increased usage of opioid medications. A caveat to this study is that while IA injections are used to identify patients with degenerative joint changes—an indication for surgery—they have not been validated to identify EA posterior ligamentous pathology, which is the primary indication for denervation. Multiple case series (summarized in Table 19, section 2c) provide additional evidence of statistically significant improvements in pain relief and/or functional measures with operative treatment.$^{145}$,$^{371}$⁻$^{14}$ In a retrospective study comparing cylindrical threaded implants to triangular titanium implants, both groups experienced a statistically significant improvement in VAS pain and ODI scores at 6 months when compared to their preoperative scores.$^{372}$ Of note, most of the available literature for MIS lateral fusion involves the use of triangular titanium implants, with most publications disclosing sponsorship by industry.

Minimally-invasive posterior approach

Criteria for surgery in the following studies included history and physical (typically at least three positive provocative tests), with most recent studies requiring a positive response to at least one local synthetic IA SIJ injection.

Sham-Controlled randomized controlled trials

We identified no published RCTs comparing the posterior MIS approach for fusion of the SIJ to sham surgery.

Pragmatic randomized controlled trials

We identified no published RCTs comparing the posterior MIS approach for fusion of the SIJ to an active treatment.

Observational studies

Details relevant to each of these studies are provided in Table 19. In brief, we identified a case series of 19 patients treated with a posterior distraction screw. The authors reported an improvement in VAS pain intensity (mean preop 8.5 to 6 postoperatively) and ODI (64.1 to 57.0%, P =.19) scores at a mean follow-up of 13 months. Additionally, 79% of patients indicated that they would “go through the procedure again in order to achieve their current results.”$^{383}$ Another report of 171 patients with a posteriorly placed MIS SI distraction screw demonstrated improvement in VAS pain and ODI scores over the 24-month study period.$^{384}$ Further, 73.3% of patients reported they felt better or much better at 24 months than prior to surgery, and when asked if they would recommend the surgery to a friend, 79% said “yes” at 12 months, and 78.5% replied affirmatively at 24 months. The results of five additional studies investigating the effectiveness of a specific posterior SIJ stabilization system have been published, all funded by the device manufacturer. Deer et al. reported results on 111 participants, of whom 47% experienced at least 80% pain relief at an average follow-up of 12 months.$^{385}$ A second

|Authors|No. of patients|Fusion method|Definition of success and primary outcomes|Other results| |---|---|---|---|---| |1. Open Surgery (Design) Buchowski, $2005^{361}$ (Observational)|20|Anterior SIJ plate|Radiographic fusion-17/20|60% would have surgery again; Statistically significant improvement occurred in all SF-36 categories except for General Health and Mental Health;| |Ledonio, $2014^{365}$ (Retrospective comparative-effectiveness)|63 patients (open: 36; MIS: 27) w/ minimum 1-yr f/u|Open ant plate vs TTI|Compared EBL, OR time, LOS, ODI between approaches|Open surgery patients had higher EBL, LOS and OR time than MIS patients; ODI scores not different between groups| |Murakami, $2018^{363}$ (Observational)|27 pts followed for at least 5 years|Anterior SIJ plate|Modified McNab Criteria: 21 unilateral pts-7 had excellent, 11 good, and 3 had fair outcomes; 4 ant and post fusion-1 excellent, 1 good, 1 fair, and 1 poor outcome; 2 pelvic ring fusions-1 good and 1 poor outcome.|Union on CT 27/27 at 1yr; in 21 unilateral pts, VAS pain score decreased from 84.4 to 26.6| |Slinkard, $2013^{362}$ (Observational)|19 pts (12 with lumbar fusion, 7 without), with at least 12-month f/u|Open ant plate|ODI and SMFA: Avg pre-op (63.1) and final (45.2) ODI and SMFA (decrease from 47.7 to 36.0) significantly changed at final follow-up; 12- and 24-month pain scores and operative parameters. MCID achieved in 86% and 82% in the MIS group at 12 and 24 months, respectively, vs 61% and 50% in the open group at 12 and 24 months|No significant difference for prior fusion patients in either measure.| |Smith, $2013^{364}$ (Retrospective comparative-effectiveness)|263 pts followed for a minimum of 12 months (149 open and 114 MIS)|Lateral open transarticular noninstrumented fusion|Undefined, but 23/26 (89%) experienced no pain at variable follow-up periods.|Open surgery required longer surgical time and was associated with greater blood loss and loss of sensation.| |Smith-Peterson, $1926^{360}$ (Retrospective)|26|Lateral open transarticular noninstrumented fusion|Undefined, but 23/26 (89%) experienced no pain at variable follow-up periods.|25/26 (96%) returned to previous occupation, 2/26 reported as "failures"| |2. Minimally-invasive Lateral Approach||||| |2a. Sham-controlled RCTs||||| |Randers, $2024^{288}$|63 (32 rec'd surgery vs 31 who rec'd sham), followed for 6 months|Sham vs TTI|≥2 pts on VAS pain scale (Surgical group 16/32, (50%); Sham group 13/31 (42%), P > .05|Avg VAS pain improvement at 6 months: 2.6 in surgery vs 1.7 in sham group; "better" or "much better" 45% in surgery vs 31% in sham group (P = NS)| |2b. Pragmatic RCTs||||| |Dengler, 2016 iMIA trial $^{366}$||CM vs TTI|||

|Authors|No. of patients|Fusion method|Definition of success and primary outcomes|Other results| |---|---|---|---|---| ||100 (49 rec'd CM vs 51 who rec'd TTI), followed for 6 months||Primary outcome referred leg pain (RLP). After 6 months, CM group had no meaningful change in RLP. RLP VAS pain score in TTI group decreased from 58 to 13.5.|Referred leg pain present in 76% of pts at baseline.| |Sturesson, 2017 iMIA trial $^{287}$|103 (51 rec'd CM vs 52 who rec'd TTI) followed for 6 months|CM vs TTI|Difference in change in self-rated LBP at 6 months. Mean LBP improved by 43.3 points in the SIJF group vs 5.7 points in the CM group ≥20-pt VAS (0-100) pain improvement: 78.8% in TTI vs 22.4% CM|Mean ODI improved by 26 percentage points in the SIJF group vs 6 points in the CM group| |Dengler, $2017^{367}$|103 (51 who received CM vs 52 who received TTI), followed for 12 months|CM vs TTI|≥20-pt VAS (0-100) pain improvement at 24 mo: 79% in TTI vs 24% CM|12-month VAS leg pain, ODI, EQ-5D all significantly better in TTI group; 21/49 (43%) of CM pts crossed over to TTI at 6 months and had similar results to TTI group| |Dengler, $2019^{368}$|103 (51 received CM vs 52 who rec'd TTI), followed for 24 months|CM vs TTI|≥20-pt VAS (0-100) improvement at 24 mo: 79% in TTI vs 24% CM|Mean VAS pain at 2 yrs improved by 45 in SIJF vs 11 in CM; mean ODI at 2 yrs improved by 26% in SIJ fusion vs 8% in CM; 12-month CT scan in 43/52 SIJF pts; 8 implants had radio-lucency (rather than bone apposition).| |Polly, 2015 INSITE trial $^{369}$|148 (102 rec'd TTI vs 46 who rec'd CM)|CM vs TTI|Improved VAS pain score and no serious adverse event or reoperation at 6 months: 81.4% vs 26.1% in CM)|At 12 mo, VAS SIJ complex pain score improved by at least 2 points in 81.6% of SIJ fusion patients vs 12.5% of CM patients. ODI improved by >15% in 72.4% of SIJ fusion vs 10% of CM group.| |Polly, 2016 INSITE trial $^{146}$|148 (102 who rec'd TTI vs 46 who rec'd CM)|CM vs TTI|SIJ complex pain improvement ≥20 points on pain VAS, ODI ≥15% over 2 years. In SIJF group, 83.1% improved vs <10% for CM group.|In SIJF group, 68.2% met ODI improvement threshold vs <10% in CM group.| |2c. Observational studies Al-Khayer, $2008^{371}$|9|Hollow modular anchorage screw|Change of at least 10% on ODI and 2 points on 0-10 VAS for pain Mean ODI decreased from 59 preoperatively to 45 postoperatively); Mean VAS (0-10) pain score decreased from 8.1 preoperatively to 4.6 postoperatively|Mean patients' satisfaction was 6.8 on a 0-10 VAS scale| |Claus, $2020^{372}$|156 (n = 74) with threaded implant and 82 with TTI)|Cylindrical threaded implant (CTI) vs TTI|VAS pain score, ODI, and Short Form-12 at 6 months and 1 year. No difference in outcomes between the groups at 6 or 12 months.|In both cohorts, there was a significant improvement in all outcomes at 6 months when compared with preoperative values.| |Cummings, $2013^{373}$|18 unilateral SIJF pts with 12 mo f/u|TTI|VAS ≥2pt: 90% at 1 yr; ODI ≥12.8%: 89% at 1 yr|(continued)|

Table 19. (continued)

|Authors|No. of patients|Fusion method|Definition of success and primary outcomes|Other results| |---|---|---|---|---| |Duhon, $2013^{375}$|94 (only 32 with 6-month f/u)|TTI|Success = VAS pain score Δ ≥ 2, no device complications, no neurological symptoms, and no reoperation: Successful outcome criteria met in 23/26 (88.5%); Success = VAS pain score decrease of at least 2 points, no device complications, no neurological symptoms, no reoperation: successful outcome criteria met in 23/26 (88.5%); Success = VAS pain score decrease of at least 2 points, no device complications, no neurological symptoms, no reoperation: Successful outcome criteria met in 127/159 @ 12 months (79.9%) and 119/149 (79.9%) at 24 months. VAS pain score ≤3 (0-10 scale): 9/29 (31%) had a successful outcome at 12 months|1/18 (5.6%) "not satisfied", 2/18 (11%) would "definitely not" have SIJF again Mean ODI score decreased from 55% to 38.9% at 6 months; 22/26 (85%) somewhat or very satisfied, 22/26 would have procedure again Pain improved from a baseline of 79.8 (12.8) points to 26.0 (26.7) points at 24 months; ODI improved from a baseline of 55.2 (11.5%) percentage points to 30.9 (20.5) at 24 months| |Duhon, $2016^{374}$|172 (149 with 24-month f/u)|TTI|Success = VAS pain score Δ ≥ 2, no device complications, no neurological symptoms, no reoperation: Successful outcome criteria met in 127/159 @ 12 months (79.9%) and 119/149 (79.9%) at 24 months. VAS pain score ≤3 (0-10 scale): 9/29 (31%) had a successful outcome at 12 months|Statistically significant change in VAS pain score and EQ-5D at 1-yr f/u (VAS score decreased from 7.83 to 4.97 and EQ-5D from 0.27 to 0.5; 23/29 (79%) "agreed" or "totally agreed" that their complaints were reduced after surgery; 24/29 (83%) would have surgery again would have surgery again| |Hermans, $2022^{145}$|29|TTI|Pain relief and normalization of SI tests-4/5 (80%) no pain|1/5 with no pain relief yet solid fusion of SIJ on CT| |Giannikas, $2004^{377}$|5|Open, lateral, Cloward bone plug fusion|Pain relief and normalization of SI tests-4/5 (80%) no pain|Avg. follow up 17 months; 2/15 with fair results had fusion on CT scan| |Khurana, $2009^{378}$|15|Hollow Modular anchorage screw|SF-36 and Majeed's functional scoring system; The mean SF-36 improved from 37 to 80 for physical function and from 53 to 86 for general health; mean Majeed's score improved from 37 to 79 postoperatively. 53% had a 20 pt or greater reduction in VAS (0-100 mm) back pain score at 1 yr; 69% had a 20 pt or greater reduction in VAS leg pain at 1 yr; 57% had a 15% or greater decrease in ODI. VAS (0-10) pain score: Mean VAS back pain score decreased from 6.6 to 3.7 Improvement in VAS pain score: Mean preoperative VAS pain score decreased from 8.7 to 0.9 at 12 months|Avg. follow up 17 months; 2/15 with fair results had fusion on CT scan Avg VAS leg pain score decreased from 63.6 to 27.7 at 12 mo; 88% had successful radiologic fusion at 12 months| |Kube, $2016^{379}$|18 pts w/ 20 SI fusions|Simmetry Screw(s)/ MIS lateral screw|53% had a 20 pt or greater reduction in VAS (0-100 mm) back pain score at 1 yr; 69% had a 20 pt or greater reduction in VAS leg pain at 1 yr; 57% had a 15% or greater decrease in ODI. VAS (0-10) pain score: Mean VAS back pain score decreased from 6.6 to 3.7 Improvement in VAS pain score: Mean preoperative VAS pain score decreased from 8.7 to 0.9 at 12 months|Avg VAS leg pain score decreased from 63.6 to 27.7 at 12 mo; 88% had successful radiologic fusion at 12 months| |Rajpal, $2019^{380}$|24 pts followed for an avg of 19 months|MIS Lateral Screw (Rialto)|VAS leg pain score decreased from 4.8 back pain score decreased from 6.6 to 3.7 Improvement in VAS pain score: Mean preoperative VAS pain score decreased from 8.7 to 0.9 at 12 months|VAS leg pain score decreased from 4.8 to 1.5; 88% would have surgery again| |Sachs, $2013^{381}$|40 pts with 1-yr follow-up after TTI|TTI|Improvement in VAS pain score: Mean preoperative VAS pain score decreased from 8.7 to 0.9 at 12 months|100% would have surgery again|

Pain Medicine, 2025, Vol. 00, No. 0

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

|Authors|No. of patients|Fusion method|Definition of success and primary outcomes|Other results| |---|---|---|---|---| |Schmidt, $2021^{376}$|19 pts (24 SIJF), followed for an avg of 58 months|TTI|Decrease in VAS pain scores and SF-36. Preoperative VAS score decreased from 7 to 3 (P = .001); SF-36 physical domain increased from 40 to 55 (P = .016)|All other SF-36 domains not significantly different| |Vanaclocha, $2018^{370}$|137 pts followed for an avg of 6 yrs (63 with conservative treatment, 47 who rec'd SI denervation, 27 who had SIJF)|CM vs Radiofrequency denervation vs TTI|VAS and ODI scores: After 4 yrs, no CM or denervation pt had ≥15% ODI improvement, all SIJF had ≥15% improvement|Pain relief in all groups at 1 month; however, relief was sustained in the SIJF group only| |Whang, 2019 INSITE $^{382}$|103 (93 available at 5-yr follow-up)|TTI|Study success defined as VAS pain score (0-100) improvement of at least 20 points in the absence of severe device-related or neurologic adverse event or revision surgery in 76 subjects at 60 months: 81.7% successful outcomes|Mean SIJ complex pain score decreased from 81.5 to 27.1; ODI decreased from 56.3 to 29.9; 85% said they would definitely undergo the procedure again at 24 months, 75% at 60 months| |3. Minimally-invasive Posterior Approach||||| |3a. Sham-controlled RCTs||||| |3b. Pragmatic RCTs||||| |3c. Observational studies||||| |Calodney, $2024^{387}$|122 pts|LinQ (bone allograft posterior SIJ stabilization system)|Responders met the following criteria: VAS pain score improvement ≥20 mm; Absence of implant-related serious adverse events; Absence of neurologic worsening related to the lumbosacral nerve roots Absence of surgical reintervention. Responder rate at 12 months (composite outcomes) when excluding lost to follow-up: 73%. Responder rate at 12 months (composite outcomes) when treating lost to follow-up as treatment failure: 54%.|When excluding participants lost to follow-up, at 12 months, 61% of participants reported at least 50% improvement in pain and 69% of participants reported an ODI improvement of at least 15 percentage points. In a worst-case scenario analysis, these proportions decreased to 46% for pain and 51% for ODI responders, respectively.|

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

|Authors|No. of patients|Fusion method|Definition of success and primary|Other results|| |---|---|---|---|---|---| |Authors|No. of patients|Fusion method|VAS pain score improvement at least 20 mm at 6 months: 47 ps|Avg. VA5 pain improvement was 34-9; avg. ODI improvement|Avg. VA5 pain improvement was 34-9; avg. ODI improvement| |Calodhey, 2022 386|69 pts|LinQ (bone allograft posterior SJ)|68%, ODI improvement of at least 15%, ODI improvement of at least 30% VA5 pain score improvement: ≥ 30% VA5 baseline: 39 ps (57%) ≤ 20% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 20% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39 ps (57%) ≤ 30% VA5 baseline: 39|||

CNY, N. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V.

Pain Medicine, 2025, Vol. 00, No. 0

study reported results at an interim 6-month time point. In this study 68% of participants experienced at least a 20-point VAS pain score improvement and 57% of participants reported an ODI improvement of at least 15 points at 6-month follow-up.$^{386}$ A third study (the final results) reported on 122 implanted participants at 12-month follow-up.$^{387}$ When excluding participants lost to follow-up, at 12 months, 61% of participants reported at least 50% improvement in pain and 69% of participants reported an ODI improvement of at least 15 percentage points from baseline. In a worst-case scenario analysis counting participants lost to follow-up as treatment failures, these proportions decrease to 46% for pain and 51% for ODI responders, respectively. One recent study evaluating the results of inferior-posterior transarticular MIS SIJ fusion in 36 patients selected based on positive response to an IA corticosteroid injection and at least 3 provocative tests reported a decline in average 0-100 VAS pain score from 79 mm to 23 mm at 12-month follow-up, and a decrease in mean ODI score from 52% to 21% at the same endpoint. The 12-month satisfaction rate in this study was 83%, with successful fusion confirmed in 83% of cases.$^{388}$ Finally, another study reported the 12-month outcomes of 50 participants following the posterior MIS procedure. At the end of the study, the authors reported an average pain score of approximately 4 points.$^{389}$

Database reviews

In one database review performed in 4666 patients who underwent minimally-invasive SIJ fusion between 2010 and 2022, the authors found that patients who were opioid-naïve before surgery required fewer postoperative opioids compared to chronic preoperative users, yet continued opioid use one year post-surgery was higher in the MIS SIJF group than in patients undergoing other spine surgeries.$^{390}$

Summary:

$\bullet$ An open approach to SIJ fusion can provide pain relief and functional benefit in carefully selected patients who are refractory to other therapies; Grade C evidence, low-to-moderate level of certainty.

* There is conflicting evidence (two positive pragmatic RCTs vs one negative sham-controlled RCT) that SIJ fusion via the minimally-invasive (MIS) lateral approach is effective at improving pain and function in carefully selected patients; Grade C evidence, very low level of certainty.

$\bullet$ The MIS posterior approach to SIJ fusion can be considered in patients with refractory intra-articular degeneration with instability who fail to respond to more conservative therapies; Grade C recommendation, low level of certainty.

Q20. How does one best select patients appropriately for sacroiliac joint fusion surgery?

Given the complex anatomy and variable presentation of SIJ complex pain, appropriate selection for surgical intervention is of upmost importance.$^{43,79}$ Surgical intervention for the SIJ should be reserved as a last treatment option, only after all conservative and less invasive interventional options have been exhausted.$^{391,392}$ In a subset of well-selected patients, SIJ fusion can provide pain relief and functional benefit with both open and minimally-invasive fusion techniques.$^{370,391-394}$ Minimally invasive SIJ fusions have been shown to be relatively safe with low revision rates.$^{392,395,396}$ Patient selection for surgical intervention should be based on a combination of clinical history, physical examination, and diagnostic SIJ injections,

after ruling out other sources of pain, and failing trials with nonoperative integrative, pharmacological and interventional treatments.$^{397}$

Clinical presentation and physical examination

Adequately identifying patients with SIJ complex pain by history and physical exam alone is challenging.$^{398}$ Whereas no single element of clinical history or physical exam finding is diagnostic of SIJ complex pain, there are some commonalities suggestive of pain originating from the SIJ.$^{66,140,282}$ Historically, patients describe unilateral pain below L5 vertebra, frequently present with a history of trauma, spondyloarthropathy, or lumbosacral fusion, and complain of pain aggravation with sitting or standing.$^{66}$ A combination of physical exam maneuvers can be more reliable than single tests for identifying possible IA SIJ-mediated symptoms.$^{140,282,399}$ Namely, three of six SIJ provocation maneuvers, including Patrick's or FABER, Gaenslen's, thigh thrust, sacral thrust, distraction, and compression should be positive, though recent studies have called into question the validity of physical exam findings.$^{73,74,140}$ In addition to identifying relevant history and physical exam findings to diagnose SIJ complex pain, other sources of pain such as radicular pain, discogenic pain, facetogenic pain, generalized pain disorders such as fibromyalgia, and autoimmune arthropathy should be ruled out.$^{400}$

Failure of nonoperative treatment

According to the International Society for the Advancement of Spine Surgery (ISASS), all patients who are considered for SIJ fusion surgery should have had at least 6 months of SIJ complex pain which impacts quality of life, and have failed nonsurgical treatments including PT, therapeutic exercise, and either an IA SIJ corticosteroid injection, or EA injection or SLBRFA if an EA component is suspected.$^{397}$

Factors associated with treatment outcome

In a secondary analysis from three RCTs comparing nonsurgical management (n = 97) to minimally-invasive fusion (n = 316), predictors of treatment failure after surgery were smoking, opioid use, younger age and shorter duration of pain, with the latter more common in individuals with EA than IA pathology.$^{401}$ Separately, the INSITE study reported no association between pain relief associated with the diagnostic IA SIJ injection and pain relief after SIJ fusion, with the parameters of the injection not noted.$^{146}$ In a presentation at the 2016 Congress of Neurological Surgeons meeting, decreased postsurgical dyspareunia (a possible symptom of nociplastic pain) and not taking opioids were the only factors associated with a successful SIJ fusion.$^{402}$

Radiologic findings

Imaging modalities such as plain radiographs, CT, and MRI can be useful for identifying sacroiliitis, SIJ arthropathy, or for monitoring treatment response in inflammatory spondyloarthropathy.$^{40,403}$ However, the absence of imaging abnormalities neither rules out the SIJ as a source of pain nor reduces the likelihood that patients have IA pathology amenable to fusion.$^{40,65}$ In one study performed in 18 patients with moderate to severe ankylosing spondylitis, no association was found between MRI abnormalities and back pain.$^{404}$ In the non-inflammatory spondyloarthropathy population, no imaging modality (including x-rays, CT, and MRI) has been shown to be superior to clinical selection to

Pain Medicine, 2025, Vol. 00, No. 0

diagnose SIJ complex pain, and no imaging modality has been shown to be superior to diagnostic SIJ injections in predicting responsiveness to therapeutic SIJ procedures.$^{405}$ As such, imaging modalities should not be solely relied upon for diagnosing SIJ complex pain or predicting responsiveness to SIJ fusion. However, imaging should be used to rule out other sources of pain, including lumbar radiculopathy, IA hip pathology, autoimmune arthropathy, tumor, or infection.

Diagnostic injections

Diagnostic injections play an important role in confirming SIJ-mediated symptoms.$^{8,64,80}$ However, a single SIJ injection has a high false-positive rate which may exceed 50% in some poorly selected populations.$^{8,40,55}$ Based on the available literature, dual local anesthetic injections alone (without cortico-steroid) result in the strictest criteria for diagnosing SIJ complex pain, while single uncontrolled injections of local anesthetic and corticosteroid result in the lowest specificity.$^{8}$ Using a stricter benchmark for a positive test also decreases the number of responders. For controlled dual anesthetic IA SIJ injections, increasing the threshold from 50% to 80% relief reduces the percentage of positive injections from 45% to 10% based on single studies. When steroid is added to the local anesthetic, the decrease in positive rate is smaller, from 33% to 27%, again based on single studies. For single anesthetic injections, 60%-62% are positive at an 80% threshold, which paradoxically compares with 54%-63% without sedation at a 50% threshold, underscoring differences in selection criteria.$^{8}$ Several SIJ fusion studies used single injections with a 50% pain relief cutoff for the diagnosis of SIJ complex pain; however, most did not specify the volume of anesthetic used for the injection.$^{286,287,369,394}$ As noted above, one study found no correlation between pain relief during diagnostic IA injections and SIJ fusion outcome, with this study employing a 50% cutoff threshold for enrollment.$^{146}$ Given that the volume of a symptomatic SIJ joint is only 1-2.7 mL,$^{129}$ low volume diagnostic injections of anesthetic only, with a documented arthrogram and absence of contrast flow to adjacent structures, is the best way to ensure target specificity.$^{8}$

There are significant EA sources of pain near the SIJ that can undermine the specificity of an IA SIJ injection.$^{139}$ The SIJ joint can communicate with the S1 foramen, and there are often defects in the anterior capsule, which can result in anesthetic extravasation outside the SIJ into muscles and/or the lumbosacral plexus, potentially leading to a false-positive response.$^{42}$ Given the anatomic complexity of the joint and necessity of confirming target-specific flow with contrast medium, diagnostic SIJ injections are typically performed using either fluoroscopic or CT guidance, not under ultrasound guidance, which frequently results in EA spread.$^{150,154,406}$ The addition of corticosteroid does not contribute to the diagnostic value of an injection and adds more volume, increasing the chance of a false-positive response from EA spread and/or systemic effects.$^{8}$ In randomized trials evaluating SIJ fusion, all studies have used 50% pain relief as the threshold for designating a diagnostic injection as positive.$^{146,286,287}$ Considering the rate of false-positive injections and the costs and risks of SIJ fusion, strict criteria for a positive anesthetic injection should be used including the use of dual IA SIJ local anesthetic injections.

Summary:

$\bullet$ Appropriate patient selection for SIJ fusion should include a combination of the following; Grade C recommendation, low level of certainty:

$\bullet$ At least 6 months of non-radicular pain below L5, consistent with SIJ pain.

* Failure of at least 12 weeks of conservative care (eg, physical therapy, therapeutic exercise, medications) and other non-surgical interventions.

* Patients who smoke should be referred for smoking cessation, and those on opioids should ideally be weaned off or tapered down.

* Physical exam indicating a positive response on at least 3 out of 6 SIJ provocation tests.

$\bullet$ History and radiologic studies that rule out other possible sources of pain, including generalized pain disorders, lumbar radiculopathy, hip pathology, autoimmune arthropathy, or destructive lesions such as tumor or infection.

* Positive responses to low-volume, dual prognostic IA SIJ injections (without corticosteroid), under fluoroscopic or CT guidance.

$\bullet$ Evidence is mixed on using >50% pain relief as the prognostic threshold to select patients for SIJ fusion (two pragmatic RCTs showed effectiveness, but a sham-controlled study did not distinguish fusion from sham outcomes). Given the invasiveness of SIJ fusion surgery, a higher pain relief threshold than 50%, along with documented functional improvement, is recommended to select patients for SIJ fusion; Grade I recommended (no consensus on number of blocks or specific cutoff was achieved in the working group.

Dissents: IPSIS, NASS and the NASS Fellowship Consortium recommend dual IA blocks with ≥75% pain relief to select patients for SIJ fusion.

Q21. What are the most common complications with sacroiliac joint fusion surgery and how can they be minimized?

Over the past 25 years, numerous reports on SIJ fusion have been published, but complication rates remain challenging to ascertain the final or designation. Studies employing stringent reporting methodologies have documented high numbers of adverse events, although many are unrelated to the procedure itself. Conversely, other studies report low-to-no complications. The most frequently reported complications associated with minimally invasive SIJF generally include lumbosacral radiculitis, persistent pain, and pseudarthrosis necessitating revision. An analysis of SIJF complications from the Manufacturer and User Facility Device Experience (MAUDE) database identified device malposition as the predominant complication, accounting for approximately 50% of all reported complications.$^{407}$ Malposition may lead to nerve and blood vessel injury, and the authors reported the root cause as surgical error, not device related, in 644 of the 1107 (58%) total complications reported in this series. Given the heterogeneity of data analyzed, further evaluation of the complications will be presented by approach and implant(s) used.

80

Pain Medicine, 2025, Vol. 00, No. 0

Open fusion$^{361}$,$^{362}$,$^{364}$,$^{365}$,$^{376}$,$^{377}$

Two case series from the same institution describe an open anterior approach to the SI joint, involving decortication for fusion and fixation with a 4.5 mm reconstruction plate. Ledonio et al.$^{365}$ reported complications in 14% of patients (3 out of 22) undergoing open fusion, including a pulmonary embolus and two revisions due to pseudarthrosis, with greater blood loss, operative time, and length of stay compared to minimally-invasive lateral fusion. Another study$^{376}$ reported a 21% complication rate (4 out of 19 patients), with three reoperations: One pulmonary embolus, one wound hematoma, one non-union, and an overlong screw causing soft tissue irritation. Buchowski et al.$^{361}$ found pseudarthrosis (15%), wound infection (10%), and painful hardware (5%) were the most common complications in their case series of 20 patients. Interestingly, earlier studies by Smith-Peterson et al.$^{360}$ in 1926 reporting on 26 patients, and Giannikas et al.$^{377}$ in 2004 describing an open lateral technique without implants in five patients, reported no complications. A retrospective study$^{364}$ comparing open posterior fusion to minimally-invasive lateral fusion found a 23% adverse event rate (34 events in 149 patients), with trochanteric bursits (n=4), postoperative neuropathy (n=4), and leg pain (n=3) being the most common complications. Notably, 44% of patients (66 out of 149) in this study required reoperation for implant removal.

Minimally-invasive lateral fusion

techniques $^{145,364,365,367,368,370,371,373-376,378-383,408$

Multiple reports have described complications related to implantation of a minimally-invasive, laterally placed interference device for fusion of the SIJ.

A prospective, multicenter RCT by Dengler et al.$^{367,368}$ provided comprehensive one- and two-year complication data for SI joint fusion. Out of 101 documented complications in the entire study population (including operative, conservative, and crossover groups), only four of 39 severe adverse events were likely procedure-related: Two cases of increased post-procedural pain, a gluteal muscle hematoma, and a case of nerve impingement requiring revision. Radiographic analysis of 43 patients in the fusion group at 12 months revealed cortical breach by implants in 8.6% (17 patients) of 198 evaluated implants. Three retrospective cohort studies involving triangular titanium implants reported varying complication rates. Smith et al.$^{364}$ reported no intraoperative complications in 114 patients treated with triangular titanium implants, with postoperative issues including falls (n=4), facet joint pain (n=4), and cellulitis (n=3); four patients returned to the operating room, three for nerve impingement and one for "radiographic findings." Ledonio et al.$^{365}$ noted two revisions for implant loosening and one pulmonary embolus in a group of 27 minimally-invasive fusions. Another study by Vanaclocha et al.$^{370}$ reported only two cases of temporary sciatic pain among 27 minimally-invasive fusions, with no revisions required. Additionally, a case report by Maxwell et al.$^{408}$ described a superior gluteal artery injury following percutaneous lateral SIJ fusion, necessitating open revision and branch ligation due to a large hematoma in the gluteal musculature and a significant hemoglobin drop within 24 h post-operation. Further case series are summarized in Table 20.

Minimally-invasive screw-based fusion techniques$^{371}$,$^{378}$,$^{380}$,$^{383}$

Screw-based, minimally-invasive SIJ fusion techniques have been reported and are summarized in Table 21. With low numbers, the complications most reported include postoperative pain, hematoma, and infection.

Posterior, minimally-invasive, percutaneous fusion$^{385}$,$^{386}$,$^{389}$

In the published observational studies that used a percutaneous, minimally-invasive, posteriorly implanted device to treat SIJ complex pain, relatively few complications were reported. In one study involving 111 treated patients, there was no comment on complications.$^{385}$ The only complication noted in another series of 50 treated patients with 12-month follow-up data was a case of device migration, treated with revision.$^{389}$ A multicenter, prospective series of 122 participants followed for 12 months and treated with the same device reported four serious adverse events although only one was deemed related to the study, which involved an aspiration event after initiation of monitored anesthesia care.$^{387}$ There was one patient in this study who reported increased SIJ pain.

Summary:

$\bullet$ SIJ fusion procedures are associated with standard postoperative complications such as infection, worsening or new-onset pain at the surgical site, hematoma, pseudarthrosis, and other device-related complications; Grade A evidence, high level of certainty.

* Postoperative neuropathy is a common and procedure-specific complication, likely due to the proximity of the L5 and S1 nerve roots to the operative field and has been reported in both open and minimally-invasive series. Proper evaluation of imaging is paramount to reducing this complication. Postoperative neuropathy appears to be more common with the lateral MIS fusion procedure than with anterior or posterior approaches; Grade C evidence, low level of certainty.

* There is evidence for a higher incidence of hematomas and neuropathic leg pain with open and lateral MIS (screw-based) SIJ fusion than with other minimally-invasive techniques; Grade C evidence, low level of certainty.

* There is a higher rate of device migration with posterior minimally-invasive SIJ percutaneous fusion than with other approaches; Grade C evidence, low level of certainty.

Discussion

Purpose of the guidelines

The present guidelines were developed to clarify evidence-based best practices in the diagnosis and treatment of SIJ complex pain. As a joint complex that interacts with the spine, pelvis, and hip, this structure is biomechanically distinct and innervated in a different fashion from other joints of the axial skeleton. Although it may be tempting to apply principles of spinal facet joint diagnosis and treatment to the SIJ complex,$^{28,29}$ doing so uniformly is inappropriate. Appreciation of the unique characteristics of the SIJ complex is vital to optimizing diagnosis and treatment. Table 22 provides a summary of statements and recommendations, and Figure 4 provides an algorithmic flow chart that can be used to guide treatment.

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Study type|No. of patients|Implant|Complications noted| |---|---|---|---|---| |$Cummings^{373}$|Retrospective Case Series|18|TTI|0 reported| |$Dengler^{368}$|Prospective RCT|103|CM vs TTI|Post op CT-198 implants-17 (8.6%) cortical breaches SIJF cohort complications:2 cases of increased SIJ complex pain1 gluteal hematoma1 case of new-onset radicular pain (req revision)1 revision for hardware looseningFusion group:2 cases of worsening pain w/ implant loosening1 case of new-onset leg pain (implant malposition)1 hematomaCM group1 hardware loosening after crossover fusion1 hematoma after crossover fusion1 case with 800 mL EBL3 cases of excessive pin advancement2 cases of post-op buttock pain2 wound infections1 case of cellulitis2 cases of PONV3 cases of neuropathic pain (required additional surgery)4 cases of hip or SIJ pain "attributed to implant"1 deep wound infection among 6 total infections8 revisions1 case of wound numbness1 case of vascular injury1 wound infection2 cases of implant loosening1 case of nerve root injury (S1)MIS TTI2 pulmonary emboli2 cases of nerve irritation requiring revision MISOPEN SURGERY1 pulmonary embolus2 revisions for 'failed implant' and 'nerve root irritation' (continued)| |$Hermans^{145}$|Retrospective Case Series|29|TTI|| |$Ledonio^{365}$|Retrospective Cohort Study|63|Anterior Plate vs TTI||

(continued)

83

|Study|Study type|No. of patients|Implant|Complications noted| |---|---|---|---|---| |Sachs and Capobianco $^{381}$|Retrospective Case Series|40|TTI|8 cases of facet joint pain2 cases of trochanteric bursitis1 case of new-onset LBP1 case of piriformis syndrome3 patients required subsequent lumbar surgeryNone reported| |Schmidt $^{376}$|Retrospective Case Series|19|TTI|Open cohort complications:21% (trochanteric bursitis, neuropathy and leg pain)66 of 149 (44%) had surgery to remove implantsMIS cohort complications:18% MIS (4 cases each of facet joint pain, falls, 2 cases each of trochanteric bursitis and piriformis syndrome, 1 case of new-onset back pain, 4 infections (1 wound infection, 3 cases of cellulitis), 2 hematomas4/114 (4%) had surgery to remove implants2 cases of temporary sciatic pain| |Smith $^{364}$|Retrospective Cohort Study|137|CM vsDenervation vsTTITTI|4/114 (4%) had surgery to remove implants2 cases of temporary sciatic pain| |Vanaclocha $^{370}$|Retrospective Cohort Study|93|TTI|3 of 103 had revision surgery1 case of trochanteric bursitis| |Whang $^{382}$|Prospective Case Series||||

CM, conservative management; PONV, postoperative nausea and vomiting; RCT, randomized controlled trial; TTI, triangular titanium implants.

Pain Medicine, 2025, Vol. 00, No. 0

|Study|Study type|No. of patients|Implant|Complications noted| |---|---|---|---|---| |Al-Khayer $^{371}$|Retrospective Case Series|9|Lateral screw|1 deep wound infection treated with debridement| |Endres & Ludwig $^{383}$|Prospective Case Series|19|Posterior screw|None reported| |Khurana $^{378}$|Retrospective Case Series|15|Lateral screw|No neurological or wound complications, no implant failures/revisions| |Kube & Muir $^{379}$|Retrospective Case Series|18|Lateral screw|1 guide wire fracture 1 case of prolonged postoperative pain 2 cases required prolonged surgery time No postoperative complications| |Rajpal & Burneikiene $^{380}$|Retrospective Case Series|24|Lateral Screw|2 subcutaneous hematomas, which resolved spontaneously 2 superficial wound infections treated with antibiotics 1 osteophyte adjacent to implant, later excised|

Table 22. Summary of statements and recommendations.

|Question|Statement/recommendation|Grade|Level of evidence| |---|---|---|---| |Q1|The prevalence of SIJ-mediated chronic LBP is estimated to be 15%-30% in patients with axial pain predominantly below L5. The prevalence of dorsal ligament-mediated pain is unknown but appears to be at least as common as intra-articular pain based on IA vs EA injection responses. The prevalence of SIJ complex pain is suspected to be higher in patients with a prior lumbar or lumbosacral fusion, estimated between 33% and 59%. Chronic LBP after fusion surgery may be related to adjacent segment stress, accelerated degeneration, missed SIJ pain before fusion, or violation of SIJ ligaments/synovium dur-ing bone grafting. Recommendation - Further studies are needed to evaluate the prevalence of dorsal liga-ment-mediated pain using EA injections and SLBBs.|B|Moderate N/A Low C Low| |Q2|There is low-quality evidence that history (eg, groin pain) and physical exam features (provocative tests) may help predict IA SIJ injection response. There is little data on using history and physical exam features to predict outcomes for patients with EA dorsal ligament pathology. Specificity of PE tests (single or clustered) is lower than sensitivity, with negative tests being more predictive than positive ones. Inter-rater reliability of SIJ provocation tests varies widely; FABER test has highest reliability. History and physical exam can help guide injections. IA injections may still be reason-able even without positive provocation tests when suspicion is strong.|B|Moderate N/A N/A N/A N/A| |Q3|Radionuclide bone scan sensitivity (13%-46%) is inadequate for diagnosis though spe-cificity is high (>90%). CT scan shows inadequate sensitivity (57.5%) and specificity (69%) for diagnosis. MRI: insufficient evidence, low sensitivity and specificity. SPECT/CT may show high sensitivity (95%) and specificity (99%) but evidence limited by using exam (not anesthetic injections) as the reference standard. SPECT/CT can identify uptake but there is insufficient evidence to distinguish IA SIJ from dorsal ligament pain.|D D I I N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A<||

(continued)

84

Pain Medicine, 2025, Vol. 00, No. 0

Table 22. (continued)

|Question|Statement/recommendation|Grade|Level of evidence| |---|---|---|---| |Q5|IA SIJ injections should use a maximum of 2.5 mL injectate to avoid capsular rupture and spread; injection may be halted at lower volumes if resistance is met. In patients with combined IA and EA pathology, volumes exceeding 3 mL may sometimes be required to treat dysfunction. Response to diagnostic IA SIJ injections can be used to select candidates for fusion surgery. Corticosteroids should not be added to diagnostic IA injections as they provide no diagnostic benefit and may increase extra-articular spread risk. SLBB local anesthetic volumes should be 0.67-1.2 mL per foraminal region at multiple depths to address neuroanatomy.|C|Low Moderate| |Q6|Fluoroscopically guided IA SIJ injections are more accurate than landmark-guided. Landmark-guided IA SIJ injections should not be performed. Image guidance should be strongly considered when extra-articular involvement is suspected. Fluoroscopy should include contrast injection to confirm intra-articular placement. Ultrasound guidance may be used in patients with contraindications to radiation/contrast (eg, pregnancy). Evidence insufficient to confirm ultrasound non-inferiority to fluoroscopy. Accuracy of ultrasound depends on proceduralist experience and patient BMI. Insufficient evidence comparing CT to fluoroscopy. CT guidance may be considered in advanced degenerative SIJ or failed fluoroscopy cases.|C|Low Moderate N/A Low N/A Low| |Q7|Extra-articular injections with corticosteroid have slightly stronger evidence of efficacy than intra-articular injections for at least one month. Evidence conflicting on superiority of EA vs IA injections at 3 months.|C|Low Low| |Q8|Preferred IA SIJ injection technique: fluoroscopic guidance, inferior joint access, contrast confirmation, and low injectate volume (<2.5 mL). If inferior access not possible, middle/superior access or CT guidance may be considered. Ultrasound-guided multi-site SLBBs may be considered for prognostic use.|B|Moderate High Moderate-High| |Q9|Fluoroscopy should be not recommended for diagnostic or prognostic SIJ/SLB injections due to increased false-positive rates. Routine sedation is not recommended for diagnostic or prognostic SIJ/SLB injections cases (needle phobia, psychiatric issues, severe vasovagal reactions). Light or moderate sedation may be used during SLBRFA depending on patient/provider preference. If sedation is required, the lowest dose of short-acting agents should be used.|C|Low Moderate| |Q10|Physical therapy and structured exercise may provide intermediate-term benefit. CBT may benefit patients with comorbid psychological issues. PT combined with CBT does not reduce risk of transition to chronic low back pain. SIJ belts may provide benefit in some patients. Insufficient evidence for chiropractic therapy effectiveness. NSAIDs and SNRIs may help; other antidepressants are not effective. Insufficient evidence for acute opioid use. No benefit from long-term opioid therapy. Acetaminophen is ineffective for SIJ complex pain.|C|Low Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N/A Low N|

(continued)

Pain Medicine, 2025, Vol. 00, No. 0

Table 22. (continued)

|Question|Statement/recommendation|Grade|Level of evidence| |---|---|---|---| |Q14|Cadaveric studies show dorsal innervation from S1-S3 in >90% of specimens; L5 and S4 branches are often involved.|B|Moderate| |Q14|S1-S3 lateral branches should always be targeted during SLBRFA.|A|High| |Q14|Consider targeting L5 and S4 lateral branches for complete denervation.|B|Moderate| |Q15|≥50% pain relief should be considered the cutoff for positive diagnostic/prognostic SIJ injections for SLBRFA selection.|B|Moderate| |Q15|Insufficient evidence to define EA prognostic cutoff before RFA.|I|N/A| |Q15|Mixed evidence for using >50% cutoff to select SIJ fusion candidates.|I|N/A| |Q15|Dual prognostic IA SIJ injections without corticosteroid and with higher thresholds recommended for fusion candidate selection.|I|N/A| |Q15|SLBBs should use 50% cutoff as prognostic threshold for RFA selection.|B|Moderate| |Q15|Therapeutic outcome success should follow IMPMPACT/FDA thresholds of ≥2 points or 30% improvement with functional benefit.|B|Moderate| |Q16|Older age (≥65), longer pain duration, higher baseline pain, and opioid use predict poorer outcomes after SLBRFA.|C|Low| |Q16|Anterior pathology or uncontrolled psychopathology predict treatment failure.|C|Low-Moderate| |Q16|Prognostic SLBBs are preferred for selecting SLBRFA candidates.|C-B|Low| |Q17|Anatomically-valid techniques (cooled monopolar, bipolar strip lesioning) improve pain/disability ≥6 months.|B|Moderate| |Q17|Without excluding invalid techniques, evidence quality drops.|C|Low| |Q17|Techniques that create larger lesions are supported for efficacy.|B|Moderate| |Q18|NSAIDs have weak evidence for preventing/treating neuritis.|C|Low| |Q18|Insufficient evidence for corticosteroids/membrane stabilizers for neuritis.|I|N/A| |Q18|Anticoagulation: consult specialist; in many cases anticoagulants may be continued.|C-B|Moderate| |Q18|SLBRFA safe in patients with implanted devices; bipolar lesioning may be safer.|B|Moderate| |Q18|Radiographs should be checked to ensure electrodes avoid hardware.|B|Moderate| |Q18|Sensory stimulation offers minimal safety or therapeutic value.|C|Low| |Q18|Motor stimulation can reduce risk of collateral nerve injury.|C|Low| |Q19|Open SIJ fusion can provide pain relief and functional improvement in selected patients.|C|Low-to-moderate| |Q19|Evidence for minimally-invasive lateral SIJ fusion is conflicting (two positive RCTs vs one negative sham RCT).|C|Very Low| |Q19|MIS posterior SIJ fusion may be considered for refractory degeneration with instability.|C|Low| |Q20|Appropriate selection should include ≥6 months non-radicular pain, failed ≥12 weeks of conservative therapy, ≥3 of 6 positive provocation tests, exclusion of other causes, and positive dual IA injections.|C|Low| |Q20|Evidence mixed on >50% cutoff for SIJ fusion candidate selection; higher thresholds with functional improvement recommended.|I|N/A| |Q21|SIJ fusion associated with standard surgical risks (infection, pain, hematoma, pseudarthrosis, device-related issues).|A|High| |Q21|Postoperative neuropathy is common, especially with lateral MIS approach.|C|Low| |Q21|Higher incidence of hematomas and neuropathic leg pain with open and lateral MIS compared to other approaches.|C|Low| |Q21|Posterior MIS SIJ fusion has higher device migration risk than other approaches.|C|Low|

CT, computed tomography EA, extra-articular; IA, intra-articular; LBP, low back pain; MRI, magnetic resonance imaging; NSAIDs, Non-steroidal anti-inflammatory drugs; SIJ, sacroiliac joint; SLBB, sacral lateral branch blocks; SLBRFA, sacral lateral branch radiofrequency ablation; SPECT, single photon emission computed tomography.

Epidemiology

Although the prevalence of other specific anatomical sources of LBP have been relatively well defined in select populations (namely, the lumbar facet joint and intervertebral disc),$^{409}$ accurate prevalence estimates are more challenging regarding the SIJ complex given its articular and EA ligamentous components, as well as intrinsic technical challenges in reliably performing diagnostic IA-only injections.$^{5}$ A literature review performed for the present guidelines demonstrated the prevalence of SIJ-mediated chronic LBP to be between 15% and 30%, but the epidemiological studies are methodologically

insufficient to estimate the prevalence of EA ligamentous pain with any precision.

Diagnosis

There are some features pertaining to clinical history (maximal pain within 1 cm of a location just inferomedial to the posterior superior iliac spine,$^{57}$ groin pain,$^{43,61,62}$ and injuries with specific mechanisms,$^{64}$ and lumbosacral fusion,$^{65}$ among others) that may increase the likelihood of a positive response to an IA SIJ local anesthetic injection, suggesting IA-mediated pain, but studies have not typically distinguished

Pain Medicine, 2025, Vol. 00, No. 0

SUSPECTED SIJ COMPLEX PAIN BASED ON HISTORY AND PHYSICAL EXAMINATION

![graph TD\n A1[Suspected Intra-Articular Pathology] --> A2[Low-volume IA injection with steroid]\n A2 --> A3[Pain relief ≥ 3 months]\n A2 --> A4[Pain relief < 3 months]\n A2 --> A5[No pain relief]\n A3 --> A6[Repeat injection as needed]\n A4 --> A11[Consider IA PRP]\n A5 --> A14[Consider EA pathology & injection]\n A6 --> A7[Response]\n A7 --> A8[Repeat as needed]\n A7 --> A9[Consider controlled low-volume IA injection without steroid before consideration of minimally-invasive fusion]\n A7 --> A10[Consider non-SIJ pathology]\n A11 --> A12[Response]\n A12 --> A13[Repeat as needed]\n A14 --> A15[Response]\n A15 --> A16[Repeat as needed]\n A15 --> A17[Consider non-SIJ pathology]\n A18[Suspected Extra-Articular Pathology] --> A19[EA or combination EA+IA injection with steroid]\n A19 --> A20[Pain relief ≥ 3 months]\n A19 --> A21[Pain relief < 3 months]\n A19 --> A22[No pain relief]\n A20 --> A23[Repeat injection as needed]\n A21 --> A24[Consider Prolotherapy OR Prognostic sacral lateral branch blocks before considering RF ablation]\n A22 --> A25[Consider IA pathology & injection]\n A23 --> A26[Response]\n A25 --> A27[Response]\n A26 --> A28[Repeat as needed]\n A27 --> A29[Consider non-SIJ pathology]\n](1/parseimages/figure37550b7168474c5bb6a9bdf08224a4f8_0_2.png) Figure 4. Flow chart demonstrating treatment of suspected sacroiliac joint complex pain based on history, physical exam and response to interventions.

between IA pain and dorsal ligament pain. Whereas a physical examination in isolation is an imperfect tool for diagnosing SIJ complex pain, the sensitivity of individual and clusters of SIJ complex provocation tests is associated with a reasonable negative predictive value for the outcome of an IA SIJ injection. Notably, there remains a lack of research on distinguishing IA-mediated pain from dorsal ligament pain by physical examination. Although the latter may elicit more tenderness due to the closer proximity to skin and younger demographic, this has not been tested in clinical trials.

Imaging is important for ruling out inflammatory, infectious, malignant, and other non-degenerative/biomechanical causes of SIJ complex pain but otherwise has minimal or unclear utility for diagnosing SIJ complex pain. This is the case regardless of imaging modality (radiographs, MRI, CT scan, radionuclide bone scan, SPECT/CT).

Given the limitations of history, physical examination, and imaging, interventional-based diagnosis remains the reference standard for diagnosing and distinguishing between IA-mediated pain and dorsal ligament pain given. Although medial branch blocks represent the reference standard for diagnosing artular lumbar facet joint pain as opposed to IA injection,$^{29}$ this is not the case for isolated IA pain, wherein IA injection represents the reference standard for diagnosis. Alternatively, SLBBs represent the reference standard for diagnosing dorsal ligament complex pain.$^{116}$

Conservative and minimally-invasive treatment results

The evidence supporting conservative treatments for SIJ complex pain is limited and largely indirect. While PT, structured exercise, and cognitive behavioral therapy (particularly for

those with comorbid psychopathology) may offer some short-term benefits, the level of certainty remains low. Indirect evidence from LBP studies suggests that NSAIDs and antidepressants may be beneficial, but the efficacy of opioids, especially for long-term use, is not supported. SIJ belts show weak evidence of potential benefit in some patients, while evidence for chiropractic treatment is insufficient. Notably, acetaminophen appears ineffective for LBP in general, with no direct evidence for SIJ complex pain. Given the paucity of high-quality, direct evidence, further research is needed to establish the effectiveness of conservative treatments specifically for SIJ complex pain.

Injection-based treatment

There is reasonable evidence that both IA and EA/ligamentous corticosteroid injections are effective in the short-term for the treatment of SIJ and/or dorsal ligament complex pain, which is in contrast to IA facet joint corticosteroid injection which does not reliably differ from placebo.$^{29}$ There is presently insufficient evidence to recommend non-corticosteroid-based injection treatments for SIJ complex pain. This is an area in which further study is needed. The accuracy of such injections relies on image guidance by either fluoroscopy of CT. Fluoroscopic guidance provides better accuracy compared to landmark-based injection. Whereas some evidence suggests ultrasound guidance is less accurate than fluoroscopic-guidance for achieving reliable IA injection,$^{150}$ the evidence is insufficient to determine non-inferiority.

Radiofrequency ablation-based treatment

It is well established that best practices for selecting individuals for lumbar facet denervation are based on the use of medial branch blocks, not IA facet joint injections.$^{29}$

Pain Medicine, 2025, Vol. 00, No. 0

Alternatively, the literature review performed for the present guidelines demonstrated that both SLBBs and IA SIJ injections have reasonable prognostic value for the predicting the outcome of SLBRFA; however, there are no data on patients who have undergone SLBRFA without prognostic blocks, or with negative blocks. Whereas SLBBs have greater face validity as a prognostic test, the present evidence indicates that either type of prognostic (SLBB or IA SIJ) injection can be reasonably used (Grade C evidence, low level of certainty). In clinical studies, since they are prognostic, SLBB are almost never used without a preceding positive SIJ injection.

It is well established that the most effective way to achieve facet joint denervation by RFA involves targeting the medial branch nerves with a large-gauge electrode, using a near-parallel technique.$^{29}$ The literature review performed for the present guideline revealed that, while many techniques have been proposed, only two techniques reliably capture the sacral lateral branch nerves necessary for effective

denervation of the dorsal ligament complex$^{121}$,$^{122}$: (1) comprehensive periforaminal large-gauge monopolar RFA; and (2) continuous bipolar-strip lesions lateral to the sacral foramina. For these anatomically valid techniques, there is grade B evidence for improving pain and disability for at least 6 months post-treatment. As such, sacral lateral branch RFA should be performed using one of these two techniques unless future evidence validates additional techniques that provide the same reliability for sacral lateral branch capture.

Sacroiliac joint fusion

Utilization of SIJ fusion has increased substantially in the past two decades. One study that queried Medicare databases found a 2351% increase in utilization of SIJ fusion between 2010 and 2020 in the United States.$^{410}$ This trend is likely to continue in the United States given coverage restrictions implemented within the past two years on other treatments for SIJ complex pain for Medicare beneficiaries, including

|Question|Organization-specific committee members|$Organizations^{a,b,c}$| |---|---|---| |Guidelines en bloc|27 approved|24 endorsed| |Prevalence of SIJ pain|27 approved|24 endorsed| |Use of history and physical exam to diagnose|27 approved|24 endorsed| |SIJ pain||| |Radiological imaging for diagnosis|27 approved|24 endorsed| |Diagnostic and prognostic value of IA injections and SLBB|27 approved|24 endorsed| |Volume and composition of SIJ IA injections|27 approved|24 endorsed| |Accuracy of using anatomical landmarks and imaging for IA injectate spread|27 approved|24 endorsed| |Relative effectiveness for IA vs EA SIJ injections|27 approved|23 endorsed, 1 abstained$^{d}$| |injections||| |Technique for IA injections and SLBB|27 approved|24 endorsed| |Use of sedation for diagnostic and prognostic blocks and RFA|27 approved|24 endorsed| |Effectiveness of non-invasive treatments for SIJ complex pain|27 approved|24 endorsed| |Evidence for non-steroid injections for SIJ complex pain|27 approved|24 endorsed| |Evidence for cryoablation|27 approved|24 endorsed| |Effectiveness of SLBRFA|27 approved|24 endorsed| |Targeted nerves for SLBRFA|27 approved|24 endorsed| |Cutoff thresholds for selecting patients for SLBRFA and SIJ fusion & designating outcomes|25 approved|24 endorsed the summary, but 3 dissented on cutoff threshold for selecting patients$^{e}$| |Ideal candidates for SLBRFA|27 approved|24 endorsed| |Optimal technique for SLBRFA|27 approved|24 endorsed| |Minimizing the most common complications from SLBRFA|27 approved|24 endorsed| |Effectiveness of surgery for SIJ pain|27 approved|24 endorsed| |Selecting patients for SIJ fusion|27 approved|24 endorsed the summary, but 3 dissented on a specific cutoff for a positive block before $surgery^{c}$| |Identifying and minimizing the most common complications from SIJ fusion|27 approved|25 endorsed| |Periodities for SIJ fusion|||

EA, extra-articular; IA, intra-articular; SIJ, sacroiliac joint; SLBB, sacral lateral branch blocks; SLBRFA, sacral lateral branch radiofrequency ablation.

a The Departments of Defense and Veterans Affairs did not vote on the guidelines but their representatives approved them.

b The American Academy of Neurology declined to affirm the benefit of the guideline citing lack of relevance to their membership.

c The American Society of Anesthesiologists, American Academy of Physical Medicine & Rehabilitation and the North American Spine Society "affirmed the benefit" of the guidelines rather than formally endorsed them. For simplicity sake, these votes are listed in the table as "endorsements", while dissents are listed as such. Since these documents were not formally approved by these societies' Boards of Directors or Houses of Delegates, they do not represent official or approved statements or policies of these Societies. Although the document is supported by the ASA, variations from the recommendations contained in the document may be acceptable based on the judgment of the responsible anesthesiologist.

d The Latin American Society for Regional Anesthesia abstained on the issue of stronger evidence for EA than IA injections.

e NASS, NASS Fellowship Consortium and IPSIS recommend dual SLBBs with ≥75% cutoff threshold, and dual IA injections with ≥75% cutoff threshold, to select patients for SLBRFA and SIJ fusion, respectively. The multisociety guidelines recommend a single block with ≥50% cutoff before SLBRFA to maximize access to care and improve sensitivity and recommend dual IA blocks with "a higher threshold than 50% pain relief" to select patients for SIJ fusion, without designating what that threshold is. For SIJ fusion, all 24 voting societies concurred on the other selection criteria.

Pain Medicine, 2025, Vol. 00, No. 0

SLBRFA. The present review found that the lateral minimally-invasive surgical (MIS) approach to SIJ fusion can be considered in patients with refractory IA degeneration with instability who fail to respond to more conservative therapies (Grade C recommendation, very low level of certainty). Additionally, there is weak evidence that both an open approach to SIJ fusion and the MIS posterior approach can provide pain relief and functional benefit in carefully selected patients who have not responded to other therapies (Grade C recommendation, low level of certainty). Appropriate selection of patients for fusion includes the following criteria: (1) At least 6 months of non-radicular pain below L5, consistent with SIJ complex pain; (2) failure of at least 12 weeks of conservative care that may include PT, therapeutic exercise and medications, as well as nonsurgical interventions; (3) physical exam indicating positive response to at least 3/6 SIJ provocative tests; (4) history and radiologic studies that rule out other possible sources of pain, including generalized pain disorders, lumbar radiculopathy, hip pathology, autoimmune arthropathy, or destructive lesions such as tumor or infection; (5) at least 50% pain improvement following low-volume, dual comparative diagnostic SIJ injections without corticosteroid, under fluoroscopic or CT guidance.

Approval of participating organizations

All voting organizations endorsed or affirmed the guidelines except one; the American Academy of Neurology (AAN) declined to affirm the guidelines citing a lack of relevance to their membership. The American Society of Anesthesiologists (ASA), American Academy of Physical Medicine and Rehabilitation (AAPMR), and North American Spine Society (NASS) affirmed the benefit of the guidelines but did not formally approve them per their bylaws. Whereas the Departments of Defense and Veterans Affairs did not per policy formally review these non-federally sponsored guidelines for endorsement, their representatives approved them. Consensus was obtained on all 21 questions by all committee members, and 18 of 21 questions by the 24 voting organizations who formally endorsed or affirmed the benefit of the guidelines (see Table 23). On one question (stronger evidence for EA than IA injections), one society, the Latin American Society of Regional Anesthesia (LASRA) abstained, believing the evidence was not strong enough to render any recommendation. On 2 recommendations, there were dissensions from 3 participating societies that overlap in membership and leadership: NASS, the NASS Fellowship Consortium, and the International Pain and Spine Intervention Society (IPSIS). These organizations all dissented on the recommendation for ≥50% cutoff from a single block to select patients for RFA (which prioritized access to care and enhancing sensitivity at the expense of specificity), instead advocating for ≥75% pain relief after 2 SLBBs; for the designation of a positive outcome, all concurred. The second question that these three organizations dissented on was related to the first, with the guidelines recommending dual blocks without steroids and an unspecified threshold exceeding 50% with documented functional improvement to select patients for SIJ fusion (ie, no cutoff percentage designated). The 3 dissenting organizations believe that the percent pain relief required should equal or exceed 75%. On the other selection criteria for SIJ fusion, the organizations concurred.

Guideline context and limitations

As with many areas of medicine, not all clinical questions have been definitively answered; yet, patients, physicians, payers, and regulatory agencies need a structural framework to make decisions and create reasonable policies. The present guidelines provide a foundation for understanding evidence-based best practices in the diagnosis and treatment of SIJ complex pain, but they are also meant to be interpreted through the lens of reasonable judgment by physicians. Clinical practice guidelines are meant to allow for variations in practice, based on case-specific patient scenarios, as well as shared decision-making between the patient and physician. Whereas a "guideline-concordant" decision may be the best choice for one patient, this is not necessarily true for the next patient.

Some of our findings are limited in part by the lack of large high-quality studies, with several of the randomized trials evaluating invasive therapies being sponsored by industry, which are more likely to yield positive results.$^{411}$ Differences in study design, selection criteria, outcome measures, follow-up, and technique amplify the issues surrounding low sample sizes and power inherent to interventional studies, and limit generalization. Although we included several societies with significant or predominant spine surgeon membership and one spine surgeon on our guidelines, their inclusion on these guidelines is underrepresented. We failed to include any physical therapists or physical therapy societies, which are also important stakeholders.

Conclusions

These international, multi-society clinical practice guidelines provide a framework for evidence-based best practices in the diagnosis and treatment of SIJ complex pain. These guidelines are meant to inform patients, physicians, payers, and regulatory agencies in arriving at clinical care decisions and in creating reasonable policy. The recommendations contained represent consensus amongst 27 different academies, societies, and other organizations (eg, U.S. Departments of Defense and Veteran Affairs) spanning much of North America, Europe, Latin America, Asia, and the South Pacific.

Supplementary material

Supplementary material is available at Pain Medicine online.

Funding

Research reported in this publication was supported by the National Institutes of Health HEAL Initiative and National Institute on Drug Abuse of the National Institutes of Health under grant number R24DA055306 (R.W.H.). SPC receives funding including effort spent working on these guidelines from the National Institutes of Health National Institute for Arthritis and Musculoskeletal and Skin Disorders (1UH3135804), U.S. Department of Defense (Neu-92-13337), and the Dept. of Health and Human Services Advanced Research Projects Agency for Health (ARPA-H), TDL 24-ICHUB-0001.

Pain Medicine, 2025, Vol. 00, No. 0

Disclaimer

This report does not represent the official view of the National Institute of Drug Abuse, the National Institutes of Health, or any part of the US Federal Government. No official support or endorsement of this article by the NIDA or NIH is intended or should be inferred.

Conflicts of interest

Z.L.M.: Research grants (not related to this work): Avanos Medical, Boston Scientific, Presidio Medical, Saol Therapeutics, Spine Biopharma, SPR Therapeutics, Stratus Medical (paid directly to the University of Utah); Consulting (not related to this work): Saol Therapeutics, Stryker, Avanos (all relationships ended 2025), R.W.H.: Cooperative (eg, U grants) NIH grants, NIH grants not related to this work; Nevro, Inc—research funds to the institution (topic—painful diabetic neuropathy) and not related to this work, and State Farm—expert/consulting not related to this work; Avanos Medical research funds to the institution (topic—SIJ Complex Pain) which ended 2023. M.A.: Consulting for Medtronic (self), Abbott (self), Boston Scientific (self); President Elect NANS; Medtronic (Institution for fellowship grant grants), Abbott (Institution for fellowship grant and research), Boston Scientific (Institution for fellowship grant and research). A.B.: No conflicts to report. A.B.: Honoraria for speaking and advising—Bioventus, Boston Scientific, Medtronic. R.C.C.: Consultant, Johnson & Johnson, 2023, course instructor. A.S.C.: Instructor (travel): International Pain and Spine Society, Instructor (travel): North American Spine Society, Educational Grant (to institution): Medtronic. T.C.D.: No conflicts to report. J.D.A.A.: No conflicts to report. J.L.d.C.: Consulting for Hospital Vera Cruz—Campinas-SP (34 years); Sobramid—Former President and now Director Board; Adora Clinic-Medical Director. S.M.H.: Ex-officio North American Neuromodulation Society. B.C. H.-P.: Vice President, Asociación Mexicana para el Estudio y Tratamiento del Dolor. N.R.K.: Consulting fees from Vertex Pharmaceuticals for unrelated topics; On Board of Directors for non-profit “Neuro Hospitality House” which provides affordable housing to patients receiving neurological cares. Institution receives research funding for unrelated topics (Institution): Virtes (Institution), INH-R. President Elect NANS; Medtronic (Institution): AsRA Pain Medicine BOD; AAPM BOD; SAAAPM BOD. M.F.E.L.B.: Conflates to report. J.Y.M.: No disclosures to report. D.A.P.: Consulted for Avanos, Boston Scientific, Medtronic, SI Bone, and Wavegate; ASRA Board Of Directors; Pain Diagnostics and Interventional Care has received research support from Avanos, Boston Scientific, Brixton, and Medtronic. D.E.R.:—PainTEQ—Teaching Faculty (Absolute Pain Management)—No stocks/ investments. Honoraria for teaching cadaver laboratory training and didactic for civilian physician practices. N.M.S.: Consulting for Averitas, Lohocla, Lundbeck, Salvia, Rapport, Schedule 1, Syneos, Syqe, Vertex (individual); Expert Testimony to various law firms and State of California (Individual and Institution); Board of Directors of American Academy of Pain Medicine (Individual); ShiraTronics (Clinical Trial, Institution). C.C.S.: Research support from Discogen. A.S.: Consulting for State Farm (individual) and Atlas Health (unpaid); PSIS BOD, Vice President (payment

Spine Society, Educational Grant (to institution): Medtronic. T.C.D.: No conflicts to report. J.L.d.C.: Consulting for Hospital Vera Cruz—Campinas-SP (34 years); Sobramid—Former President and now Director Board; Adora Clinic-Medical Director. S.M.H.: Ex-officio North American Neuromodulation Society. B.C. H.-P.: Vice President, Asociación Mexicana para el Estudio y Tratamiento del Dolor. N.R.K.: Consulting fees from Vertex Pharmaceuticals for unrelated topics; On Board of Directors for non-profit “Neuro Hospitality House” which provides affordable housing to patients receiving neurological cares. Institution receives research funding for unrelated topics from Shiratronics and NIH. L.R.K.: Consulting—FUS mobile (Institution). Vertex (Institution), Averitas (Institution), NIH (Institution); ASRA Pain Medicine BOD; AAPM BOD; SAAAPM BOD. M.F.E.L.B.: Confiles to report. J.Y.M.: No disclosures to report. D.A.P.: Consulted for Avanos, Boston Scientific, Medtronic, SI Bone, and Wavegate; ASRA Board Of Directors; Pain Diagnostics and Interventional Care has received research support from Avanos, Boston Scientific, Brixton, and Medtronic. D.E.R.:—PainTEQ—Teaching Faculty (Absolute Pain Management)—No stocks investments. Honoraria for teaching cadaver laboratory training and didactic for civilian physician practices. N.M.S.: Consulting for Averitas, Lohocla, Lundbeck, Salvia, Rapport, Schedule 1, Syneos, Syqe, Vertex (individual); Expert Testimony to various law firms and State of California (Individual and Institution); Board of Directors of American Academy of Pain Medicine (Individual); ShiraTronics (Clinical Trial, Institution). C.C.S.: Research support from Discogen. A.S.: Consulting for State Farm (individual) and Atlas health (unpaid); PSIS BOD, Vice President (payment

for travel), IPSIS Ultrasound Committee (payment for travel), NASS DEI committee (unpaid) and NASS Section Chair, Comprehensive Episodes of Spine Care (unpaid); NIH grant, ASTRAL lumbar radiofrequency trial (Institution). K.S.: Vice president World Institute of Pain Benelux chapter. D.-A.T.: No conflicts to report. N.T.: No conflicts to report. M.F.V.: Director of Professional Affairs, Faculty of Pain Medicine, ANZCA (Paid role concerned with liaison with external stakeholders and professional documents for FPM) - No direct COI with the topic of the Guidelines. J.V.Z.: President-elect of the World Institute of Pain (unpaid), Liaison Algology Belgian Professional Society of Anesthesia and Reanimation (unpaid). A.W.: United States Association for the Study of Pain (Board of Directors (term completed), unpaid); American Society of Anesthesiologists Abstract Review Committee on Chronic and Cancer Pain (Vice Chair, unpaid); received funding from the VA and royalties from Elsevier for Pain Management Board Review book (paid to University) received some speaker honoraria from the Michigan Society of Anesthesiologists. S.P.C.: Avanos, SPR Therapeutics, Vertex, Persica, Sword Health: Consulting Scilex: Consulting (ended 2024).

References

1. Cohen SP, Chen Y, Neufeld NJ. Sacroiliac joint pain: A comprehensive review of epidemiology, diagnosis and treatment. Expert Rev Neurother. 2013;13:99-116. https://doi.org/10.1586/ern.12.148

2. Gartenberg A, Nessim A, Cho W. Sacroiliac joint dysfunction: Pathophysiology, diagnosis, and treatment. Eur Spine J. 2021;30:2936-2943. https://doi.org/10.1007/s00586-021-06927-9

3. Szadek KM, Hoogland PV, Zuurmond WW, de Lange JJ, Perez RS. Nociceptive nerve fibers in the sacroiliac joint in humans. Reg Anesth Pain Med. 2008;33:36-43. https://doi.org/10.1016/j.rampr.2007.07.011

4. Szadek KM, Hoogland PV, Zuurmond WW, De Lange JJ, Perez RS. Possible nociceptive structures in the sacroiliac joint carti-lage: An immunohistochemical study. Clin Anat. 2010;23:192-198. https://doi.org/10.1002/ca.20908

5. Schneider BJ, Rosati R, Zheng P, McCormick ZL. Challenges in diagnosing sacriliac joint pain: A narrative review. PM R. 2019;11:S40±45. https://doi.org/10.1002/pmrj.12175

6. Cohen SP, Bicket MC, Kurihara C, et al. Fluoroscopically guided vs landmark-guided sacroiliac joint injections: A randomized controlled study. Mayo Clin Proc. 2019;94:628-642. https://doi.org/10.1016/j.maoycp.2018.08.038

7. Khalil IA, Horn P, Kanaan A, et al. Efficacy of peri-articular versus intra-articular sacroiliac joint blocks under fluoroscopy guidance: A multicentric study. Chonic Pain Manag J. 2022;6:141. https://doi.org/10.29011/2576-957X.100041

8. Kennedy DJ, Engel A, Kreiner DS, Nampiaparamil D, Duszynski B, MacVicar J. Fluoroscopically guided diagnostic and therapeutic intra-articular sacroiliac joint injections: A systematic review. Pain Med. 2015;16:1500-1518. https://doi.org/10.1111/pme.12833

9. Luukkainen R, Nissilä M, Asikainen E, et al. Periarticular cortosteroid treatment of the sacroiliac joint in patients with seronegative spondylarthropathy. Clin Exp Rheumatol. 1999;17:88-90.

10. Luukkainen RK, Wennerstrand PV, Kautiainen HH, Sanila MT, Asikainen EL. Efficacy of periarticular corticosteroid treatment of the sacroilacil joint in non-spondylarthropathic patients with chronic low back pain in the region of the sacroiliac joint. Clin Exp Rheumatol. 2002;20:52:54.

11. Maugars Y, Mathis C, Berthelot JM, Charlier C, Prost A. Assessment of the efficacy of sacroiliac corticosteroid injections in

90

Pain Medicine, 2025, Vol. 00, No. 0

spondylarthropathies: A double-blind study. Br J Rheumatol. 1996;35:767-770. https://doi.org/10.1093/rheumatology/35.8.767 Colen S, Haverkamp D, Mulier M, van den Bekerom MP. Hyaluronic acid for the treatment of osteoarthritis in all joints except the knee: What is the current evidence? BioDrugs. 2012;26:101-112. https://doi.org/10.2165/11630830-000000000-00000 13. Goodwin B, Averell N, Al-Shehab U, et al. Efficacy of platelet-rich plasma for sacroiliac joint dysfunction: A qualitative systematic review with pooled analysis. Regen Med. 2023;18:505-514. https://doi.org/10.2217/rme-2022-0211 14. Cohen SP, Kapural L, Kohan L, et al. Cooled radiofrequency ablation versus standard medical management for chronic sacroiliac joint pain: A multicenter, randomized comparative effectiveness study. Reg Anesth Pain Med. 2024;49:184-191. https://doi.org/10.1136/rapm-2023-104568 15. Dutta K, De Sy, Bhattacharyya P, Agarwal S, Dev P. Comparison of efficacy of lateral branch pulsed radiofrequency denervation and intraarticular depot methylprednisolone injection for sacroiliac joint pain. Pain Phys. 2018;21:489-496. 16. Juch JNS, Maas ET, Ostelo R, et al. Effect of radiofrequency denervation on pain intensity among patients with chronic low back pain: The mint randomized clinical trials. JAMA. 2017;318:68-81. https://doi.org/10.1001/jama.2017.7918 17. Lowe M, Okunlola O, Raza S, et al. Radiofrequency ablation as an effective long-term treatment for chronic sacroiliac joint pain: A systematic review of randomized controlled trials. Cureus. 2022;14:e26327. https://doi.org/10.7759/cureus.26327 18. Salman OH, Gad GS, Mohamed AA, Raeaf HH, Abedlhat AM. Randomized, controlled blind study comparing sacroiliac intra-articular steroid injection to radiofrequency denervation for sacroiliac joint pain. Egyptian J Anaesstheia. 2016;32:219-225. 19. Ferrante FM, King LF, Roche EA, et al. Radiofrequency sacroiliac joint denervation for sacroiliac syndrome. Reg Anesth Pain Med. 2001;26:137-142. https://doi.org/10.1053/rapm.2001.21739 20. Gevargez A, Groenemeyer D, Schirp S, Braun M. CT-guided percutaneous radiofrequency denervation of the sacroiliac joint. Eur Radiol. 2002;12:1360-1365. https://doi.org/10.1007/s00330-001-1257-2 21. Cohen SP, Hurley RW, Buckenmaier CC, 3rd, Kurihara C, Morlando B, Dragovich A. Randomized placebo-controlled study evaluating lateral branch radiofrequency denervation for sacroiliac joint pain. Anesthesiology. 2008;109:279-288. https://doi.org/10.1097/ALN.0b013e31817f4c7c 22. Patel N, Gross A, Brown L, Gekht G. A randomized, placebo-controlled study to assess the efficacy of lateral branch neurotomy for chronic sacroiliac joint pain. Pain Med. 2012;13:383-398. https://doi.org/10.1111/j.1526-4637.2012.01328.x 23. Ramasubba C, Cohen SP. Cooled sacroiliac radiofrequency denervation for the treatment of pain secondary to tumor infiltration: A case-based focused literature review. Pain Physician. 2013;16:1-8. 24. van Tilburg CW, Schuurmans FA, Strons DL, Groeneweg JG, Huygen FJ. Randomized sham-controlled double-blind multicenter clinical trial to ascertain the effect of percutaneous radiofrequency treatment for sacroiliac joint pain: Three-month results. Clin J Pain. 2016;32:921-926. https://doi.org/10.1097/ajip.0000000000000351 25. Yin W, Willard F, Carreiro J, Dreyfuss P. Sensory stimulation-guided sacroiliac joint radiofrequency neurotomy: Technique based on neuroanatomy of the dorsal sacral plesus. Spine (Pbila Pa 1976). 2003;28:249-2425. https://doi.org/10.1097/01.Brs.0000085360.03758.C3 26. Maccagnano G, Noia G, Cassano GD, et al. Thermal versus cooled radiofrequency in patients with sacroiliac joint pain: A systematic review of the literature and pooled analysis of clinical outcomes. J Neurosurg Sci. 2022;66:485-493. https://doi.org/10.23736/s0390-5616.22.05525-4

27. Mehta V, Poply K, Husband M, Anwar S, Langford R. The effects of radiofrequency neurotomy using a strip-lesioning device on patients with sacroiliac joint pain: Results from a single-center, randomized, sham-controlled trial. Pain Phys. 2018;21:607-618. 28. Hurley RW, Adams MCB, Barad M, et al. Consensus practice guidelines on interventions for cervical spine (facet) joint pain from a multispeciality international working group. Pain Med. 2021;22:2443-2524. https://doi.org/10.1093/pm/pnab2819. Cohen SP, Bhaskar A, Bhatia A, et al. Consensus practice guidelines on interventions for lumbar facet joint pain from a multispeciality, international working group. Reg Anesth Pain Med. 2020;45:424-467. https://doi.org/10.1136/rapm-2019-101243 30. USPSTF: Grade Definitions. U.S. Preventive Services Task Force. 2024. Accessed August 5, 2024. https://www.uspreventiveservices-taskforce.org/uspstf/about-uspstf/methods-and-processes/grade-definitions 31. Brouwers MC, Kho ME, Brownman GP, et al. AGREE Next Steps Consortium. AGREE II: Advancing guideline development, reporting and evaluation in health care. CMAJ. 2010;182: E839-E842. https://doi.org/10.1503/cmaj.090449 32. Cohen SP, Bhatia A, Buvanendran A, et al. Consensus guidelines on the use of intravenous ketamine infusions for chronic pain from the American Society of Regional Anesthesia and Pain Medicine, the American Academy of Pain Medicine, and the American Society of Anesthesiologists. Reg Anesth Pain Med. 2018;43:521-546. https://doi.org/10.1097/aap.00000000000000808 33. Deer TR, Narouze S, Provenzano DA, et al. The Neurostimulation Appropriateness Consensus Committee (NACC): Recommendations on bleeding and coagulation management in neurostimulation devices. Neuromodulation. 2017;20:51-62. https://doi.org/10.1111/ner.12542 34. Deer TR, Pope JE, Hayek SM, et al. The Polyanalgesic Consensus Conference (PACC): Recommendations for intrathecal drug delivery: guidance for improving safety and mitigating risks. Neuromodulation. 2017;20:155-176. https://doi.org/10.1111/ner.12579 35. Helm li S, Simopulos TT, Stojanovic M, Abdi S, El Terany MA. Effectiveness of thermal annular procedures in treating discogenic low back pain. Pain Physician. 2017;20:447-470. 36. GRADE. GRADE Working Group. 2024. Accessed August 5, 2024. https://www.adeworkinggroup.org/ 37. CEBM. OCEBM Levels of Evidence. https://www.cebm.ox.ac.uk/resources/levels-of-evidence/ocebm-levels-of-evidence 38. Cohen SP, Wallace M, Rauck RL, Stacey BR. Unique aspects of clinical trials of invasive therapies for chronic pain. Pain Rep. 2019;4:e687. https://doi.org/10.1097/pr9.00000000000000687 39. Wong CK, Mak RY, Kwok TS, et al. Prevalence, incidence, and factors associated with non-specific chronic low back pain in community-dwelling older adults aged 60 years and older: A systematic review and meta-analysis. J Pain. 2022;23:509-534. https://doi.org/10.1016/j.jpain.2021.07.012 40. Simopoulos TT, Manchikanti L, Singh V, et al. A systematic evaluation of prevalence and diagnostic accuracy of sacroiliac joint interventions. Pain Phys. 2012;15:E305-E344. 41. Zheng P, Schneider BJ, Yang A, McCormick ZL. Image-guided sacroiliac joint injections: An evidence-based review of best practices and clinical outcomes. PM R. 2019;11: S98-S104. https://doi.org/10.1002/pmrj.12191 42. Fortin JD, Washington WJ, Falco FJ. Three pathways between the sacroiliac joint and neural structures. AJNR Am J Neuroradiol. 1999;20:1429-1434. 43. Schwarzer AC, April CN, Bogduk N. The sacroiliac joint in chronic low back pain. Spine (Phila Pa 1976). 1995;20:31-37. https://doi.org/10.1097/0007632-199501000-00007 44. Cohen SP, Hameed H, Kurihara C, et al. The effect of sedation on the accuracy and treatment outcomes for diagnostic injections: A randomized, controlled, crossover study. Pain Med. 2014;15:588-602. https://doi.org/10.1111/pme.12389

Pain Medicine, 2025, Vol. 00, No. 0

91

45. Telli H, Hüner B, Kuru Ö. Determination of the prevalence from clinical diagnosis of sacroiliac joint dysfunction in patients with lumbar disc hernia and an evaluation of the effect of this combination on pain and quality of life. Spine (Phila Pa 1976). 2020;45:549-554. https://doi.org/10.1097/brs.000000000000003309 46. Vleeming A, Pool-Goudzwaard AL, Hammudoghlu D, Stoeckart R, Snijders CJ, Mens JM. The function of the long dorsal sacroiliac ligament: Its implication for understanding low back pain. Spine (Phila Pa 1976). 1996;21:556-562. https://doi.org/10.1097/00007632-199603010-00005 47. Zhu W, Zhao Q, Ma R, et al. Anatomical study of the innervation of different parts of the posterior ligamentous region of the sacroiliac joint. Reg Anesth Pain Med. 2021;46:410-415. https://doi.org/10.1136/rapm-2020-102366 48. Yang AJ, McCormick ZL, Zheng PZ, Schneider BJ. Radiofrequency ablation for posterior sacroilac joint complex pain: A narrative review. PM R. 2019;11:510-S5113. https://doi.org/10.1002/pmrj.12200 49. Ivanov AA, Kiapour A, Ebraheim NA, Goel V. Lumbar fusion leads to increases in angular motion and stress across sacroiliac joint: A finite element study. Spine (Phila PA 1976). 2009;34:1862-9. https://doi.org/10.1097/BRS.0b013e3181978ea3 50. Kiapour A, Joukar A, Elgafy H, Erbulut DU, Agarwal AK, Goel VK. Biomechanics of the Sacroiliac Joint: Anatomy, Function, Biomechanics, Sexual Dimorphism, and Causes of Pain. Int J Spine Surg. 2020;14:3-13. https://doi.org/10.14444/6077 51. Ebraheim NA, Elgafy H, Semaan HB. Computed tomographic findings in patients with persistent sacroiliac pain after posterior iliac graft harvesting. Spine (Phila Pa 1976). 2000;25:2047-2051. https://doi.org/10.1097/00007632-200008150-00008 52. Delawi D, Dhert WJ, Castelein RM, Verbout AJ, Oner FC. The incidence of donor site pain after bone graft harvesting from the posterior iliac crest may be overestimated: A study on spine fracture patients. Spine (Phila PA 1976). 2007;32:1865-1868. https://doi.org/10.1097/BRS.0b013e318107674e 53. Liliang PC, Lu K, Liang CL, Tsai YD, Wang KW, Chen HJ. Sacroiliac joint pain after lumbar and lumbosacral fusion: Findings using dual sacroiliac joint blocks. Pain Med. 2011;12:565-570. https://doi.org/10.1111/j.1526-4637.2011.01087.x 44. Katz V, Schofferman J, Reynolds J. The sacroiliac joint: A potential cause of pain after lumbar fusion to the sacrum. J Spinal Disord Tech. 2003;16:96-99. https://doi.org/10.1097/00024720-200302000-00015 55. DePalma MJ, Ketchum JM, Saullo TR. Etiology of chronic low back pain in patients having undergone lumbar fusion. Pain Med. 2011;12:732-739. https://doi.org/10.1111/j.1526-4637.2011.01098.x 56. Maigne JY, Planchon CA. Sacroiliac joint pain after lumbar fusion. A study with anesthetic blocks. Eur Spine J. 2005;14:654-658. https://doi.org/10.1007/s00586-004-0692-6 57. Fortin JD, Falco FJ. The Fortin finger test: An indicator of sacroiliac pain. Am J Orthop (Belle Mead NJ). 1997;26:477-480. 58. Murakami E, Aizawa T, Noguchi K, Kanno H, Okuno H, Uozumi H. Diagram specific to sacroiliac joint pain site indicated by one-finger test. J Orthop Sci. 2008;13:492-497. https://doi.org/10.1007/s00776-008-1280-0 59. van der Wurff P, Buijs EJ, Groen GJ. Intensity mapping of pain referral areas in sacroiliac joint pain patients. J Manipulative Physiol Ther. 2006;29:190-195. https://doi.org/10.1016/j.jmpt.2006.01.007 60. Fukui S, Nosaka S. Pain patterns originating from the sacroiliac joints. J Anesth. 2002;16:245-247. https://doi.org/10.1007/s005400200033 61. Kurosawa D, Murakami E, Aizawa T. Referred pain location depends on the affected section of the sacroiliac joint. Eur Spine J. 2015;24:521-527. https://doi.org/10.1007/s00586-004-0692-6 62. Kurosawa D, Murakami E, Ozawa H, et al. A diagnostic scoring system for sacroiliac joint pain originating from the posterior ligament. Pain Med. 2017;18:228-238. https://doi.org/10.1093/pm/pnw117

63. Jung JH, Kim HI, Shin DA, et al. Usefulness of pain distribution pattern assessment in decision-making for the patients with lumbar zygapophysical and sacroiliac joint arthropathy. J Korean Med Sci. 2007;22:1048-1054. https://doi.org/10.3346/jkms.2007.22.6.1048 64. Chou LH, Slipman CW, Bhagia SM, et al. Inciting events initiating injection-proven sacroiliac joint syndrome. Pain Med. 2004;5:26-32. https://doi.org/10.1111/j.1526-4637.2004.04009.x 65. Cohen SP. Sacroilac joint pain: A comprehensive review of anatomy, diagnosis, and treatment. Anesth Analg. 2005;101:1440-1453. https://doi.org/10.1213/01.Ane.0000180831.60169.Ea 66. Young S, April C, Laslett M. Correlation of clinical examination characteristics with three sources of chronic low back pain. Spine J. 2003;3:460-465. https://doi.org/10.1016/s1529-9430(03)00151-7 67. Dreyfuss P, Michaelsen M, Pauza K, McLarty J, Bogduk N. The value of medical history and physical examination in diagnosing sacroiliac joint pain. Spine (Phila PA 1976). 1996;21:2594-2602. https://doi.org/10.1097/00007632-199611150-00009 68. Laslett M, Young SB, April CN, McDonald B. Diagnosing painful sacroiliac joints: A validity study of a McKenzie evaluation and sacroiliac private vocation tests. Aust J Physiother. 2003;49(2):89-97. https://doi.org/10.1016/s0004-9514(14)60125-2 69. Bittar M, Deodhar A. Axial spondyloarthritis: A review. JAMA. 2023;333:408-420. https://doi.org/10.1001/jama.2024.20917 70. Kim YH, Yao Z, Kim K, Park WM. Quantitative investigation of ligament strains during physical tests for sacroiliac joint pain using finite element analysis. Man Ther. 2014;19:235-241. https://doi.org/10.1016/j.math.2013.11.003 71. Han CS, Hancock MJ, Sharma S, et al. Low back pain of disc, sacroiliac joint, or facet joint origin: A diagnostic accuracy systematic review. EClinical Medicine. 2023;59:101960. https://doi.org/10.1016/j.eclinm.2023.101960 72. Mekhail N, Saweris Y, Sue Mehanny D, Makarova N, Guirguis M, Costandi S. Diagnosis of sacroiliac joint pain: predictive value of three diagnostic clinical tests. Pain Pract. 2021;21:204-214. https://doi.org/10.1111/papr.12950 73. Cohen SP, Doshi TL, Kurihara C, et al. Multicenter study evaluating factors associated with treatment outcome for low back pain injections. Reg Anesth Pain Med. 2022;47:89-99. https://doi.org/10.1136/rapm-2021-103247 74. Schneider BJ, Ehsanian R, Rosati R, Huynh L, Levin J, Kennedy DJ. Validity of physical exam maneuvers in the diagnosis of sacroiliac joint pathology. Pain Med. 2020;21:255-260. https://doi.org/10.1093/pm/onz183 75. Vleeming A, de Vries HJ, Mens JM, van Wingerden JP. Possible role of the long dorsal sacroiliac ligament in women with peripartum pelvic pain. Acta Obstet Gynecol Scand. 2002;81:430-436. https://doi.org/10.1034/j.1600-0412.2002.810510.x 76. Saueressig T, Owen PJ, Diemer F, Zebisch J, Belavy DL. Diagnostic accuracy of clusters of pain provocation tests for detecting sacroiliac joint pain: Systematic review with meta-analysis. J Orthop Sports Phys Ther. 2021;51:422-431. https://doi.org/10.2519/jospt.2021.10469 77. van der Wurf P, Hagmeijer RH, Meyne W. Clinical tests of the sacroiliac joint. A systematic methodological review. Part 1: Reliability. Man Ther. 2000;5:30-36. https://doi.org/10.1054/math.1999.0228 78. Kokmeyer DJ, Van der Wurff P, Aufdemkampe G, Fickenscher TC. The reliability of multitest regimens with sacroiliac pain provocation tests. J Manipulative Physiol Ther. 2002;25:42-48. https://doi.org/10.1067/mmt.2002.120418 79. Slipman CW, Jackson HB, Lipetz JS, Chan KT, Lenrow D, Vresilovic EJ. Sacroiliac joint pain referral zones. Arch Phys Med Rehabil. 2000;81:334-338. https://doi.org/10.1016/s0003-9993 (00)90080-7 80. Maigne JY, Aivaliklis A, Pfefer F. Results of sacroiliac joint double block and value of sacroiliac pain provocation tests in 54 patients with low back pain. Spine (Phila PA 1976). 1996;21:1889-1892. https://doi.org/10.1097/00007632-199608150-00012

Pain Medicine, 2025, Vol. 00, No. 0

81. Broadhurst NA, Bond MJ. Pain provocation tests for the assessment of sacroiliac joint dysfunction. J Spinal Disord. 1998;11:341-345. 82. Slipman CW, Sterenfeld EB, Chou LH, Herzog R, Vresilovic E. The predictive value of provocative sacroiliac joint stress maneuvers in the diagnosis of sacroiliac joint syndrome. Arch Phys Med Rehabil. 1998;79:288-292. https://doi.org/10.1016/s0003-0993 (98)90008-9 83. Szadek KM, van der Wurff P, van Tulder MW, Zuurmond WW, Perez RS. Diagnostic validity of criteria for sacroiliac joint pain: A systematic review. J Pain. 2009;10:354-368. https://doi.org/10.1016/j.jpain.2008.09.014 84. Stanford G, Burnham RS. Is it useful to repeat sacroiliac joint provocative tests post-block? Pain Med. 2010;11:1774-1776. https://doi.org/10.1111/j.1526-4637.2010.00968.x 85. Nejati P, Sartaj E, Imani F, Moeineddin R, Nejati L, Safavi M. Accuracy of the diagnostic tests of sacroiliac joint dysfunction. J Chiropr Med. 2020;19:28-37. https://doi.org/10.1016/j.jcm.2019.12.002 86. McCombe PF, Fairbank JC, Cockersoon BC, Pynsent PB. 1989 Volvo Award in clinical sciences. Reproducibility of physical signs in low-back pain. Spine (Phila Pa 1976). 1989;14:908-918. https://doi.org/10.1097/00007632-198909000-00002 87. Laskett M, Williams M. The reliability of selected pain provocation tests for sacroiliac joint pathology. Spine (Phila PA 1976). 1994;19:1243-1249. https://doi.org/10.1097/00007632-199405310-00009 88. Potter NA, Rothstein JM. Inteterster reliability for selected clinical tests of the sacroiliac joint. Phys Ther. 1985;6:51:671-1675. https://doi.org/10.1093/ptj/65.11.1671 89. Strender LE, Sjoblam A, Sundell K, Ludwig R, Taube A. Interexaminer reliability in physical examination of patients with low back pain. Spine (Phila PA 1976). 1997;22:814-820. https://doi.org/10.1097/00007632-199704010-00021 90. Bertilson BC, Bring J, Sjöblam A, Sundell K, Strender LE. Inter-examiner reliability in the assessment of low back pain (LBP) using the Kirkaldy-Willis classification (KWC). Eur Spine J. 2006;15:1695-1703. https://doi.org/10.1007/s00586-005-0050-301.9100.7 91. Tuite MJ. Sacroiliac joint imaging. Semin Musculoskelet Radiol. 2008;12:72-82. https://doi.org/10.1055/s-2008-1067939 92. Lee KH, Lee RW, Lee KH, Park W, Kwon SR, Lim MJ. The development and validation of an ai diagnostic model for sacroili-bi. It: A deep-learning approach. Diagnostics (Basi). 2023;13:272-279. https://doi.org/10.3390/diagnostics13243643 93. Cusi M, Saunders J, Van der Wall H, Fogelman I. Metabolic disturbances identified by SPECT-CT in patients with a clinical diagnosis of sacroiliac joint incompetence. Eur Spine J. 2013;22:1674-1682. https://doi.org/10.1007/s00586-013-2725-5 94. Poddubnyy D, Sieper J. Similarities and differences between nonradiographic and radiographic axial spondyloarthritis: A clinical, epidemiological and therapeutic assessment. Curr Opin Rheumatol. 2014;26:377-383. https://doi.org/10.1097/bor.00000000000000071 95. Protopopov M, Profit F, Wichah 5, et al. Comparing MRI and conventional radiography for the detection of structural changes indicative of axial spondyloarthritis in the ASAS cohort. Rheumatology (Oxford). 2023;62:1631-1635. https://doi.org/10.1093/rheumatology/keac432 96. Poddubnyy D, Gaydukova I, Hermann KG, et al. Magnetic resonance imaging compared to conventional radiographs for detection of chronic structural changes in sacroiliac joints in axial spondyloarthritis. J Rheumatol. 2013;40:1557-1565. https://doi.org/10.3899/rheum.130141 97. Puddubny K, Jurik AG, Schiøttz-Christensen B, et al. MRI abnormalities of sacroiliac joints in early spondylarthropathy: A 1-year follow-up study. Scand J Rheumatol. 2004;33:332-338. https://doi.org/10.1080/03009740410005881 98. Yoon EC, Kim JS, Lim CH, et al. Visual scoring of sacroiliac joint/sacrum ratios of single-photon emission computed tomography; computed tomography images affords high sensitivity and negative

predictive value axial spondyloarthritis. Diagnostics (Basel). 2023;13:1725. https://doi.org/10.3390/diagnostics13101725 99. Vanaclocha V, Jordá-Gómez P, Saiz-Sapena N, Vanaclocha L, Kennedy J. Diagnostic accuracy of clinical examination to distinguish sacroiliac joint pain as a cause of chronic low back pain. Br J Neurosurg. 2024;1-8. https://doi.org/10.1080/02688697.2024.2433492 100. Slipman CW, Sterenfeld EB, Chou LH, Herzog R, Vresilovic E. The value of radionuclide imaging in the diagnosis of sacroiliac joint syndrome. Spine (Phila PA 1976). 1996;21:2251-2254. https://doi.org/10.1097/00007632-199610010-00013 101. Maigne JY, Boulahdour H, Chatellier G. Value of quantitative radionuclide bone scanning in the diagnosis of sacroiliac joint syndrome in 32 patients with low back pain. Eur Spine J. 1998;7:328-331. https://doi.org/10.1007/s005860050083 102. Elgafy H, Semaan HB, Ebraheim NA, Coombs RJ. Computed tomography findings in patients with sacroiliac pain. Clin Orthop Relat Res. 2001;382:112-118. https://doi.org/10.1097/00003086-200101000-00017 103. Fritz J, Henes JC, Thomas C, et al. Diagnostic and interventional MRI of the sacroiliac joints using a 1.5-T open-bore magnet: A one-stop-shopping approach. AJR Am J Roentgenol. 2008;191:1717-1724. https://doi.org/10.2214/ajr.08.1075 104. Tofuku K, Koga H, Komiya S. The diagnostic value of single-photon emission computed tomography/computed tomography for severe sacroiliac joint dysfunction. Eur Spine J. 2015;24:859-863. https://doi.org/10.1007/s00586-014-3375-y.105. Saunders J, Cusi M, Van der Wall H. What's old is new again: The sacroiliac joint as a cause of lateralizing low back pain. Tomography. 2018;4:72-77. https://doi.org/10.18383/j.tom.2018.00011 106. Bradley KC. The anatomy of backache. Aust N Z J Surg. 1974;44:227-232. https://doi.org/10.1111/j.1445-2197.1974.tb04409.x.107. Cox M, Ng, Mashriqi F, et al. Innervation of the anterior sacroiliac joint. World Neurosurg. 2017;107:750-752. https://doi.org/10.1016/j.wneu.2017.08.062 108. Cox RC, Fortin JD. The anatomy of the lateral branches of the sacral dosral rami: Implications for radiofrequency ablation. Pain Phys. 2014;17:459-464. 109. Grob KR, Nehuber WL, Kissing RO. [Innervation of the sacroiliac joint of the human]. Z Rheumatol. 1995;54:117-122. 110. Horwitz MT. The anatomy of (A) the lumbosaral nerve plexus-its relation to variations of vertebral segmentation, and (B), the posterior sacral nerve plexus. Anat Rec. 1939;74:91-107. 111. Ikeda R. [Innervation of the sacroiliac joint. Macroscopical and histological studies]. Nihon Ika Daigaku Zasshi. 1991;58:587-596. https://doi.org/10.1272/jnms1923.58.587 112. McGrath MC, Zhang M. Lateral branches of dorsal sacral nerve plexus and the long posterior sacroiliac ligament. Surg Radiol Anat. 2005;27:327-330. https://doi.org/10.1007/s00276-005-0331-x 113. Roberts SL, Burnham RS, Ravichandiran K, Agur AM, Loh EY. Cadaveric study of sacroiliac joint innervation: Implications for diagnostic blocks and radiofrequency ablation. Reg Anesth Pain Med. 2014;39:456-464. https://doi.org/10.1097/aap.00000000000000156 114. Solonen KA. The sacroiliac joint in the light of anatomical, roent- genological and clinical studies. Acta Orthop Scand Suppl. 1957;27:1-127. 115. Steinke H, Saito T, Kuehner J, et al. Sacroiliac innervation. Eur Spine J. 2022;31:2831-2843. https://doi.org/10.1007/s00586-022-07353-1 116. Vilensky JA, O'Connor BL, Fortin JD, et al. Histologic analysis of neural elements in the human sacroiliac joint. Spine (Phila PA 1976). 2002;27:1202-1207. https://doi.org/10.1097/00007632-200206010-00012 117. Dreyfuss P, Henning T, Malladi N, Goldstein B, Bogduk N. The ability of multi-site, multi-depth sacral lateral branch blocks to anes- thatize the sacroiliac joint complex. Pain Med. 2009;10:679-688. https://doi.org/10.1111/j.1526-4637.2009.00631.x

Pain Medicine, 2025, Vol. 00, No. 0

118. Dreyfuss P, Snyder BD, Park K, Willard F, Carreiro J, Bogduk N. The ability of single site, single depth sacral lateral branch blocks to anesethetize the sacroiliac joint complex. Pain Med. 2008;9:844-850. https://doi.org/10.1111/j.1526-4637.2008.00517.x 119. Cohen SP, Strassels SA, Kurihara C, et al. Outcome predictors for sacroiliac joint (lateral branch) radiofrequency denervation. Reg Anesth Pain Med. 2009;34:206-214. https://doi.org/10.1097/AAP.0b013e3181958f4b 120. CMS.LCD-Sacroiliac Joint Injections and Procedures (L39383). Centers for Medicare and Medicaid Services. 2024. Accessed July 17, 2024. https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdId=39383&ver=9 121. Lee DW, Cheney C, Sherwood D, et al. The effectiveness and safety of sacral lateral branch radiofrequency neurotomy (SLBRFN): A systematic review. Interv Pain Med. 2023;2:100259. 122. Roberts SL, Stout A, Loh EY, Swain N, Dreyfuss P, Agur AM. Anatomical comparison of radiofrequency ablation techniques for sacroiliac joint pain. Pain Med. 2018;19:1924-1943. https://doi.org/10.1093/pm/px329 123. Schneider BJ, Huynh L, Levin J, Rinkaekan P, Kordi R, Kennedy DJ. Does immediate pain relief after an injection into the sacroiliac joint with anesthetic and corticosteroid predict subsequent pain relief? Pain Med. 2018;19:244-251. https://doi.org/10.1093/pm/px104 124. North RB, Kidd DH, Zahurak M, Piantadosi S. Specificity of diagnostic nerve blocks: A prospective, randomized study of sciatica due to lumbosacral spine disease. Pain. 1996;65:77-85. https://doi.org/10.1016/0304-3959(95)00170-0 125. Carette S, Marcoux S, Truchon R, et al. A controlled trial of corticosteroid injections into facet joints for chronic low back pain. N Engl J Med. 1991;325:1002-1007. https://doi.org/10.1056/NEJM199110033251405 126. Barnsley L, Lord SM, Wallis BJ, Bogduk N. Lack of effect of intraarticular corticosteroids for chronic pain in the cervical zyg-apophyseal joints. N Engl J Med. 1994;330:1047-1050. https://doi.org/10.1056/NEJM199404143301504 127. Yilmaz E, D'Antoni AV, Olewnik L, et al. Are the nerves supplying the anterior sacroiliac joint nociceptive? Folia Morphol (Warsz). 2023;82:96-101. https://doi.org/10.5603/FM.a2022.0009 128. Manchikanti L, Singh V, Pampati V, et al. Evaluation of the relative contributions of various structures in chronic low back pain. Pain Phys. 2001;4:308-316. 129. Fortin JD, April CN, Ponthieux B, Pier J. Sacroiliac joint: pain referral maps upon applying a new injection/arthrography technique. 103; Fortin JD, Dwyer AP, West S, Pier J. Sacroiliac joint: Pain referral maps upon applying a new injection/arthrography technique. Part I: Asymptomatic volunteers. Spine (Phila PA 1976). 1994;19:1475-1482. 131. Engel A, MacVicar J, Bogduk N. A philosophical foundation for diagnostic blocks, with criteria for their validation. Pain Med. 2014;15:998-1006. https://doi.org/10.1111/pme.12436 132. Vleeming A, Schuenke MD, Masi AT, Carreiro JE, Danneels L, Willard FH. The sacroiliac joint: An overview of its anatomy, function and potential clinical implications. J Anat. 2012;221:537-567. https://doi.org/10.1111/j.1469-7580.2012.01564.x 133. Fortin JD, Tolchin RB. Sacroiliac arthrograms and postarthography computerized tomography. Pain Phys. 2003;6:287-290. 134. Visser LH, Woudenberg NP, de Bont J, et al. Treatment of the sacroiliac joint in patients with leg pain: A randomized-controlled trial. Eur Spine J. 2013;22:2310-2317. https://doi.org/10.1007/s00586-013-2833-2 135. Sakamoto N, Yamashita T, Takebayashi T, Sekine M, Ishii S. An electrophysiologic study of mechanoreceptors in the sacroiliac joint

and adjacent tissues. Spine (Phila PA 1976). 2001;26:E468-E471. https://doi.org/10.1097/00007632-200110150-00008 136. Murakami E, Kurosawa D, Aizawa T. Treatment strategy for sacroiliac joint-related pain at or around the posterior superior iliac spine. Clin Neurol Neurosurg. 2018;165:43-46. https://doi.org/10.1016/j.clinneuro.2017.12.017 137. Murakami E, Tanaka Y, Aizawa T, Ishizuka M, Kokubun S. Effect of periarticular and intraarticular lidocaine injections for sacroiliac joint pain: Prospective comparative study. J Orthop Sci. 2007;12:274-280. https://doi.org/10.1007/s00776-007-1126-1 138. Bernard TN, Jr., Cassidy, J.D. The sacroiliac joint syndrome: Pathophysiology, diagnosis, and management. In: Frymoyer JW, et al. ed. The Adult Spine: Principles and Practice, 2nd ed. Lippincott-Raven Publishers; 1997:2343-2363. 139. Borowsky CD, Fagen G. Sources of sacroiliac region pain: Insights gained from a study comparing standard intra-articular injection with a technique combining intra- and peri-articular injection. Arch Phys Med Rehabil. 2008;89:2048-2056. https://doi.org/10.apmr.2008.06.006 140. Laslett M, April CN, McDonald B, Young SB. Diagnosis of sacroiliac joint pain: Validity of individual provocation tests and composites of tests. Man Ther. 2005;10:207-218. https://doi.org/10.1016/j.math.2005.01.003 141. Shanthanna H, Busse J, Wang L, et al. Addition of corticosteroids to local anaesthetics for chronic non-cancer pain injections: A systematic review and meta-analysis of randomised controlled trials. Br J Anaesth. 2020;125:779-801. https://doi.org/10.1016/j.bja.2020.06.062 142. Karabacakoglu A, Karaköse S, Ozerbil OM, Odev K. Fluoroscopy-guided intraarticular corticosteroid injection into the sacroiliac joints in patients with ankylosing spondylitis. Acta Radiol. 2002;43:425-427. https://doi.org/10.1080/j.1600-0455.2002.430415.x 143. Maugars Y, Mathis C, Vilon P. Prost A. Corticosteroid injection of the sacroiliac joint in patients with seronegative spondylarthropathy. Arthritis Rheum. 1992;35:564-568. https://doi.org/10.1002/Art.1780350512 144. Cohen SP, Larkin TM, Barna SA, Palmer WE, Hecht AC, Stojanovic MP. Lumbar discography: A comprehensive review of outcome studies, diagnostic accuracy, and principles. Reg Anesth Pain Med. 2005;30:163-183. https://doi.org/10.1016/j.rapm.2004.10.006 145. Hermans SMM, Knoef RJH, Schuermans VNE, et al. Double-center observation study of minimally invasive sacroiliac joint fusion for sacroiliac joint dysfunction: One-year results. J Orthop Surg Res. 2022;17:570. https://doi.org/10.1186/s13018-022-03466-x 146. Polly DW, Swofford J, Whang PG, et al. INSITE Study Group. Two-year outcomes from a randomized controlled trial of minimally invasive sacroiliac joint fusion vs. non-surgical management for sacroiliac joint dysfunction. Int J Spine Surg. 2016;10:28. https://doi.org/10.14444/3028 147. Rosenberg JM, Quint TJ, de Rosayro AM. Computerized tomographic localization of clinically-guided sacroiliac joint injections. Clin J Pain. 2000;16:18-21. https://doi.org/10.1097/00002508-200003000-00004 148. Waisbrod H, Krainick JU, Gerbershagen HU. Sacroiliac joint arthrodesis for chronic lower back pain. Arch Orthop Trauma Surg (1978). 1987;106:238-240. https://doi.org/10.1007/fb00450461 149. Schütz U, Grob D. Poor outcome following bilateral sacroiliac joint fusion for degenerative sacroiliac joint syndrome. Acta Orthop Belg. 2006;72:296-308. 150. Stelzer W, Stelzer D, Stelzer E, et al. Success rate of intra-articular sacroiliac joint injection: fluoroscopy vs ultrasound guidance—A cadaveric study. Pain Med. 2019;20:1890-1897. https://doi.org/10.1093/pm/pnz059 151. Soneji N, Bhatia A, Seib R, Tumber P, Dissanayake M, Peng PW. Comparison of fluoroscopy and ultrasound guidance for

Pain Medicine, 2025, Vol. 00, No. 0

sacroiliac joint injection in patients with chronic low back pain. Pain Pract. 2016;16:337-544. https://doi.org/10.1111/papr.12304 152. Hansen HC. Is fluoroscopy necessary for sacroiliac joint injections? Pain Phys. 2003;6:155-158. 153. Hasoon J, Chung M, Yazdi C, Robinson CL. Evaluation of land-mark-based and fluoroscopic-guided sacroiliac joint injections-A pilot series. Orthop Rev (Pavia). 2025;17:143090. https://doi.org/10.2965/001c.143090 154. Jee LH, Lee JH, Park KD, Ahn J, Park Y. Ultrasound-guided versus fluoroscopy-guided sacroiliac joint intra-articular injections in the noninflammatory sacroiliac joint dysfunction: A prospective, randomized, single-blinded study. Arch Phys Med Rehabil. 2014;95:330-337. https://doi.org/10.1016/j.apmr.2013.09.02155 155. De Luigi AJ, Saini V, Mathur R, Saini A, Yokel N. Assessing the accuracy of ultrasound-guided needle placement in sacroiliac joint injections. Am J Phys Med Rehabil. 2019;98:666-670. https://doi.org/10.1097/phm.00000000000001167 156. Fouad AZ, Ayad AE, Tawfik KAW, Mohamed EA, Mansour MA. The success rate of ultrasound-guided sacroiliac joint injections in sacroiliitis: Are we getting better? Pain Pract. 2021;21:404-410. https://doi.org/10.1111/papr.12967 157. Pekkafahli MZ, Kiralp MZ, Baekim CC, et al. Sacroiliac joint injections performed with sonographic guidance. J Ultrasound Med. 2003;22:553-559. https://doi.org/10.7863/jum.2003.22.6.553 158. Hartung W, Ross CJ, Straub R, et al. Ultrasound-guided sacroiliac joint injection in patients with established sacroiliitis: Precise IA injection verified by MRI scanning does not predict clinical outcome. Rheumatology (Oxford). 2010;49:1479-1482. https://doi.org/10.1039/rheumatology/epf424 159. Hynes JP, Flhatharta MN, Ryan JW, MacMahon PJ, Eustace SJ, Kavanagh EC. Complications in image-guided musculoskeletal injections. Skeletal Radiol. 2021;50:343-349. https://doi.org/10.1007/s00256-020-03565-y 160. Klauser A, De Zordo T, Feuchtner G, et al. Feasibility of ultrasound-guided sacroiliac joint injection considering sonoatomic landmarks at two different levels in cadavers and patients. Arthritis Rheum. 2008;59:1618-1624. https://doi.org/10.1002/art.24204 161. Perry JM, Colberg RE, Dault SL, BeonP, DP, Tresgallo RA. 37d. A cadaveric study assessing the accuracy of ultrasound-guided sacroiliac joint injections. PM R. 2016;8:1168-1172. https://doi.org/10.1016/j.pmr.2016.05.002 162. Bessar AAA, Arnaout MM, Basha MAA, Shaker SE, Elsayed AE, Bessar MA. Computed tomography versus fluoroscopic guided-sacroiliac joint injection: A prospective comparative study. Insights Imaging. 2021;12:38. https://doi.org/10.1186/s13244-021-00982-y 163. Arner J, Cakir B, Reichel H, Lattig F. Radiation dose reduction in CT-guided sacroiliac joint injections to levels of pulsed fluoroscopy: A comparative study with technical considerations. J Pain Res. 2012;5:265-269. https://doi.org/10.2147/prj.S34429 164. Althoff CE, Bollow M, Feist E, et al. CT-guided corticosteroid injection of the sacroiliac joints: Quality assurance and standardized prospective evaluation of long-term effectiveness over six months. Clin Rheumatol. 2015;34:1079-1084. https://doi.org/10.1007/s0067-015-2937-7 165. Bronsard N, Pelletier Y, Darmante H, Andréani O, de Peretti F, Trojani C. Sacroiliac joint interference after lumbosacal fusion. Orthop Traumatol Surg Res. 2020;106(6):1233-1238. https://doi.org/10.1016/j.otsr.2020.05.012 166. Paik NC. Intraarticular sacroiliac joint injection under computed tomography fluoroscopic guidance: A technical note to reduce procedural time and radiation dose. Cardiovasc. Intervent Radicol. 2016;39:1057-1060. https://doi.org/10.1007/s00270-015-1268-z 167. Oren NC, Ali S, Mok J, Lu ZF, Lee SK. Sacroiliac joint injections with the guidance of three-dimensional cone beam computed tomography (3D-CBCT) and fluoroscopy fusion. J Neurointerv

Surg. 2021;13:191-195. https://doi.org/10.1136/neurintsurg-2020-016212 168. Tantawy MF, Nazim WM. Comparison between intra-articular and combined intra- and periarticular sacroiliac injection: A prospective randomized controlled clinical trial. J Neurosurg Sci. 2024;68:294-300. https://doi.org/10.23736/s0390-5616.22.05581-3 169. Nacey NC, Patrie JT, Fox MG. Fluoroscopically guided sacroiliac joint injections: Comparison of the effects of intraarticular and periarticular injections on immediate and short-term pain relief. AJR Am J Roentgenol. 2016;207:1055-1061. https://doi.org/10.2214/ajr.15.15779 170. Bowen V, Cassidy JD. Macroscopic and microscopic anatomy of the sacroiliac joint from embryonic life until the eighth decade. Spine (Phila PA 1976). 1981;6:620-628. https://doi.org/10.1097/00007632-198111000-00015 171. Roberts SL Sacroiliac joint anatomy. Phys Med Rehabil Clin N Am. 2021;32:703-724. https://doi.org/10.1016/j.pmr.2021.05.007 172. Roberts SL, Burnham RS, Agur AM, Loh EY. A cadaveric study evaluating the feasibility of an ultrasound-guided diagnostic block and radiofrequency ablation technique for sacroiliac joint pain. Reg Anesth Pain Med. 2017;42:69-74. https://doi.org/10.1097/aap.0000000000000515 173. Schneider BJ, Ehsanian R, Levin J, Huynh L, Kennedy DJ, McCormick ZL. Intra-articular sacroiliac joint needle placement: ultrasound, fluoroscopy, and the criterion standard. Am J Phys Med Rehabil. 2020;99: 79. https://doi.org/10.1097/phm.00000000000001288 174. Scholten PM, Patel SI, Christos PJ, Singh JR. Short-term efficacy of sacroiliac jointorticosteroid injection based on arthrographic contrast patterns. PM R. 2015;7:385-391. https://doi.org/10.1016/j.pmr.2014.10.007 175. Hong SH, Chung H, Lee CH, Kim YH. A prospective randomized noniferiority trial comparing upper and lower one-third joint approaches for sacroiliac joint injections. Pain Physician. 2018;21:251-258. 176. Kurosawa D, Murakami E, Aizawa T. Fluoroscopy-guided sacroiliac intraarticular injection via the middle portion of the joint Pain Med. 2017;18:1642-1648. https://doi.org/10.1093/pm/pnw297 177. Kurosawa D, Murakami E, Aizawa T, Watanabe T, Yabe Y. Criteria for identifying technically difficult cases when performing sacroiliac intraarticular injections based on the grade of sacroiliac arthogram. Pain Med. 2020;21:2105-2110. https://doi.org/10.1093/pm/pm/aa120 178. Park J, Park HJ, Moon DE, Sa GJ, Kim YH. Radiologic analysis and clinical study of the upper one-third joint technique for fluoroscopically guided sacroiliac joint injection. Pain Phys. 2015;18:495-503. 179. Finlayson RJ, Etheridge JB, Elgueta MF, et al. A randomized comparison between ultrasound- and fluoroscopy-guided sacral lateral branch blocks. Reg Anesth Pain Med. 2017;42:400-406. https://doi.org/10.1097/AAP.0000000000000569 180. Kohan L, Salajegheh R, Hamill-Ruth RJ, Yerra S, Butz J. A review and survey of policies utilized for interventional pain procedures: A need for consensus. J Pain Res. 2017;10:625-634. https://doi.org/10.2147/jpr.S126851 181. Hynes A, Vydra D, Clements N, et al. Current trends in the use of sedation during spine intervention procedures: A survey study. Pain Med. 2020;21:1086-1088. https://doi.org/10.1093/pm/pnz194 182. Kennedy DJ, Schneider B, Smuck M, Plastaras CT. The use of moderate sedation for the secondary prevention of adverse vaso-vagal reactions. Pain Med. 2015;16:673-679. https://doi.org/10.1111/pme.12632 183. Malave B, Vrooman B. Vasovagal reactions during interventional pain management procedures—A review of pathophysiology, incidence, risk factors, prevention, and management. Med Sci (Basel). 2022;10:39. https://doi.org/10.3390/medsci10030039

Pain Medicine, 2025, Vol. 00, No. 0

184. Cucuzzella TR, Delport EG, Kim N, Marley J, Pruitt C, Delport AG. A survey: Conscious sedation with epidural and zygapophyseal injections: Is it necessary? Spine J. 2006;6:364-369. https://doi.org/10.1016/j.spinee.2005.09.005 185. Manchikanti L, Giordano J. Re: Kim N, Delport E, Cucuzzella T, Marley J, Pruitt C. Is sedation indicated before spinal injections? Spine 2007; 32: E748-E52, Spine (Phila PA 1976), 2008;33:1157-1158; author reply 1158. https://doi.org/10.1097/BRS.0b013e3181669ec 186. Smith HS, Colson J, Sehgal N. An update of evaluation of intravenous sedation on diagnostic spinal injection procedures. Pain Phys. 2013;16:52E27-28-5228. 187. Cohen SP, Raja SN. Pathogenesis, diagnosis, and treatment of lumbar zygapophysical (facet) joint pain. Anesthesiology. 2007;106:591-614. https://doi.org/10.1097/000000542-200703000-00024 188. Kubulus C, Schmitt K, Albert N, et al. Awake, sedated or anaesthetised for regional anaesthesia block placements? A retrospective registry analysis of acute complications and patient satisfaction in adults. Eur J Anaesthesiol. 2016;33:715-724. https://doi.org/10.1097/eja.000000000000495 189. Kim N, Delport E, Cucuzzella T, Marley J, Pruitt C. Is sedation indicated before spinal injections? Spine (Phila PA 1976). 2007;32: E748-E752. https://doi.org/10.1097/BRS.0b013e31815b7e9f 190. Diehn FE, Geske JR, Amrami KK, et al. An audit of transforaminal epidural steroid injections without sedation: Low patient dissatisfaction and low vasovagal rates. Pain Med. 2013;14:994-998. https://doi.org/10.1111/pm.12092. 191. Frölich MA, Zhang K, Ness TJ. Effect of sedation on pain perception. Anesthesiology. 2013;118:8(61-621. https://doi.org/10.1097/ALN.0b013e318281592d 192. Geisser ME, Ranavaya M, Haig AJ, et al. A meta-analytic review of surface electromyography among persons with low back pain and normal, healthy controls. J Pain. 2005;6:711-726. https://doi.org/10.1016/j.ipain.2005.06.008 193. Long DM, BenDebba M, Torgerson WS, et al. Persistent back pain and sciatica in the United States: Patient characteristics. J Spinal Disord. 1996;9:40-58. 194. Chen Y, Vu TH, Chinchilli VM, et al. Clinical and technical factors associated with knee radiofrequency ablation outcomes: A multicenter analysis. Reg Anesth Pain Med. 2021;46:298-304. https://doi.org/10.1136/rapm-2020-102017 195. Erdek MA, Halpert DE, Gonzalez Fernandez M, Cohen SP. Assessment of celiac plexus block and neurolysis outcomes and technique in the management of refractory visceral cancer pain. Pain Med. 2010;11:92-100. https://doi.org/10.1111/j.1526-4637.2009.00756.x 196. Sawang S, Kimpee P, Itthichaikulth W, et al. Analgesic effectiveness of methoxyflurane inhaler during genicular nerve block in knee osteoarthritis: A randomized controlled trial. Reg Anesth Pain Med. 2024. https://doi.org/10.1136/rapm-2024-105777 197. American Society of Anesthesiologists CoPM. Statement on Anesthetic Care During Interventional Pain Procedures in Adults. American Society of Anesthesiologists. 2024. Accessed July 17, 2024. https://www.asahq.org/standards-and-practice-paraments-statement-on-anesthetic-care-during-interventional-pain-procedures-for-adults 198. Schneider BJ, McCormick Z, O'Brien D, Bunch M, Smith CC, Spine Intervention Society's Patient Safety Committee. Patient safety: Conscious sedation. Pain Med. 2018;19:1872-1873. https://doi.org/10.1093/pm-pn/04y08 199. Fitzgibbon DR, Rathmell JP, Michna E, Stephens LS, Posner KL, Domino KB. Malpractice claims associated with medication management for chronic pain. Anesthesiology. 2010;11:94-98-956. https://doi.org/10.1097/ALN.0b013e3181cdef98 200. Gajraj NM. Avoid excessive sedation during cervical injections. Anesthesiology. 2005;102:869; author reply 870. https://doi.org/10.1097/00000542-200504000-00032 201. Rathmell JP, Michna E, Fitzgibbon DR, Stephens LS, Posner KL, Domino KB. Injury and liability associated with cervical

procedures for chronic pain. Anesthesiology. 2011;114:918-926. https://doi.org/10.1097/ALN.0b013e31820fc7f2 202. CMS LCD 39462 - Sacroiliac Joint Injections and Procedures. Center for Medicare & Medicare Services. 2024. Accessed July 17, 2024. https://www.cms.gov/medicare-coverage-database/view/lcd.aspx?lcdid=39462&bc=0 203. Dreyfuss P, Cohen S, Chen AS, Bohart Z, Bogduk N. Is immediate pain relief after a spinal injection procedure enhanced by intravenous sedation? PM R. 2009;1:60-63. https://doi.org/10.1016/j.pmr.2008.10.006 204. Manchikanti L, Damron KS, Rivera JJ, et al. Evaluation of the effect of sedation as a confounding factor in the diagnostic validity of lumbar facet joint pain: A prospective, randomized, double-blind, placebo-controlled evaluation. Pain Phys. 204;7:411-417. 205. Healthcare Cae. Cigna Medical Coverage Policies - Musculoskeletal Sacroiliac Joint Procedures. eviCore healthcare. 2024. Accessed July 17, 2024. https://www.evicore.com/sites/default/Eleclimical guidelines/2023-08/CignaCMM-203-SacroiliacJoint-Coated/Procedures/V102023-eff05312023_pub02162023.pdf#:~:text=A%20minimum%20of%20four%20%284%29%20weeks%20of%20noninvasive,participation%20in%20a%20rehabilitative%20or%20home%20process%20program 206. Kreiner D5, Matz P, Bono CM, et al. Guideline summary review: An evidence-based clinical guideline for the diagnosis and treatment of low back pain. Spine J. 2020;20:998-1024. https://doi.org/10.1016/j.spinee.2020.04.006 207. Qaseem A, Wilt TJ, McLean RM, et al. Clinical Guidelines Committee of the American College of Physicians. Noninvasive treatments for acute, subacute, and chronic low back pain: A clinical practice guideline from the American College of Physicians. Ann Intern Med. 2017;166:514-530. https://doi.org/10.7326/m16-2367 208. MacVar J, Kreiner DS, Duszynski B, Kennedy DI. Appropriate use criteria for fluorosocopically guided diagnostic and therapeutic sacroilac interventions: Results from the spine intervention society convened multispecially collaborative. Pain Med. 2017;18:2081-2095. https://doi.org/10.1093/pm/pnx253 209. Ramiro S, Nikiporou E, Sepriano A, et al. ASAS-EULAR recomendations for the management of axial spondyloarthritis: 2022 update. Ann Rheum Dis. 2023;82:19-34. https://doi.org/10.1136/ard-2022-223296 210. Enthoven WT, Roelofs PD, Deyo RA, van Tulder MW, Koes BW. Non-steroidal anti-inflammatory drugs for chronic low back pain. Cochrane Database Syst Rev. 2016;2:Cd012087. https://doi.org/10.1002/14651858.Cd012087 211. Migliorini F, Maffulli N, Eschweiler J, et al. The pharmacological management of chronic lower back pain. Expert Opin Pharmacother. 2021;22:109-119. https://doi.org/10.1080/14656566.2020.1817384 212. Chou R, Deyo R, Friedly J, et al. Systemic pharmacologic therapies for low back pain: A systematic review for an American College of Physicians Clinical Practice Guideline. Ann Intern Med. 2017;166:480-492. https://doi.org/10.7326/m16-2485 213. Bediawi MK, Sari I, Wallis D, et al. Clinical efficacy of celecoxib compared to acetaminophen in chronic nonspecific low back pain: Results of a randomized controlled trial. Arthritis Care Res (Hoboken). 2016;68:845-852. https://doi.org/10.1002/acr.22753 214. Ferreira GE, McLachlan AJ, Lin CC, et al. Efficacy and safety of antidepressants for the treatment of back pain and osteoarthritis: Systematic review and meta-analysis. BMJ. 2021;372:m4825. https://doi.org/10.1136/bmj.m4825 215. Enke O, New HA, New CH, et al. Anticonvulsants in the treatment of low back pain and lumbar radicular pain: A systematic review and meta-analysis. CMAJ. 2018;190:E786-E793. https://doi.org/10.1037/jcom.171333 216. Frerick H, Keitel W, Kuhn U, Schmidt S, Bredehorst A, Kuhlmann M. Topical treatment of chronic low back pain with a capsicum plaster. Pain. 2003;106:59-64. https://doi.org/10.1016/s0304-3959(03)00278-1

96

Pain Medicine, 2025, Vol. 00, No. 0

217. Khankhel N, Friedman BW, Baer J, et al. Topical diclofenac versus oral ibuprofen versus diclofenac+ ibuprofen for emergency department patients with acute low back pain: A randomized 1016/j.annemergmed.2024.01.037 218. Jones CMP, Day RO, Koes BW, et al. OPAL Investigators Coordinators. Opioid analgesia for acute low back pain and neck pain (the OPAL trial): A randomised placebo-controlled trial. Lancet. 2023;402:304-312. https://doi.org/10.1016/s0140-6736 (23)00404-x 219. Cd. Overdose Prevention - United States Dispensing Rate Maps. Center for Disease Control. Updated March 19, 2024. 2024. Accessed July 17, 2024. https://www.cdc.gov/overdose-prevention/data-research/facts-stus/as-dispensing-rate-maps.html?CDC_AAArefVal=https://www.cdc.gov/drugoverdose/rxrate-maps 220. Deyo RA, Smith, Johnson ES, et al. Opioids for back pain patients: Primary care prescribing patterns and use of services. J Am Board Fam Med. 2011;24:717-727. https://doi.org/10.3122/ajabfm.2011.06.100232 221. Degenhardt L, Grebely J, Stone J, et al. Global patterns of opioid use and dependence: Harms to populations, interventions, and future action. Lancet. 2019;394:1560-1579. https://doi.org/10.1016/s0140-6736(19)32229-9 222. Vleeming A, Buyruk HM, Stoeckart R, Karamursel S, Snijders CJ. An integrated therapy for peripartum pelvic instability: A study of the biomechanical effects of pelvic belts. Am J Obstet Gynecol. 1992;166:1243-1247. https://doi.org/10.1016/s0002-9378(11)90615-2 223. Hammer N, Möbius R, Schleifenbaum S, et al. Pelvic belt effects on health outcomes and functional parameters of patients with sacroiliac joint pain. PLoS One. 2015;10:0136375. https://doi.org/10.1371/journal.pone.0136375 224. Cibulka MT, Delitto A. A comparison of two different methods to treat hip pain in runners. J Orthop Sports Phys Ther. 1993;17:172-176. https://doi.org/10.2519/jospt.1993.17.4.172 225. Osterbauer PJ, De Boer KF, Widmaier R, Petermann E, Fuhr AW. Treatment and biomechanical assessment of patients with chronic sacroiliac joint syndrome. J Manipulative Physiol Ther. 1993;16:82-90. 226. Al-Subahi M, Alayat M, Alshehri MA, et al. The effectiveness of physiotherapy interventions for sacroiliac joint dysfunction: A systematic review. J Phys Ther Sci. 2017;29:1689-1694. https://doi.org/10.1589/jpts.29.1689 227. Trager RJ, Baumann AN, Rogers H, et al. Efficacy of manual therapy for sacroiliac joint pain syndrome: A systematic review and meta-analysis of randomized controlled trials. J Man Manipler Ther. 2024;32:561-572. https://doi.org/10.1080/10669817.2024.2316420 228. Macedo LG, Maher CG, Latimer J, McAuley JH. Motor control exercise for persistent, nonspecific low back pain: A systematic review. Phys Ther. 2009;89:9-25. https://doi.org/10.2522/ptj.20080103 229. Prather H, Bonnette M, Hunt D. Nonoperative treatment options for patients with sacroiliac joint pain. Int J Spine Surg. 2020;14:35-40. https://doi.org/10.14444/6082 230. Delitto A, Patterson CG, Stevans JM, et al. Stratified care to prevent chronic low back pain in high-risk patients: The TARGET trial. A multi-site pragmatic cluster randomized trial. EClinicalMedicine. 2021;34:100795. https://doi.org/10.1016/j.eclinm.2021.100795 231. Liu L, Pan L, Tian M, et al. Comparison of efficacy between acu-puncture therapies in improving sacroiliac joint malposition: A systematic review and meta-analysis. Biomed Res Int. 2022;2022;9485056. https://doi.org/10.11535/2022/29485056 232. Cherkin DC, Sherman KJ, Ajins, Al. et al. A randomized trial comparing acupuncture, simulated acupuncture, and usual care for chronic low back pain. Arcb Intern Med. 2009;169:858-866. https://doi.org/10.1001/archinternmed.2009.65

23.33. Vas J, Aranda JM, Modesto M, et al. Acupuncture in patients with acute low back pain: A multicentre randomised controlled clinical trial. Pain. 2012;153:1883-1889. https://doi.org/10.1016/j.pain.2012.05.033 234. Mu J, Furlan AD, Lam WY, Hsu MY, Ning Z, Lao L. Acupuncture for chronic nonspecific low back pain. Cochrane Database Syst Rev. 2020;12:Cd013814. https://doi.org/10.1002/14651858.Cd013814 235. Ho EK, Chen L, Simic M, et al. Psychological interventions for chronic, non-specific low back pain: Systematic review with network meta-analysis. BMJ. 2022;376:e067718. https://doi.org/10.1136/bmj-2021-06-07718 236. Bouvard B, Legrand E, Audran M, Chappard D. Glucocorticoid-induced osteoporosis: A Review. Clin Rev Bone Miner Metab. 2010;8:15-26. 237. Even JL, Crosby CG, Song Y, McGirt MJ, Devin CJ. Effects of epidural steroid injections on blood glucose levels in patients with diabetes mellitus. Spine (Phila PA 1976). 2012;37:E46-E50. https://doi.org/10.1097/BRS.0b013e31821fd21f 238. Younes M, Neffati F, Touzi M, et al. Systemic effects of epidural and intra-articular glucocorticoid injections in diabetic and non-diabetic patients. It Bone Spine. 2007;74:472-476. https://doi.org/10.1016/j.jbspin.2006.10.009 239. Wernecke C, Braun HJ, Dragoo JL. The effect of intra-articular corticosteroids on articular cartilage: A systematic review. Orthop J Sports Med. 2015;3:2325967115581163. https://doi.org/10.1177/2325967115581163 240. Chen AS, Solberg J, Smith C, et al. Intra-articular platelet rich plasma vs corticosteroid injections for sacroiliac joint pain: A double-blinded, randomized clinical trial. Pain Med. 2022;23:1266-1271. https://doi.org/10.1093/pm/pnab332 241. Navani A, Gupta D. Role of intra-articular platelet-rich plasma in sacroiliac joint pain. Tech Reg Anesth Pain Manage. 2015;19:54-59. 242. Singla V, Batra YK, Bharti N, Goni VG, Marwaha N. Steroid vs. platelet-rich plasma in ultrasound-guided sacroiliac joint injection for chronic low back pain. Pain Pract. 2017;17:782-791. https://doi.org/10.111/papr.12526 243. Eldin MM, Sorour OO, Hassan ASA, Baraka M, Ahmed MF. Percutaneous injection of autologous platelet-rich fibrin versus platelet-rich plasma in sacroiliac joint dysfunction: An applied comparative study. J Back Musculoskelet Rehabil. 2019;32:511-518. https://doi.org/10.3233/bmr-181366 244. Lai A, Shou J, Train SA, Barrett T. The durability and efficacy of cryopreserved human umbilical cord tissue allograft for the supplementation of cartilage defects associated with the sacroiliac joint: A case series. Reports. 2023;6:12. 245. Abdelkreem M, Hussein K, Moshrif A. Ultrasound-guided injection of low molecular weight hyaluronic acid versus steroid for inflammatory and degenerative sacroiliitis: A single-blinded randomized controlled trial. Muscle Ligaments Tendons J. 2022;12:615-620. 246. Srejc U, Calvillo O, Kabakibou K. Viscosupplementation: A new concept in the treatment of sacroiliac joint syndrome: A preliminary report of four cases. Reg Anesth Pain Med. 1999;24:84-88. https://doi.org/10.1016/s1098-7339(99)90170-0 247. Hoffman MD, Agnish V. Functional outcome from sacroiliac joint protherapy in patients with sacroiliac joint instability. Complement Ther Med. 2018;37:64-68. https://doi.org/10.1016/j.ctim.2018.01.014 248. Kim WM, Lee HG, Jeong CW, Kim CM, Yoon MH. A randomized controlled trial of intra-articular protherapy versus steroid injection for sacroiliac joint pain. J Altern Complement Med. 2010;16:1285-1290. https://doi.org/10.1089/acm.2010.0031 249. Lee JD, Lee DW, Jeong CW, Lee HG, Yoon MH, Kim WM. Effects of intraarticular protherapy on sacroiliac joint pain. Korean J Pain. 2009;22:229-233. 250. Altman RD, Devji T, Bhandari M, Fierlinger A, Niazi F, Christensen R. Clinical benefit of intra-articular saline as a

Pain Medicine, 2025, Vol. 00, No. 0

97

comparator in clinical trials of knee osteoarthritis treatments: A systematic review and meta-analysis of randomized trials. Semin Arthritis Rheum. 2016;46:151-159. https://doi.org/10.1016/j.semarthrit.2016.04.003 251. Cusi M, Saunders J, Hungerford B, Wesley-Roth T, Lucas P, Wilson S. The use of prolotherapy in the sacroiliac joint. Br J Sports Med. 2010;44:100-104. https://doi.org/10.1136/bjsm.2007.04.0424 252. Ko GD, Mindra S, Lawson GE, Whitmore S, Arseneau L. Case series of ultrasound-guided platelet-rich plasma injections for sacroiliac joint dysfunction. J Back Musculoskelet Rehabil. 2017;30:363-370. https://doi.org/10.3233/bmr-160734 253. Frederico TN, Ferraro LHC, Lemos JD, Sakata RK. Chemical neurolysis of the lateral branches of the sacral dorsal rami for the treatment of chronic pain in the sacroiliac joint: Case report and description of the technique. Pain Pract. 2022;22:134-136. https://doi.org/10.1111/papr.13046 254. Ward S, Jenson M, Royal MA, Movva V, Bhakta B, Gunyea I. Fluoroscopy-guided sacroiliac joint injections with phenol ablation for persistent sacroilitis: A case series. Pain Pract. 2002;23:332-335. https://doi.org/10.1046/j.1533-2500.2002.02043.x 255. Weksler N, Klein M, Gurevitch B, et al. Phenol neurolicase for severe chronic nonmalignant pain: Is the old also obsolete? Pain Med. 2007;8:332-337. https://doi.org/10.1111/j.1526-4637.1023336. https://2006.00228.x 256. Lee JH, Lee SH, Song SH. Clinical effectiveness of botulinum toxin A compared to a mixture of steroid and local anesthetics as a treatment for sacroiliac joint pain. Pain Med. 2010;11:692-700. https://doi.org/10.1111/j.1526-4637.2010.00838.x 257. Kumar R. Therapeutic use of botulinum toxin in pain treatment. Neuronal Signal. 2018;2:N2:20180058. https://doi.org/10.1042/29:20180058 258. Matak I, Bölcskei K, Bach-Rojecky L, Helyes Z. Mechanisms of botalinum toxin aina effect on pain. Toxins (Basel). 2019;8:459. https://doi.org/10.3390/toxins11080459 259. Zhou L, Kambin P, Casey KF, et al. Mechanism research of cryoanalgesia. Neurol Res. 1995;17:307-311. https://doi.org/10.1080/01616412.1995.11740333 260. Parvez B, Pathak V, Schubert CM, Wood M. Comparison of lesion sizes produced by cryoablation and open irrigation radio-frequency ablation catheters. J Cardiovasc Electrophysiol. May 2008;19:528-534. https://doi.org/10.1111/j.1540-8167.2007.01072.x 261. Sahoo RK, Das G, Pathak L, Dutta D, Roy C, Bhatia A. Cryoneulysis of innervation to sacroiliac joints: Technical description and initial results—A case series. A Pract. 2021;15: e01427. https://doi.org/10.1213/xaa.00000000000001427 262. Bellini M, Barbieri M. Percutaneous cryoanalgesia in pain management: A case-series. Aanestheesial Intensive Ther. 2015;47:333-335. https://doi.org/10.5603/ait.2015.0045 263. Das G, Das S, Sahoo R, Shreyas S, Kanthi B, Sharma VS. Efficacy of cryoneurolysis versus intra-articular steroid in sacroiliac joint pain: A retrospective, case-control study. Indian J Anaeth. 2023;67:1004-1008. https://doi.org/10.4103/ija.ija.938_22 264. Truong K, Meier K, Ahrens LC, et al. Cryoneulysis versus radiofrequency ablation outcome on pain experience in chronic low back pain (COPE): A single-blinded randomised controlled trial. RMD Open. 2024;10:004196. https://doi.org/10.1136/rmdopen-2024-004196 265. Diep D, Mittal N, Sangha H, Farag J. Cryoneurolysis for non-cancer knee pain: A scoping review. Interv Pain Med. 2023;2:100247. 266. King W, Ahmed SU, Baisden J, et al. Diagnosis and treatment of posterior sacroiliac complex pain: A systematic review with comprehensive analysis of the published data. Pain Med. 2015;16:257-265. https://doi.org/10.1111/pme.12630 267. Zheng Y, Gu M, Shi D, Li M, Ye L, Wang X. Tomography-guided palisade sacroiliac joint radiofrequency neurotomy versus celecoxib for ankyloving spondylitis: A open label, randomized,

and controlled trial. Rheumatol Int. 2014;34:1195-1202. https://doi.org/10.1007/s00296-014-2959-5 268. Cánovas Martínez L, Orduña Valls J, Paramés Mosquera E, Lamelas Rodríguez L, Rojas Gil S, Domínguez García M. Sacroiliac joint pain: Prospective, randomised, experimental and comparative study of thermal radiofrequency with sacroiliac joint block. Rev Esp Anestesiol Reanim. 2016;63:267-272. https://doi.org/10.1016/j.redar.2015.08.003 269. Cosman ER, Jr., Gonzalez CD. Bipolar radiofrequency lesion geometry: Implications for palisade treatment of sacroiliac joint pain. Pain Pract. 2011;11:3-22. https://doi.org/10.1111/j.1533-2500.2010.00400.x 270. Pino CA, Hoeft MA, Hofsess C, Rathmell JP. Morphologic analysis of bipolar radiofrequency lesions: Implications for treatment of the sacroiliac joint. Reg Anesth Pain Med. 2005;30:335-338. https://doi.org/10.1016/j.rapm.2005.03.014 271. Provenzano DA, Lutton EM, Somers DL. The effects of fluid injection on lesion size during bipolar radiofrequency treatment. Reg Anesth Pain Med. 2012;37:267-276. https://doi.org/10.1097/AAP.0b013e3182493bd8 272. Cohen SP, Kapural L, Kohan L, et al. Cooled radiofrequency ablation provides extended clinical utility in the management of chronic sacroiliac joint pain: 12-month follow-up results from the observational phase of a randomized, multicenter, comparative-effectiveness crossover study. Reg Anesth Pain Med. 2025. https://doi.org/10.1136/rapm-2024-106315 273. Eissa MA, moustafa a, Sawan MAS, Nazim w, Salah AM. Comparison of efficacy of lateral branch radiofrequency denerva- tion and intraarticular depot methylprednisolone injection for sacroiliac joint pain. Pan Arab J Neurosurg. 2022;17:43-49. https://doi.org/10.21608/pajn.2022.112431.1043 274. Robinson TJG, Roberts SL, Burnham RS, Loh E, Agur AM. Sacro-iliac joint sensory block and radiofrequency ablation: Assessment of bony landmarks relevant for image-guided procedures. Biomed Res Int. 2016;2016:1432074. 275. Illez OG, Atna A, Ulger EB, Kulcu DG, Ozkan FU, Aktas I. The transitional vertebra and sacroiliac joint dysfunction association. Eur Spine J. 2018;27:187-193. https://doi.org/10.1007/s00586-016-4879-4 276. Kim YH, Lee PB, Lee CJ, Lee SC, Kim YC, Huh J. Dermatome var- tion of lumbosacral nerve roots in patients with transitional lum- bosacral vertebrae. Anesth Analg. 2008;106:1279-1283, table of contents. https://doi.org/10.1213/ane.0b013e3181605635 277. Mahato NK. Complete sacralization of L5 vertebrae: Traits, dimensions, and load bearing in the involved sacra. Spine J. 2010;10:610-615. https://doi.org/10.1016/j.spinee.2010.04.01278. Dworkin RH, Turk DC, Wyrwich KW, et al. Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMPACT recommendations. J Pain. 2008;9:105-121. https://doi.org/10.1016/j.jpain.2007.09.005 279. Hägg O, Fritzell P, Nordwall A, Swedish Lumbar Spine Study Group The clinical importance of changes in outcome scores after treatment for chronic low back pain. Eur Spine J. 2003;12:12-20. https://doi.org/10.1007/s00586-002-0464-0 280. CMS LCD 39402 - Sacroiliac Joint Injections and Procedures. Centers for Medicare and Medicaid Services. Updated 10/5/ 2023. 2024. Accessed July 23, 2024. https://www.cms.gov/medi- care-coverage-database/view/lcd.aspx?lcdid=39402&ver=10& 281. Irwin RW, Watson T, Minick RP, Ambrosius WT. Age, body mass index, and gender differences in sacroiliac joint pathology. Am J Phys Med Rehabil. 2007;86:37-44. https://doi.org/10.1097/phm.0b013e31802b8554 282. van der Wurff P, Bünijs EJ, Groen GJ. A multitest regimen of pain provocation tests as an aid to reduce unnecessary minimally inva- sive sacroiliac joint procedures. Arch Phys Med Rehabil. 2006;87:10-14. https://doi.org/10.1016/j.apmr.2005.09.023 283. Lilang PC, Lu K, Weng HC, Liang CL, Tsai YD, Chen HJ. The therapeutic efficacy of sacroiliac joint blocks with triamcinolone acetonide in the treatment of sacroiliac joint dysfunction without

98

Pain Medicine, 2025, Vol. 00, No. 0

spondyloarthropathy. Spine (Phila Pa 1976). 2009;34:896-900. https://doi.org/10.1097/BRS.0b013e31819e2c78 284. Abdelrady MM, Lam KH, Shabaan N, et al. Selective ultrasound-guided nerve root block improves outcomes for disectomy in patients with cervical disc disease: A randomized, controlled, single-blinded study. Minerva Anestesiol. 2024;90:748-758. https://doi.org/10.23736/s0375-9393.24,17989-8 285. Amatto A, Burnham T, Teramoto M, Burnham R. The effectiveness and predictive factors of Sacroiliac Joint Radiofrequency Neurotomy success - A retrospective cohort study. Interv Pain Med. 2023;2:100271. https://doi.org/10.1016/j.inpm.2023.100271 286. Whang P, Cher D, Polly D, et al. Sacroiliac joint fusion using triangular titanium implants vs. non-surgical management: six-month outcomes from a prospective randomized controlled trial. Int J Spine Surg. 2015;9:6. https://doi.org/10.14444/2006 287. Sturesson B, Kools D, Pflugmacher R, Gasbarrini A, Prestamburgo D, Dengler J. Six-month outcomes from a randomized controlled trial of minimally invasive SI joint fusion with triangular titanium implants vs conservative management. Ear Spine J. 2017;26:708-719. https://doi.org/10.1007/s00586-016-4599-9 288. Randers EM, Gerdhem P, Stuge B, et al. The effect of minimally invasive sacroiliac joint fusion compared to sham operation: A double-blind randomized placebo-controlled trial. EclinicalMedicine. 2024;68:102438. https://doi.org/10.1016/j.eclinm.2024.102438 289. NASS. Sacroiliac Joint injections and Radiofrequency Ablation. Accessed February. 14, 2025. Sacroiliac Joint Injections and Radiofrequency Ablation. Accessed February. 14, 2025. Sacroiliac Joint Injections and Radiofrequency Ablation. 2026;1016/0304-3959(86)90115-6 291. Cohen SP, Hurley RW, Christo PJ, Winkley J, Mohiuddin MM, Stojanovic MP. Clinical predictors of success and failure for lumbar facet radiofrequency denervation. Clin J Pain. 2007;23:45-52. https://doi.org/10.1097/01.ajp.0000210941.04182.ea 292. Ferrante FM, Wilson SP, Iacobo C, Orav EJ, Rocco AG, Lipson S. Clinical classification as a predictor of therapeutic outcome after cervical epidural steroid injection. Spine (Pbila PA 1976). 1993;18:730-736. https://doi.org/10.1097/00007632-199305000-00010 293. Cohen SP, Wang EJ, Roybal A, Chen Y. Factors predicting outcomes from chronic pain management interventions. BMJ Med. 2025;4:e001143. https://doi.org/10.1136/hbim-2024-001143 294. Cohen SP, Abdi S. Lateral branch blocks as a treatment for sacroiliac joint pain: A pilot study. Reg Anesth Pain Med. 2003;28:113-119. https://doi.org/10.1053/rapm.2003.50029 295. Peebles R, Jonas CE. Sacroiliac joint dysfunction in the athlete: Diagnosis and management. Curr Sports Med Rep. 2017;16:336-342. https://doi.org/10.1249/jsr.0000000000000410 296. Burnham RS, Yasui Y. An alternate method of radiofrequency neurotomy of the sacroiliac joint. A pilot study of the effect on pain, function, and satisfaction. Reg Anesth Pain Med. 2007;32:12-19. https://doi.org/10.1016/j.rapm.2006.08.008 297. Casiano VE, Sarwan G, Dydyk AM, Varacallo M. Back Pain. StatPearls. StatPearls Publishing Copyright © 2024, StatPearls Publishing LLC.; 2024. 298. Yang AJ, Wagner G, Burnham T, McCormick ZL, Schneider BJ. Radiofrequency ablation for chronic posterior sacroiliac joint complex pain: A comprehensive review. Pain Med. 2021;22: S9-S13. https://doi.org/10.1093/pm/nab021 299. Cohen SP, Bajwa ZH, Kraemer JJ, et al. Factors predicting success and failure for cervical facet radiofrequency denervation: A multi-center analysis. Reg Anesth Pain Med. 2007;32:495-503. https://doi.org/10.1016/j.rapm.2007.05.009 300. Cohen SP, Williams KA, Kurihara C, et al. Multicenter, randomized, comparative cost-effectiveness study comparing 0, 1, and 1, diagnostic medial branch (facet joint nerve) block treatment paradigms before lumbar facet radiofrequency denervation.

Anesthesiology. 2010;113:395-405. https://doi.org/10.1097/ALN.010c131e33a5 301. Body BS, Sperry BP, Harper K, Macadaeg K, McCormick ZL. The relationship between patient demographic and clinical characteristics and successful treatment outcomes after basivertebral nerve radiofrequency ablation: A Pooled Cohort Study of Three Prospective Clinical Trials. Pain Med. 2022;23:S21-S13. https://doi.org/10.1093/pm/pnac050 302. Caragea M, Woodworth T, Curtis T, et al. Genicular nerve radio-frequency ablation for the treatment of chronic knee joint pain: A real-world cohort study with evaluation of prognostic factors. Pain Med. 2023;24:1332-1340. https://doi.org/10.1093/pm/pnad095 303. Keefe FJ, Rumble ME, Scipio CD, Giordano LA, Perri LM. Psychological aspects of persistent pain: Current state of the science. J Pain. 2004;5:195-211. https://doi.org/10.1016/j.jpain.2004.02.576 304. Woolf CJ, Salter MW. Neuronal plasticity: Increasing the gain in pain. Science (1979). 2000;288:1765-1769. https://doi.org/10.1126/scice.288.5472.1765 305. Kapural L. Sacroiliac joint radiofrequency denervation: Who benefits? Reg Angesth Pain Med. 2009;34:185-186. https://doi.org/10.1097/AAP.0b013e31819a94da 306. Schwarzer AC, April CN, Derby R, Fortin J, Kine G, Bogduk N. The false-positive rate of uncontrolled diagnostic blocks of the lumbar zygapophysial joints. Pain. 1994;58:195-200. https://doi.org/10.1016/0304-3959(94)90199-6 307. Simopoulos TT, Manchikanti L, Gupta S, et al. Systematic review of the diagnostic accuracy and therapeutic effectiveness of sacroiliac joint interventions. Pain Phys. 201;18:E713-E756. 308. Buijs EJ, Kamphuis ET, Groen GJ. Radiofrequency treatment of sacroiliac joint-related pain aimed at the first three sacral dorsal rami: A minimal approach. Pain Clin. 2004;16:139-146. 309. Vallejo R, Benyamin RM. Kramer J, Stanton G. Joseph NJ. Pulsed radiofrequency denervation for the treatment of sacroiliac joint syndrome. Pain Med. 2006;7:429-434. https://doi.org/10.1111/j.1526-4637.2006.00143.x 310. Edwards RR, Dworkin RH, Turk DC, et al. Patient phenotyping chronic pain treatments: IMPMACT recomendations. Pain. 2016;157:1851-1871. https://doi.org/10.1097/pn.0000000006002 311. Stout A, Dreyfuss P, Swain N, Roberts S, Loh E, Agur A. Proposed optimal fluoroscopic targets for cooled radiofrequency neotomy of the sacral lateral branches to improve clinical outcomes: An anatomical study. Pain Med. 2018;19:1916-1923. https://doi.org/10.1093/pm/pnx287 312. Sam J, Pastrak M, Duda L, et al. Clinical radiofrequency ablation outcomes of combined sensory nerve branch and dorsal entry root zone complex lesions for sacroiliac joint complex pain. Adv Ther. 2022;39:3539-3546. https://doi.org/10.1007/s12325-022-02183-5 313. Saffarian M, Christolias G, Babaria V, et al. Spine Intervention Society's Patient Safety Committee FactFinders for patient safety: Motor stimulation testing in lumbar radiofrequency neurotomy and radiofrequency neurotomy in patients with posterior hardware. Interv Pain Med. 2023;2:100170. 314. Cohen SP, Strassels SA, Kurihara C, et al. Does sensory stimulation threshold lumbar facet radiofrequency denervation outcomes? A prospective clinical correlational study. Anesth Analg. 2011;113:1233-1241. https://doi.org/10.1213/ANE.0b013e31822dd379 315. Shi W, Rudra R, Vu T-N, Gordin Y, Smith R, Kunselman AR. Sensory stimulation threshold: A viable tool to improve the outcome of lumbar facet radiofrequency denervation?. J Pain Res. 2021;14:2113-2119. https://doi.org/10.2147/JPR.S319622 316. Cedeno DL, Vallejo A, Kelley CA, Tilley DM, Kumar N. Comparison of lesion volumes and shapes produced by a radio-frequency system with a cooled, a protruding, or a monopolar probe. Pain Phys. 2017;20:E915-E922.

Pain Medicine, 2025, Vol. 00, No. 0

99

317. Cosman ER, Jr., Dolensky JR, Hoffman RA. Factors that affect radiofrequency heat lesion size. Pain Med. 2014;15:2020-2036. https://doi.org/10.1111/pm.12566 318. Zachariah C, Mayeux J, Alas G, et al. Physiological and functional responses of water-cooled versus traditional radiofrequency ablation of peripheral nerves in rats. Reg Anesth Pain Med. 2020;45:792-798. https://doi.org/10.1116/rapm-2020-101361 319. Kapural L, Minerali A, Sanders M, Matea M, Dua S. Cooled radiofrequency ablation provides prolonged pain relief compared to traditional radiofrequency ablation: A real-world, large retrospective clinical comparison from a single practice. J Pain Res. 2022;15:2577-2586. https://doi.org/10.2147/jpr.S373877 320. Tinnirello A, Barbieri S, Todeschini M, Marchesini M. Conventional (Simplicity III) and Cooled (Sinergy) Radiofrequency for Sacroiliac Joint Denervation: One-Year Retrospective Study Comparing Two Devices, Pain Med. 2017;18:1731-1744. https://doi.org/10.1093/pm/pnw333 321. Desai MJ, Safriel Y. MRI for in vivo analysis of ablation zones formed by cooled radiofrequency neurotomy to treat chronic joint pain across multiple axial spine sites. J Pain Res. 2022;15:423-430. https://doi.org/10.2147/jpr.S342795 322. Cheng J, Chen SL, Zimmerman N, Dalton JE, LaSalle G, Rosenquist R. A new radiofrequency ablation procedure to treat sacroiliac joint pain. Pain Phys. 2016;19:603-615. 323. De Andrés Ares J, Gisanz F. Randomized pragmatic pilot trial comparing perpendicular thin electrode versus parallel thick electrode approaches for lumbar medial branch neurotomy in facetogenic low back pain. Pain Pract. 2020;20:889-907. https://doi.org/10.1111/papr.12928 324. Vanneste T, Belba A, Kallewaard JW, et al. Comparison of cooled versus conventional radiofrequency treatment of the genicular nerves for chronic knee pain: A multicenter non-inferiority randomized pilot trial (COCOGEN trial), Reg Anesth Pain Med. 2023;48:197-204. https://doi.org/10.1136/rapm-2022-104054 325. De Salvatore S, Russo F, Vadala G, et al. Interventional treatments for low back pain due to sacroiliac joint dysfunction: A systematic review of the literature. Eur Spine J. 2024;33:1407-1423. https://doi.org/10.1007/s00586-024-081-09 326. Hegarty D. Clinical outcome following radiofrequency denervation for refractory sacroiliac joint dysfunction using the simplicity III probe: A 12-month retrospective evaluation. Pain Phys. 2016;19:E129-E135. 327. Dobrogowski J, Wrosek A, Wordliczek J. Radiofrequency denervation with or without addition of pentoxifylline or methylprednisolone for chronic lumbar zygapophysical joint pain. Pharmacol Rep. 2005;57:475-480. 326. Singh JR, Miccio VF, Jr., Modi DJ, sein MT. The impact of local steroid administration on the incidence of neuritis following lumbar facet radiofrequency neurotomy. Pain Phys. 2019;22:69-74. 329. Fitzpatrick B, Hilton K, Suer M, Poliak-Tunis M, Hetzel S. The impact of local corticosteroid administration on the incidence of post-neurotonomy neuritis: A prospective investigation. Pain Phys. 2022;25:E121-E126. 330. Kaye AD, Brouillette AE, Howe CA, et al. Efficacy of steroid facet joint injections for axial spinal pain and post radiofrequency ablation neuritis: A systematic review. Curr Pain Headache Rep. 2025;29:53. https://doi.org/10.1007/s11916-025-01369-0 331. Welsh SW, Varner S, Golfeiz H, Hernandez M, MT, Andrade K, Moser FG. The role of gab�antepin in the prevention of postoperative neuritis following radiofrequency ablation of cervical and lumbar medial branch nerves. Indian J Basic Appl Med Res. 2014;3:501-506. Accessed July 18, 2024, https://www.jibamr.com/assets/images/issues/pdf/501-506.pdf. 332. Ma K, Yiqun M, Wu T, et al. Efficacy of diclofenac sodium in pain relief after conventional radiofrequency denervation for chronic facet joint pain: A double-blind randomized controlled trial. Pain Med. 2011;12:27:35. https://doi.org/10.1111/j.1526-4637.2010.00978.x

333. Roark C, Whicher S, Abosch A. Reversible neurological symptoms caused by diathermy in a patient with deep brain stimulators: Case report. Neurosurgery. 2008;62:E256; discussion E256. https://doi.org/10.1227/01.Neu.0000311085.73519.B4 334. Barbieri M, Bellini M. Radiofrequency neurotomy for the treatment of chronic pain: Interference with implantable medical devices. Anaethesiol Intensive Ther. 2014;46:162-165. https://doi.org/10.5603/ait.2014.0029 335. Osborne MD. Radiofrequency neurotomy for a patient with deep brain stimulators: Proposed safety guidelines. Pain Med. 2009;10:1046-1049. https://doi.org/10.1111/j.1526-4637.2009.00686.x 336. ASA. Practice advisory for the perioperative management of patients with cardiac implantable electronic devices: Pacemakers and implantable cardioverter-defibrillators: An updated report by the American society of anesthesiologists task force on perioperative management of patients with cardiac implantable electronic devices. Anesthesiology. 2011;114:247-261. https://doi.org/10.0913e3181fbe7f6 337. Burnham T, Hilgenhurst G, McCormick ZL. Second-degree skin burn from a radiofrequency grounding pad: A case report and review of risk-mitigation strategies. PM R. 2019;11:1139-1142. https://doi.org/10.1002/pmrj.12143 338. McCormick ZL, Chung B, Smith C, Spine Intervention Society's Patient Safety Committee Internal skin burn due to novel radio-frequency ablation technology. Pain Med. 2018;19:1497-1498. https://doi.org/10.1093/pm/pny017 339. McCormick ZL, Patel J, Smith C. Inadvertent internal skin burn caused by radiofrequency electrodes during ablation procedures. Pain Med. 2020. https://doi.org/10.1093/pm/pnz375 340. McCormick ZL, Smith CC, Engel AJ, Spine Intervention Society's Patient Safety Committee Preventing External Skin Burns During Thermal Radiofrequency Neurotomy. Pain Med. 2019;20:8523. 853. https://doi.org/10.1093/pm/pny274 341. Bautista A, Dadabayev A, Rosenquist E, Cheng J. Bipolar radio-frequency neurotomy to treat neck and back pain in patients with automatic implantable cardioverter defibrillator. Pain Phys. 2016;19:E505-E509. 342. Smith C, DeFrancesch F, Patel J, Spine Intervention Society's Patient Safety Committee Radiofrequency neurotomy for facet joint pain in patients with permanent pacemakers and defibrillators. Pain Med. 2019;20:411-412. https://doi.org/10.1093/pm/pny213 343. Turtle J, Miller S, Yang A, Hurley RW, Spina N, McCormick ZL. Considerations for lumbar medial branch nerve radiofrequency at spinal motion segments adjacent to a fusion construct. Pain Med. 2023;24:165-170. https://doi.org/10.1093/pm/pnac118 344. Gazelka HM, Welch TL, Nassr A, Lamer TJ. Safety of lumbar spine radiofrequency procedures in the presence of posterior pedicle screws: Technical report of a cadaver study. Pain Med. 2015;16:877-880. https://doi.org/10.1111/pme.12678 345. Lamer TJ, Smith J, Hoelzer BC, Mauck WD, Qu W, Gazelka HM. Safety of lumbar spine radiofrequency procedures in patients who have posterior spinal hardware. Pain Med. 2016;17:1634-1637. https://doi.org/10.1093/pm/pnv078 346. Christolias G, Patel J, Nguyen MC, Smith CC, Miller DC, McCormick ZL, On Behalf of the Spine Intervention Society's Patient Safety Committee. Radiofrequency Neurotomy in Patients with Posterior Spinal Hardware. Spine Intervention Society; 2021. Accessed July 18, 2024. https://cdn.ymaws.com/ww.ipsismed.org/resource/resrmgr/factfinder/21/12.ff.rfnpos. hardware.pdf 347. Kaye AD, Manchikanti L, Novitch MB, et al. Responsible, safe, and effective use of antithrombotics and anticoagulants in patients undergoing interventional techniques: American Society of Interventional Pain Physicians (ASIPP) Guidelines. Pain Phys. 2019;22:S75-S128. 348. Narouze S, Benzon HT, Provenzano D, et al. Interventional Spine and Pain Procedures in Patients on Antiplatelet and

Pain Medicine, 2025, Vol. 00, No. 0

Anticoagulant Medications (Second Edition): Guidelines From the American Society of Regional Anesthesia and Pain Medicine, the European Society of Regional Anaesthesia and Pain Therapy, the American Academy of Pain Medicine, the International Neuromodulation Society, the North American Neuromodulation Society, and the World Institute of Pain. Reg Anesth Pain Med. 2018;43:25-262. https://doi.org/10.1097/aap.0000000000000700 349. Berstein J, Furman MB, Gilhool L, Kurup A, Bednarek C, Caplan H. The risks of continuing or discontinuing anticoagulants in diverse interventional pain procedures. [abstract]. Pain Med Meeting Abstract. 2018;19:1885-1886. 350. Endres S, Shufelt A, Bogduk N. The risks of continuing or discontinuing anticoagulants for patients undergoing common interven- tional pain procedures. Pain Med. 2017;18:403-409. https://doi.org/10.1093/pm/pnw108 351. Desai MJ, Dua S. Perineural hematoma following lumbar trans- foraminal steroid injection causing acute-on-chronic lumbar radi- culopathy: A case report. Pain Pract. 2014;14:271-277. https:// doi.org/10.1111/papr.12047 352. Gungor S, Aiyer R. Epidural hematoma development contral- teral to dura after lumbar transforaminal epidural steroid injec- tion. Pain Manag. 2017;7:367-375. https://doi.org/10.2217/pmt- 2017-0012 353. Kim GE, Hong SJ, Kang SS, Ki HJ, Park JH. Epidural hematoma treated by aspiration after transforaminal epidural steroid injec- tion – A case report. Anesth Pain Med (Seoul). 2021;16:184-190. https://doi.org/10.17085/apm.20085 354. Kim SI, Lee DH, Kim SH, Cho YH. Spinal epidural hematoma occurring at a distance from the transforaminal epidural injection site: A case report. Medicine (Baltimore). 2019;98:16654. https://doi.org/10.1097/md.00000000000016654 355. Schneider BJ, Miller DC, Mattie R, McCormick ZL, Smith C. Anticoagulants for lumbar epidural steroid injection. Pain Med. 2020;21:1294-1295. https://doi.org/10.1093/pm/pnaa087 356. Endres S, Hefti K, Schlimchen E, Bogduk N. Update of a study of not ceasing anticoagulants for patients undergoing injection pro- edures for spinal pain. Pain Med. 2020;21:918-921. https://doi.org/10.1093/pm/nz354 357. Smith CC, Schneider B, McCormick ZL, et al. Standards Division of the Spine Intervention Society. Risks and benefits of ceasing or continuing anticoagulant medication for image-guided proce- dures for spine pain: A systematic review. Pain Med. 2018;19:438-448. https://doi.org/10.1093/pm/px152 358. Society SI. Practice Guidelines for Spinal Diagnostic & Treatment Procedures. 2nd edition. Spine Intervention Society; 2014. 359. Smith CC, Miller, D.C., Saffarian, M., McCormick, ZL. I. on behalf of the Spine Intervention Society's Patient Safety Committee. Anticoagulants and Antiplatelet Agents for Lumbar Medial Branch Radiofrequency Neurotomy. Spine Intervention Society; 2020. December 2020. Accessed July 18, 2024. https:// cdn.ymaws.com/ww.ipsismed.org/resource/resgr/factfinder/ 2020/sisfactfinder2012rfa ant.pdf 360. Smith-Peterson MN, Rogers WA, End-result study of arthrodesis of the sacroiliac joint for arthritis. Traumatic and non-traumatic. J Bone It Surg Am. 1926;8:118-136. 361. Buchowski JM, Kebaish KM, Sinkov V, Cohen DB, Sieber AN, Kostuik JP. Functional and radiographic outcome of sacroiliac arthrodesis for the disorders of the sacroiliac joint. Spine J. 2005;5:520-528; discussion 529. https://doi.org/10.1016/j. spinee.2005.02.022 362. Slinkard N, Agel J, Swiontkowski MF. Documentation of out- comes for sacroiliac joint fusion: Does prior spinal fusion info- ence the outcome? Eur Spine J. 2013;22:2318-2324. https://doi.org/10.1007/s00586-013-2968-1 363. Murakami E, Kurosawa D, Aizawa T. Sacroiliac joint arthrodesis for chronic sacroiliac joint pain: An anterior approach and clinical

outcomes with a minimum 5-year follow-up. J Neurosurg Spine. 2018;29:279-285. https://doi.org/10.3171/2018.1.Spine17115 364. Smith AG, Capobianco R, Cher D, et al. Open versus minimally invasive sacroiliac joint fusion: A multi-center comparison of perioperative measures and clinical outcomes. Ann Surg Innov Res. 2013;7:14. https://doi.org/10.1186/1750-1164-7-14 365. Ledonio CG, Polly DW, Jr., Swiontkowski MF. Minimally invasive versus open sacroiliac joint fusion: Are they similarly safe and effective? Clin Orthop Relat Res. 2014;472:1831-1838. https://doi.org/10.1007/s11999-014-3499-8 366. Dengler J, Sturesson B, Kools D, et al. The iMIA Study Group. Refered leg pain originating from the sacroiliac joint: 6-month outcomes from the prospective randomized controlled iMIA trial. Acta Neurochir (Wien). 2016;158:2219-2224. https://doi.org/10.1007/s00701-016-2953-7 367. Dengler JD, Kools D, Pflugmacher R, et al. 1-year results of a randomized controlled trial of conservative management vs. minimally invasive surgical treatment for sacroiliac joint pain. Pain Phys. 2017;20:537-550. 368. Dengler J, Kools D, Pflugmacher R, et al. Randomized trial of sacroiliac joint arthrodesis compared with conservative management for chronic low back pain attributed to the sacroiliac joint. J Bone Jt Surg Am. 2019;101:400-411. https://doi.org/10.2106/jbjs.18.00022 369. Polly DW, Cher DJ, Wine KD, et al. INSITE Study Group. Randomized controlled trial of minimally invasive sacroiliac joint fusion using triangular titanium implants vs nonsurgical management for sacroiliac joint dysfunction: 12-Month outcomes. Neurosurgery. 2015;77:674-690. discussion 690-1. https://doi.org/10.1227/neu.0000000000000988 370. Vanaclocha V, Herrera JM, Sáiz-Sapena N, Rivera-Paz M, Verdú-López F. Minimally invasive sacroiliac joint fusion, radio-frequency dervation, and conservative management for sacroiliac joint pain: 6-Year comparative case series. Neurosurgery. 2018;82:48-55. https://doi.org/10.1093/neuros/nyx185 371. Al-Khayer A, Hegarty J, Hahn D, Grevitt MP. Percutaneous sacroiliac joint arthrodesis: A novel technique. J Spinal Disord Tech. 2008;21:359-363. https://doi.org/10.1097/BSD.0b013e318145ab96 372. Claus CF, Lytle E, Kaufmann A, et al. Minimally invasive sacroiliac joint fusion using triangular titanium versus cylindrical threaded implants: A comparison of patient-reported outcomes. World Neurosurg. 2020;133:e745-e750. https://doi.org/10.1016/j.wneu.2019.09.150 373. Cummings J, Jr., Capobianco RA. Minimally invasive sacroiliac joint fusion: One-year outcomes in 18 patients. Ann Surg Innov Res. 2013;7:12. https://doi.org/10.1186/1750-1164-7-12 374. Duhon BS, Bitan F, Lockstadt H, Kovalsky D, Cher D, Hillen T, SIFI Study Group. Triangular titanium implants for minimally invasive sacroiliac joint fusion: 2-Year follow-up from a prospective multicenter trial. Int J Spine Surg. 2016;10:13. https://doi.org/10.14444/3013 375. Dubon BS, Cher DJ, Wine KD, Lockstadt H, Kovalsky D, Soo CL. Safety and 6-month effectiveness of minimally invasive sacroiliac joint fusion: A prospective study. Med Devices (Auckl). 2013;6:219-229. https://doi.org/10.2147/mder.S55197 376. Schmidt GL, Bologna M, Schorr R. Patient reported clinical outcomes of minimally invasive sacroiliac joint arthrodesis. Orthop Surg. 2021;13:71-76. https://doi.org/10.1111/os.12832 377. Giannikas KA, Khan AM, Karski MT, Maxwell HA. Sacroiliac joint fusion for chronic pain: A simple technique avoiding the use of metalwork. Eur Spine J. 2004;13:253-256. https://doi.org/10.1007/s00586-003-0620-1 378. Khurana A, Guha AR, Mohanty K, Ahuja S. Percutaneous fusion of the sacroiliac joint with hollow modular anchorage screws: Clinical and radiological outcome. J Bone Joint Surg Br. 2009;91:627-631. https://doi.org/10.1302/0301-620x91b5.21519 379. Kube RA, Muir JM. Sacroiliac joint fusion: One year clinical and radiographic results following minimally invasive sacroiliac joint

References

© The Author(s) or their employers 2025. Published concurrently by Oxford University Press on behalf of the American Academy of Pain Medicine and American Society of Regional Anesthesia & Pain Medicine 2025, published by BMJ Group.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY- NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non- commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non- commercial. See https://creativecommons.org/licenses/by-nc/4.0/.

Pain Medicine, 2025, 00, 1–101

https://doi.org/10.1093/pm/pnaf129

Special Article

相关指南

• 肿瘤患者心身症状临床管理中国专家共识 · 综合与多学科 2025
• 2025 EAS共识声明：临床分期指导代谢紊乱及其后遗症的治疗 · 综合与多学科 2025
• 标准与规范 | 口服抗凝药居家管理中国专家共识（2024版） · 综合与多学科 2024
• 2024 ASCO GI 摘要集 · 综合与多学科 2024
• 继发性肢体淋巴水肿整合治疗专家共识（2025版） · 综合与多学科 2025
• 2023 ESMO 摘要集 (LBA+O+MO+POSTER) · 综合与多学科 2023
• （2026.V1）NCCN临床实践指南：成人癌痛 · 综合与多学科 2026
• 多汗诊治与管理专家共识(2025) · 综合与多学科 2025