Assessment: Use of epidural steroid injections to treat radicular lumbosacral pain

Chronic back pain and its associated disabilities represent an important health problem.¹ The rising prevalence of obesity may increase the impact of chronic back pain. The competitive nature of the modern workplace places individuals with less than perfect health and, in particular, those with painful conditions at a disadvantage. Workplace accommodation may not be an option for many occupations and, even where possible, is frequently linked with economic losses for employee and employer alike.

In 1998, individuals with back pain in the United States were estimated to have incurred total health care expenditures of $90.7 billion.² Inpatient care accounted for 31% of the expenditure, followed by expenditure for office-based visits (26%), prescription drugs (15.6%), and outpatient services (13.1%). Emergency department visits and home health visits each accounted for 3%. Of the $90.7 billion total expenditures incurred by these individuals, the expenditures attributable directly to the back pain totaled approximately $26.3 billion,² of which 42% were for office-based visits, 18% for outpatient services, 17% for inpatient care, 15% for prescription drugs, and 4% for emergency room visits. The estimated cost of treatments for spinal pain (medical therapy, epidural steroid injections, spinal cord stimulation, and intrathecal narcotics) for 1990 was at least $13 billion, growing by 7% per year.³ Medicare part B claims in 1999 for 40.4 million covered individuals were $49.9 million for lumbar epidural steroid injections, $8.5 million for lumbar facet or peri-facet joint injections, and $5.6 million for cervical or thoracic epidural steroid injections.⁴

Low back pain may occur without or with radicular features (the latter often referred to as sciatica). In the strictest sense, sciatica refers to pain running down the posterior aspect of the lower extremity. A less restrictive usage to refer to lower back pain with radiation is found in the literature reviewed for this report. A structural cause for sciatica, such as a herniated disc or foraminal stenosis, may or may not be found with investigations. Abnormalities on imaging may be seen in asymptomatic subjects, thus it may not be correct to infer a causal role for radiologic structural changes even if they are concordant with the distribution of sciatica.

Reports of epidural corticosteroid injections to treat sciatica date back to the 1950s.^5,6 Their use has increased over time despite limited quality data, as reflected by conflicting reviews of their efficacy and safety.^7–9 These reviews varied in terms of criteria for inclusion of patients, study design, types of interventions, outcome measures, and use of additional treatments. A recent review (2004) by the Technology Assessment Committee of the Institute of Clinical Systems Improvement (ICSI)¹⁰ focused on fluoroscopically guided, transforaminal epidural steroid injections in radicular lumbar pain. Although it used an evidence-based approach, the rating system is different from that of the American Academy of Neurology (AAN). It concluded that, even though results based on limited data appeared to be promising, there was insufficient evidence to comment on the efficacy of transforaminal epidural steroid injections in radicular lumbar pain.¹⁰

A recent editorial¹¹ discussing the role of placebo-controlled trials emphasizes that pain treatments considered effective based on uncontrolled clinical observations or case studies may be found to be ineffective when tested within well-designed placebo-controlled studies. To avoid this error, higher quality evidence requires performance of studies incorporating rigorous case definition, use of controls (placebo or active), use of a standardized efficacy scale, masking of patient and evaluator regarding treatment, and gathering of safety data in different treatment arms. The data should permit calculation of the number of patients needed to treat in order to make one more patient better than what would be obtained with a control treatment (placebo or alternative active treatment).

A listing of important questions regarding epidural steroids appears in table 1. However, the number of high quality of studies was limited. Therefore, the question “What is the evidence to support use of epidural steroid injections in radicular lumbosacral pain to produce pain relief?” was dealt with first. Other endpoints considered within the higher quality studies so identified were considered, as were endpoints identified by reviewers for which there were high quality data.

Methods.

Results.

Discussion.

Recommendations and conclusions.

1. Epidural steroid injections may result in some improvement in radicular lumbosacral pain when determined between 2 and 6 weeks following the injection, compared to control treatment (Level C, Class I-III evidence). The average magnitude of effect is small, and the generalizability of the observation is limited by the small number of studies, limited to highly selected patient populations, the few techniques and doses studied, and variable comparison treatments.

2. In general, epidural steroid injections for radicular lumbosacral pain have shown no impact on average impairment of function, on need for surgery, or on long-term pain relief beyond 3 months. Their routine use for these indications is not recommended (Level B, Class I-III evidence).

3. Data on use of epidural steroid injections to treat cervical radicular pain are inadequate to make any recommendation (Level U).

Recommendations for future research.

1. Further studies of the efficacy of epidural steroids for radicular lumbosacral pain should be well-designed, meeting the following criteria: a) clear case definition; b) clear measures of outcome using standardized tools, with function as the primary measure and clinically meaningful improvement in pain^23–25 as a secondary measure; c) use of a control group (placebo or active); d) prospective design; e) randomization; f) double-blind study design, so that neither patient nor assessor of measure of outcome knows the treatment arm; or triple-blind, if physician administering the epidural steroids also does not know what treatment is administered; g) adequate power; and h) adequate statistical analysis.

2. Studies of use of epidural steroids to treat radicular cervical pain or non-radicular low back or cervical pain should also be designed rigorously, meeting similar criteria.

3. The principal questions to be answered are as follows:

   • (a)What is the degree of efficacy, expressed in terms of magnitude of effect, duration of effect, and percent of patients who achieve clinically meaningful improvement, in comparison to alternative treatments?

   • (b)Using a controlled design: are there predictors of lack of efficacy or poor efficacy? Consider studying first patients without putative predictors of poor efficacy.

   • (c)How many treatments are appropriate, and at what intervals?

   • (d)How frequent are complications, and what are they?

4. Initially, it will be necessary to standardize some of the variables reflected in the questions in table 1, such as a specific technical approach, the minimal competency of the treating physician, and utilization of additional therapies.

5. Subsequently, research can be directed to evaluate the role of these variables. In particular, different techniques will need to be assessed using standardized methodology.

Mission statement of TTA.

The Therapeutics and Technology Assessment Subcommittee (TTA) produces evidence-based statements that assess the safety, utility, and effectiveness of new, emerging, or established therapeutic agents or technologies in the field of neurology. These are developed through a rigorous process of defining the topic, evaluating and rating the quality of the evidence, and translating the conclusions of the evidence into practical recommendations that can help to guide the practice of Neurology.

Disclaimer.

This statement is provided as an educational service of the American Academy of Neurology. It is based on an assessment of current scientific and clinical information. It is not intended to include all possible proper methods of care for a particular neurologic problem or all legitimate criteria for choosing to use a specific procedure. Neither is it intended to exclude any reasonable alternative methodologies. The AAN recognizes that specific patient care decisions are the prerogative of the patient and the physician caring for the patient, based on all of the circumstances involved.

Conflict of interest statement.

The American Academy of Neurology is committed to producing independent, critical and truthful clinical practice guidelines (CPGs). Significant efforts are made to minimize the potential for conflicts of interest to influence the recommendations of this CPG. To the extent possible, the AAN keeps separate those who have a financial stake in the success or failure of the products appraised in the CPGs and the developers of the guidelines. Conflict of interest forms were obtained from all authors and reviewed by an oversight committee prior to project initiation. AAN limits the participation of authors with substantial conflicts of interest. The AAN forbids commercial participation in, or funding of, guideline projects. Drafts of the guideline have been reviewed by at least three AAN committees, a network of neurologists, Neurology peer reviewers, and representatives from related fields. The AAN Guideline Author Conflict of Interest Policy can be viewed at www.aan.com. With regards to this specific report, all authors have stated that they have nothing to disclose. One of the authors performs epidural steroid injections.

Appendix 1

Therapeutics and Technology Assessment subcommittee members: Yuen T. So, MD, PhD (Co-Chair); Janis Miyasaki, MD, FAAN (Co-Chair); Douglas S. Goodin, MD (ex-officio); Carmel Armon, MD, MHS, FAAN (ex-officio); Richard M. Dubinsky, MD, MPH, FAAN; Mark Hallett, MD, FAAN; Cynthia L. Harden, MD; Michael A. Sloan, MD, MS, FAAN; James C. Stevens, MD, FAAN; Fenwick T. Nichols, III, MD; Kenneth J. Mack, MD, PhD; Paul W. O'Connor, MD; Vinay Chaudhry, MD, FAAN.

Appendix 2

AAN classification of evidence for therapeutic intervention

Class I. Prospective, randomized, controlled clinical trial with masked outcome assessment, in a representative population. The following are required: a) primary outcome(s) clearly defined; b) exclusion/inclusion criteria clearly defined; c) adequate accounting for dropouts and cross-overs with numbers sufficiently low to have minimal potential for bias; and d) relevant baseline characteristics are presented and substantially equivalent among treatment groups or there is appropriate statistical adjustment for differences.

Class II. Prospective matched group cohort study in a representative population with masked outcome assessment that meets a-d above OR a RCT in a representative population that lacks one criteria a-d.

Class III. All other controlled trials (including well-defined natural history controls or patients serving as own controls) in a representative population, where outcome is independently assessed, or independently derived by objective outcome measurement.*

Class IV. Evidence from uncontrolled studies, case series, case reports, or expert opinion.

*Objective outcome measurement: an outcome measure that is unlikely to be affected by an observer's (patient, treating physician, investigator) expectation or bias (e.g. blood tests, administrative outcome data).

Appendix 3

Classification of recommendations

A= Established as effective, ineffective, or harmful for the given condition in the specified population. (Level A rating requires at least two consistent Class I studies.)

B= Probably effective, ineffective, or harmful for the given condition in the specified population. (Level B rating requires at least one Class I study or at least two consistent Class II studies.)

C= Possibly effective, ineffective, or harmful for the given condition in the specified population. (Level C rating requires at least one Class II study or two consistent Class III studies.)

U= Data inadequate or conflicting; given current knowledge, treatment is unproven.