Breaking Bad (Tissue): Epidural Adhesiolysis and Its Outcomes

Medicine is at a crossroads as we approach the third decade of the 21st century. Intellectual and technologic advances have progressed at such a dramatic pace that much of what physicians learned only a decade ago is now obsolete. The dark side of this is that these innovations come at a cost, which includes unnecessary tests and procedures that exact a steep personal and financial toll on patients and society. For some conditions such as certain types of cancer, the benefits outweigh the financial costs. However, for low back pain, dramatic increases in the number, complexity, and costs of interventions have not been accompanied by corresponding decreases in prevalence or disability rates,1–3 and some research suggests that higher procedure rates may actually be associated with inferior outcomes.4 This augurs for publication of negative trials, which can help conserve resources and prevent unnecessary interventions, but which tend to be cited less frequently. The requirement for registration of clinical trials by many journals, including Anesthesia & Analgesia, and systematic reviews that now consider unpublished, negative data sets, has produced significant strides in remedying publication bias, although the problem still exists.5 This issue of Anesthesia & Analgesia includes a negative comparative-effectiveness study by Choi et al6 evaluating epidural lysis of adhesions (LOA) for an all-to-common condition—persistent pain after spine surgery. Epidural LOA, also known as epidural neuroplasty, was first described in the late 1980s as a means to disrupt perineural fibrosis or scarring and facilitate the spread of analgesic substances to pain generators in the epidural space.7 It is most commonly performed to alleviate pain and disability in the limbs, neck, or back associated with failed back surgery syndrome (FBSS or postlaminectomy persistent pain syndrome) or spinal stenosis.8 These conditions share pathologic characteristics that include reduction in central neuraxial or foraminal cross-sectional areas, and/or epidural scar tissue resulting in compression of the spinal cord, cauda equina, or exiting nerve roots. As our population ages and invasive, expensive interventions become more commonplace, the prevalence and socioeconomic burden of both FBSS (prevalence of 10%–50% patients after surgery)9 and spinal stenosis (prevalence 19%–47% in the elderly) have risen commensurately over the past 2 decades.10 The lack of any nonsurgical, reliable treatment options for these conditions despite hundreds of studies evaluating medications, physical therapy, and other interventions is frustrating for both patients and health care providers. This disappointment is compounded by the lack of clarity regarding indications, procedural aspects of interventions, and data evaluating both short- and long-term outcomes. LOA is offered to patients when other non- or less invasive modalities have been ineffective. It can involve injection of 1 or more of the following substances: radio-opaque contrast, local anesthetics, saline (isotonic or hypertonic), hyaluronidase, and steroids. This is most frequently accomplished by introducing a stiff, “navigable” catheter into the epidural space (usually cervical or lumbar) under fluoroscopic (and/or endoscopic) guidance that can then be maneuvered to the putative area of pathology. The benefits of LOA are believed to be secondary to dissolution of adhesions and prevention of recurrence through inhibition of fibroblasts11 and greater access for analgesic substances in the perineural spaces. Mechanical lysis of scar tissue with a catheter may or may not be necessary for epidural LOA to be effective.12 Despite its long history of use, several questions remain unanswered regarding potential responders, effectiveness, efficacy, longevity of analgesic benefit, and safety of LOA. The aforementioned study by Choi et al6 was a retrospective attempt to address some of these issues by examining patients who received LOA at 2 hospitals. The authors evaluated patient response to LOA performed with hypertonic saline infusions of either 5% or 10% concentration with the 2 cohorts being time-based and sequential. The rationale behind hypertonic saline is that it may prevent the reformation of scar tissue. However, the evidence for this is preliminary,11 and there is even stronger evidence that hyperosmotic solutions are associated with neurotoxicity, because they may affect enzyme and other cell functions and in essence “dessicate” nervous tissue.13 Clinical consequences were the focus of the authors; they selected a numeric rating scale pain score as the primary outcome variable and reported complications and procedure-related pain. Despite the retrospective nature, the relatively large cohort in this study (543 patients) should theoretically control for factors that might otherwise predispose to type I and type II statistical errors. Inclusion and exclusion criteria were robust and the authors performed multivariate analysis to gauge the impact of noninterventional factors on analgesic outcomes, yet no significant differences between groups were detected with the caveat that there was a nonsignificant trend toward higher intraprocedural pain scores in the 10% group. Given the similar benefit, the authors concluded that 5% hypertonic saline may be preferable for infusion during LOA given its theoretically improved safety margin in the event of inadvertent dural puncture resulting in intrathecal administration. The authors should be commended for collecting and analyzing data from this large cohort, but there are several limitations that warrant attention. In addition to the known drawbacks of retrospective studies (eg, observer and recall bias, missing data, inability to control for cointerventions, lack of standardization of study variables, etc), there are potential confounding factors to consider. These include liberal use of repeat injections of epidural steroids in patients in both groups after LOA, lack of data regarding oral analgesic use, and lack of comparison against epidural steroid injections or LOA with 0.9% saline. The absence of data on functional ability14 as assessed by validated tools (eg, the Oswestry Disability Index or the Roland-Morris Disability Questionnaire) is also a significant limitation of this study. The quest to evaluate different concentrations of saline for LOA is not new. A study conducted in 83 patients published in 1999 compared 0.9% and 10% saline with or without hyaluronidase for LOA. All participants obtained significant relief up to 12 months after the intervention with a trend being noted for decreased additional treatments in subjects treated with hypertonic saline (with or without hyaluronidase).15 Recent reviews of published literature of epidural LOA have come to conflicting conclusions. One review concluded that there was weak positive evidence that LOA is more effective than conventional caudal epidural steroid injections for FBSS and spinal stenosis and that LOA is more effective than sham adhesiolysis and conservative management for lumbosacral radiculopathy.8 Helm et al,12 in their systematic review and meta-analysis based on the results of 7 randomized controlled trials and 3 observational studies, concluded that there was strong evidence for the short- and long-term efficacy of percutaneous adhesiolysis in the treatment of chronic refractory low back and lower extremity pain secondary to FBSS and spinal stenosis. However, there were several limitations in this review including heterogeneous comparators in trials (epidural steroid injections, placebo procedures, physical therapy) and “double counting” of data from 2 of the included trials. Another recent narrative review assessed the evidence for LOA in FBSS and spinal stenosis, finding LOA to be more effective than epidural steroid injections in this population. Administration of a high volume of injectate and hypertonic saline in the epidural space was associated with better outcomes, and the authors found some support for use of hyaluronidase.16 If we accept the premise that LOA confers analgesia, then the next question to answer is, “Who benefits from this intervention?” A retrospective study showed better outcomes in patients older than 80 years of age and, surprisingly, those on or seeking disability or workers’ compensation.17 Variables that affect the outcome of epidural LOA can be divided into patient-related (ie, pathology, presentation, coexisting conditions) and procedure-related factors. Patient-related variables that may impact outcomes include age, gender, psychologic (eg, anxiety, depression, catastrophization), and physical (functional limitation) factors. Pathology may influence treatment results, but the impact of disease processes on analgesic response may be difficult to ascertain in a retrospective or small prospective study considering that FBSS is by far the most common indication for this procedure.17 Another patient-related factor that may have an adverse impact on outcomes of epidural LOA is the presence of aberrant central pain processing (also known as impairment of diffuse noxious inhibitory control) in patients with persistent pain.18 Pain phenotypes and measures of disease burden (eg, neuropathic versus nociceptive character, duration, intensity) are crucial when efficacy is evaluated with predominantly axial pain and higher pain scores typically associated with poorer outcomes. Procedure-related factors include number and duration of procedures (the initial regimen was infusion of hypertonic saline over 3 days), volume and composition of the injectate (eg, larger volumes, use of hypertonic saline and/or hyaluronidase, dose of steroids), and the equipment (navigable catheter, epidural endoscope, laser technology) used to mechanically disrupt scar tissue and guide medication administration. Another important outcome to consider is rate of adverse events. Although LOA is widely considered a safe procedure compared with alternative interventions such as reoperation, the complication rates are higher than for conventional epidural steroid injections.8,16 The incidence of dural puncture is approximately 4% for LOA19 versus 1% to 2% for epidural steroid injections. Other serious complications include shearing of the catheter, potential for trauma to the nerve roots, mass effect of the high volume of injectate causing nerve injury, epidural hematoma, and infection resulting in transient or long-term neurologic deficits.20 We live in an era of comparative-effectiveness research where payers want to see value for their money. Efficiency of a treatment can be defined as measurable improvement in outcomes or “outputs” such as analgesia, functional restoration, and return to a productive life divided by “inputs” (costs related to the treatment and its adverse effects). Is LOA an “efficient” treatment? Unfortunately, this question cannot be answered without first knowing whether the procedure is efficacious. Only after efficacy is determined can we begin to answer questions such as whether higher concentrations of saline are better than lower concentrations, the endoscopic approach is superior to the fluoroscopic approach, or adjuvants such as hyaluronidase should be used. High-quality randomized trials that include relevant analgesic and associated outcomes,14 and results that are confirmed by investigators from around the world rather than 1 or 2 centers, will help us to better understand the indications, efficacy, and safety of this frequently used technique that needs solid evidence to support its use. DISCLOSURES Name: Anuj Bhatia, MBBS, MD, FRCA, FRCPC, FIPP, FFPMRCA, EDRA, CIPS. Contribution: This author wrote the manuscript. Name: Ariana Nelson, MD. Contribution: This author wrote the manuscript. Name: Steven P. Cohen, MD. Contribution: This author wrote the manuscript. This manuscript was handled by: Honorio T. Benzon, MD.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.