Complications of Sacroiliac Joint Injection and Lateral Branch Blocks, Including Water-Cooled Rhizotomy

Published on 27/02/2015 by admin

Filed under Anesthesiology

Last modified 22/04/2025

Print this page

This article have been viewed 3646 times

Chapter 16 Complications of Sacroiliac Joint Injection and Lateral Branch Blocks, Including Water-Cooled Rhizotomy

Jason E. Pope

Chapter Overview

Chapter Synopsis: Sacroiliac pain affects diverse patient populations and can be difficult to accurately diagnose. One line of treatment for sacroiliac pain in carefully selected patients can be joint injection or lateral branch block, including cooled radiofrequency (RF) rhizotomy. The most common complications arising from sacroiliac joint injections include local pain and self-resolving neuritis. Infectious risks, including viral infection, can arise but are not common. A possible complication specialized to sacroiliac interventions is suppression of the pituitary–adrenal axis.

Important Points:

Appropriate Accreditation Council for Graduate Medical Education mentored subspecialty training in interventional pain management is vital to ensure patient-centered care.

Accurate and expertly placed needles in appropriately selected patients are essential to ensure accurate diagnosis and successful treatment.

The sacroiliac joint is a large structure with rich, variable innervation. Numerous techniques have been described for intraarticular, extraarticular, lateral branch, and RF treatments.

RF treatments are not created equal, and an appreciation for each underscores their different applications and associated risks.

Common complications, regardless of RF modality used, include transient increased localized pain and usually self-limiting neuritis.

Infectious complications are uncommon if appropriate sterile technique and avoidance of multidosing from single-dose vials are used.

Injection of steroids or enteral antiinflammatory medications may help with postprocedural pain and neuritis.

Burn injuries can be avoided by understanding basic circuitry systems, careful equipment use, and correct electrosurgical unit return plate positioning.

Introduction

Estimates of sacroiliac joint sources of back and leg pain have been estimated to be between 10% and 38% using compared diagnostic injections, with a false-positive rate estimated between 0% and 53.8%.¹^–³ Not only have diagnostic provocative tests failed to be accurate,^4,⁵ but there have been numerous efforts to treat sacroiliac joint pain, including intraarticular injections, extraarticular injections, radiofrequency (RF) treatments, fusion, and prolotherapy⁶^–⁹ with level II-3 evidence for both short- and long-term relief.²

Sacroiliac pain or dysfunction has been implicated in diverse patient populations and associated morbidities, including pediatric low back pain, pregnancy, cancer, infection, ankylosing spondylitis, and inflammatory bowel disease.¹⁰^–¹³ Controversy surrounds diagnostic accuracy and technique¹⁴^–²¹ because some advocate intraarticular injections but others advocate extraarticular injections or lateral branch blocks before RF treatment. Summarily, interventions are directed to either the afferent nociceptive nerves supplying the joint or the actual joint itself. A corollary can be drawn to zygapophyseal treatments because intraarticular or median branch blocks are used before RF treatment, just as sacroiliac intraarticular injection or lateral branch blocks are performed before using RF. Although widely accepted clinically, this treatment algorithm has recently been questioned.³

As with any treatment plan, complication avoidance begins with patient selection. Treatment of patients with local infection near the injection site, coagulopathy, allergy to injectate, or comorbidities or conditions that prevent fluoroscopic needle guidance or consent should be avoided. A clear understanding of spinal anatomy and utilization of image guidance is vital to ensure both quality treatment and reduced patient morbidity and mortality. Furthermore, it should be understood that appropriate training within Accreditation Council for Graduate Medical Education accredited programs and mentorship is pivotal to ensure treatment success; interventional hobbyists only serve to undermine accessibility of these valuable therapies to patients.

Before proceeding, readers are directed to the chapters that correspond to sacroiliac joint injections; lateral branch blocks; and traditional (see Fig. 15-2), pulsed, and cooled RF treatments (see Fig. 15-5). A brief review of the differences in RF modalities are listed in Table 16-1, and reviewed elsewhere in the text (see Chapter 7). Other modalities to treat sacroiliac joint pain include fusion and prolotherapy,²² this chapter focuses on complications specific to RF neurotomy.

Table 16-1 Radiofrequency Summary

Even after appropriate safeguards and training, significant complications have been described in scattered case reports. Theoretical risks are listed in Box 16-1, and they are typically localized or systemic in nature.

Selected Complications

Meta-analysis of treatment outcomes is difficult because the treatment arm is highly variable regarding RF technique, inconsistent patient selection, outcome endpoints, and definitions of success. Although there is a plethora of literature describing sacroiliac joint interventions, few describe complications.2–4,9,10,13,23–29

Postprocedure Pain or Neuritis

Transient postprocedural pain often follows RF treatments and has been described in numerous studies for lumbar facetogenic interventions; however, few describe sacroiliac lateral branch denervation. Cooled RF and traditional thermal RF have been accompanied by postprocedure local pain (Table 16-2), typically of a transient nature,^6,^30,³¹ and Vallejo et al⁹ reported that no complications arose from pulsed RF treatments of the lateral branches.

Table 16-2 Reported Sacroiliac Radiofrequency Complications

No published study has compared the efficacy and complications of cooled versus traditional RF. In an unpublished retrospective analysis of 88 patients at the Cleveland Clinic, there was no statistically significant difference in duration of pain relief, and anecdotally, more patients who underwent cooled RF reported transient postprocedure localized back pain. Kapural et al ⁶ described transient itching, numbness, and pain.

Steroids are injected after denervation to lessen postprocedural pain. This may seem counterintuitive because the goal in thermal rhizotomy is to create a histologically detectable lesion, blocking neural afferent nociception. Dobrogowsi et al ³² investigated strategies to reduce the inflammatory pain associated with the lesioning using pentoxifylline or methylprednisolone. In a randomized prospective trial, patients were randomized to 1 mL of intraoperative methylprednisolone, pentoxifylline, or saline. No “severe local tenderness” was reported in either the methylprednisolone group or the pentoxifylline group.³² Other authors³³ contend that 3-day dosage of enteral diclofenac is effective in reducing procedural pain after conventional RF neurotomy of lumbar median branches.

False-Positive Results

False-positive results have been reported to be 0% to 53% using controlled diagnostic injections or post-block provocative sacroiliac maneuvers.^2,⁴^–⁶ ^,34 These problems may result from numerous factors, including oversedation, large-volume injectate, inaccurate needle placement, or anatomic innate defects in sacroiliac ventral or dorsal capsule. Similarly, false-negative results or treatment failures can result if interosseous or dorsal sacroiliac joint ligaments are not anesthetized because they are known to contribute to these pain complaints. Furthermore, anterior sacroiliac joint innervation is largely ignored as a significant contribution to sacroiliac joint pain.

There have been numerous techniques suggested to enter the sacroiliac joint for arthrography,¹⁴^–¹⁹ to perform lateral branch blocks,^5,^8,^34,³⁵ and for denervation.^6,^9,^11,^27,^31,³⁶^–⁴⁰ Whatever the method of diagnosing sacroiliac joint pain, accuracy hinges on small-volume injectate and controlled diagnostic comparative blocks, although cost effectiveness has recently come into question.³ Cohen et al³ suggested that the cost of RF treatment without diagnostic injection was almost $9000 cheaper, although indirect results suggest a better treatment outcome with dual diagnostic injections before neurotomy, keeping a keen eye on potential placebo cofounders.

Infections

There are scattered reports of infection for both sacroiliac injection and RF therapy. Some suggest that the thermal nature of RF treatment may decrease the chance of infection.³ Others suggest that steroids may innately increase the infectious potential because they are immunosuppressive.³⁸ Interestingly, only one report of infectious complication after block therapy was found after an exhaustive Medline search with search terms “sacroiliac and infection.” Meydani et al⁴¹ report a herpes simplex virus type II outbreak (Fig. 16-1) after sacroiliac injection therapy (Table 16-3), confirmed by unroofing vesicle and DNA analysis.

Fig. 16-1 Herpetic outbreak.

(From Meydani A, Schwartz A, Foye PM, Patel AD: Herpes simplex following intra-articular sacroiliac corticosteroid injection, Acta Dermatoven APA 16(3):135-137, 2009.)

Table 16-3 Reported Sacroiliac Joint Block Lateral Branch Block Complications

Cohen et al ³¹ described one superficial infection after RF treatment of 77 patients that resolved after a course of oral antibiotics. Single-use vials should always be used for individual patients because this was a major source of contamination in an Enterobacter aerogenes and Klebsiella pneumoniae bacteremia outbreak after sacroiliac joint injection reported in 2008.⁴²

There have been no reported cases of epidural or paraspinal abscess after sacroiliac joint interventions. The only serious infections described in the literature are those of hematogenic septic arthritis remote from sacroiliac joint interventions.⁴³

Bleeding

The incidence of hematoma after sacroiliac joint injection is unknown but is likely less than for spinal or epidural anesthetic procedures, estimated to be one in 222,000 and one in 150,000, respectively.²⁸ In fact, no cases have been reported. However, if epidural hematoma is suspected, it should be promptly investigated and treated. Symptom and sign presentation is dependent on the anatomic location of the bleed, and a high index of suspicion is needed for diagnosis. Early neurosurgical consultation is essential because prognosis requires expeditious evacuation within 8 hours of neurologic deficit presentation.²⁹

Burn Injuries

No burn injuries have been reported with sacroiliac joint denervation; however, the potential can be extrapolated from the facetogenic literature. Burn injuries were the result of equipment malfunction, insulation breaks within the electrodes, unipolar electrosurgical unit return plate, or unknown causes.⁴⁴^–⁴⁶ Appreciation of circuit differences between grounded and ungrounded systems is encouraged as operating room power supplies are typically isolated, ungrounded systems, where office buildings are typically grounded systems. By design, isolated systems are more difficult to deliver aberrant current. Furthermore, burn severity is related to current density, either greater the current or the smaller the area applied.

Pituitary–Adrenal Axis Suppression

Although numerous studies have demonstrated pituitary–adrenal suppression with epidural or intraarticular administration of glucocorticoids,⁴⁷^–⁵¹ none has investigated axis depression from sacroiliac interventions. From the available data, hypothalamic pituitary adrenal axis suppression could be assumed to be depressed for 4 weeks, and insulin sensitivity may be adversely affected for 1 week. Cushing syndrome, steroid myopathy, aseptic meningitis, and anaphylactoid reactions have been reported after single epidural steroid injections.^52,⁵³ Other glucocorticoid side effects, including musculoskeletal, ophthalmologic, gastrointestinal, cardiovascular, secondary cortisol deficiency, gynecologic, neurologic, hematologic, psychiatric, and dermatologic, have been described.⁵⁴