Access to the PPM Journal and newsletters is FREE for clinicians.
10 Articles in Volume 14, Issue #2
How Safe Is Epidural Steroid Injection? Examining Drug-Related Factors
How Important Is Evidence-Based Medicine in Epidural Injection for Low Back Pain?
Current Access to Opioids—Survey of Chronic Pain Patients
Opioid Prescribing Part 2: Appropriate Documentation of Follow-up Visits
Neuropathic Pain: A Literature Review
Translating Chronic Pain Research Into Practice: Chronic Pain and the Brain
Intractable Pain: Time To Understand and Use the Term (Again)
Are Antibiotics a Treatment Option for Low Back Pain?
Genetic Mutations in Cytrochrome P 450 2D6
Light Exercise May Lead To Faster Recovery After Concussion

How Important Is Evidence-Based Medicine in Epidural Injection for Low Back Pain?

New study sheds light on day-to-day practice versus guidelines, with variations in practice related to indications, the duration of conservative treatment before initiating injections, and the time between injections.

There are several evidence-based (EB) guidelines that address the use of epidural injections for the treatment of low back pain. But how influential are these guidelines to the everyday decision-making of physicians? To study this question, the Accreditation Association for Ambulatory Health Care (AAAHC) Institute for Quality Improvement, a subsidiary of AAAHC, conducted a study of epidural low back injections (LBI) in the ambulatory setting to see how guidelines influence day-to-day practices.

The study results suggest that while most of the practices fall within the parameters of one or more guidelines, a significant proportion do not. Areas in which day-to-day practice differs from EB guidelines included the indications/contraindications for the procedure, use and duration of conservative therapy, and number and frequency of injections.1 This article will summarize the AAAHC study results, as well as give context to the findings.

Why Evidence-Based Guidelines

Imperfect adherence to guidelines is not unique to pain management. Many guidelines are equivocal or contradictory, there often is not enough evidence to reach definitive conclusions, and there is evidence that physicians, in general, often do not follow guidelines. Moreover, in rapidly evolving technologies such as interventional pain management, the “latest” guidelines may not reflect the latest research.

Despite these concerns, guidelines play an increasing role in patient care and reimbursement decisions. This is primarily because federal and state agencies and third-party payers look to EB recommendations to guide efforts to improve quality of care and halt the precipitous increase in health care costs.

This trend has important implications for the treatment of chronic back pain. From 1997 through 2006, there was a 49% increase in the number of patients seeking spine-related treatment for back and neck pain problems, and a 65% increase in health care expenditures related to these conditions.2 A dramatic increase in interventional techniques for treating low back pain has been a major driver of these increases in costs. A recent analysis of the Medicare population documents an increase in interventional procedures for low back pain from 1,460,495 in 2000 to 4,815,673 in 2011, an increase of more than 200%. The vast majority of these interventions included epidural injections, facet joint injections, or other nerve blocks. For example, the number of epidural injections grew from 832,000 to 2.3 million over the last decade (Figure 1).3


Day-to-Day Practice for Epidural LBI

The AAAHC Institute study offers a snapshot of the practice of LBI for pain, as performed in the ambulatory setting. In 2006, 60% of LBIs were performed in an ambulatory setting.4,5 The performance measures were reported from a sampling of 623 patients treated in 31 ambulatory organizations, which performed a total of 65,000 LBI procedures per year. These organizations included freestanding ambulatory surgery centers (ASCs) and office-based practices. Multi- and single-specialty organizations were represented as well as those owned and operated by physicians or joint ventures.

The results of the survey were compared to guideline recommendations for:

  • Types of injections
  • Indications/symptoms for which LBI is appropriate
  • Contraindications
  • Duration of conservative therapy prior to LBI
  • The number and frequency of injections per patient

Types of Injections

There are 3 common methods for delivering steroid into the epidural space: the interlaminar, caudal, and transforaminal approaches (Table 1). According to the North American Spine Society, an interlaminar epidural spinal injection (ESI), involves placing the needle into the posterior epidural space and delivering the steroid over a wider area. Similarly, the caudal approach uses the sacral hiatus to allow for needle placement into the very bottom of the epidural space.6 With both approaches, the steroid spreads over several segments and can cover both sides of the spinal canal. With a transforaminal ESI, often referred to as a “selective nerve root block,” the needle is placed alongside the nerve as it exits the spine and medication is placed into the “nerve sleeve.” The medication then travels up the sleeve and into the epidural space from the side, allowing for more concentrated delivery of medications into the affected area.6

The type of procedures reflected in the survey included 49% lumbar/sacral caudal epidural injections, 38% lumbar/sacral transforaminal injections, and 17% paravertebral facet joint injections. Of note, several patients received multiple types of LBIs. Of all of the injections, 75% were considered therapeutic, 7% diagnostic, and 18% both therapeutic and diagnostic.


There is significant dispute regarding the use of LBI, including what conditions and symptoms are considered appropriate indications. In fact, even when investigators have used the same criteria, there have been conflicting conclusions.7-10

Symptoms that often are considered indicative of LBI include nonspecific back pain, limited range of motion or function, radiating pain, disc herniation (possible positive straight leg raise), post-lumbar surgery or laminectomy syndrome, and spinal stenosis. “Red flags,” such as cauda equina and suspicion of cancer, should be distinguished from other low back-related pain and treated accordingly.

Additionally, there is conflicting evidence regarding the long-term efficacy of LBIs.8-15 The most recent EB review of these procedures, the 2013 update of the Interventional Pain Management (IPM) guidelines of the American Society of Interventional Pain Physicians,16 concluded that, overall, the evidence was fair to good for 62% of diagnostic and 52% of therapeutic interventions that were assessed. The relative efficacy of the various types of epidural injections and the rated evidence were considered to be good for patients with a herniated disk or sciatica, but less so for other patient symptoms/history. Table 2 reviews the IPM Guidelines for different indications/symptoms.

The vast majority of patients in the survey, 80%, were diagnosed with radiculitis and/or disc herniation, and had a variety of presenting symptoms including: radiating pain, 65% (n=403); localized pain, 61% (n=378); herniated disc, 57% (n=354); limited range of motion, 46% (n=287); weakness, 26% (n=165); spinal tenderness, 22% (n=139); and positive straight leg test, 12% (n=77).

Furthermore, 78% of patients rated the severity of their symptoms as greater than 5 on a scale of 0 to 10 on the visual analog scale (0= no pain, 10 = most severe pain).


Commonly cited relative contraindications to LBI include signs of local or systemic infection; current use of anticoagulants; allergies to contrast agents, local anesthetics, or corticosteroids; prior complications with injections; pregnancy; diabetes; and congestive heart failure.10-12 The only absolute contraindications in the IPM guidelines are severe spinal stenosis resulting in intraspinal obstruction, infection, and predominantly psychogenic pain.16

In our study, 17% of patients (103/623) had one or more relative contraindications to LBI, but they still received treatment. For the 84 patients within this group for whom a pre- and post-procedure pain rating was recorded, 73% (61 patients) reported a decrease in their pain rating from pre- to post-procedure. Three of the 8 complicated cases in this study included patients with contraindications listed in Table 3.

Conservative Therapy Prior to LBI

Low back pain with/without lumbar radiculopathy often is treated conservatively with non-surgical treatments including a combination of rest, medication, and home exercise or structured physical therapy. Unless conservative therapy is not possible because of pain causing functional disability or other contraindications to physical therapy or non-steroidal anti-inflammatory drugs (NSAIDs), it usually is recommended prior to use of LBI. In our study, 29% of patients (172/602) had one or more of these contraindications, which prevented the use of conservative therapy.

The length of time recommended for conservative treatment varies considerably from one guideline to the next.10-15
The IPM guidelines suggest that LBI is potentially indicated with low back pain that has persisted for 3 months or longer. In our study, the length of conservative treatment was recorded for 67% of cases (287/430), with the majority of patients undergoing LBI after less than 3 months of conservative therapy (62%) (Figure 2).



Number/Frequency of Injections

In general, the number and frequency of epidural injections are not uniform among the various guidelines.11-13 The IPM guidelines for therapeutic epidural injections suggest that there should be 2 months or longer between each injection, provided that greater than 50% relief is obtained for 2 months.16 Epidural injections should be repeated in accordance with medical necessity criteria, and it is suggested that they be limited to a maximum of 4 per year. These guidelines also conclude, however, that the indications, frequency, and total number of interventions have been poorly addressed in the literature.

In our study, the number and frequency of pain injections also varied considerably among practitioners. For cases documented with information on prior pain injections, 72% (431/601) indicated that the patients had already had injections for their current symptoms. For 406 cases for which the information was available, 69 reported that they had experienced significant relief (50% to 100%) from previous injections. Seventeen percent of patients (75/431) had more than 3 injections in the prior 6 months, while 11% of patients (49/431) had more than 6 injections in the prior 12 months.

For 417 cases, information was submitted regarding whether the patient had had any complications from previous injections. Of these, 3% of patients experienced complications. For 4% of cases (16/381), patients re-injured their backs since the last injection. As many as 62% of these patients did not meet the IPM criteria of 2 month interval between injections (Figure 3).




While most procedures reported in our study were performed in compliance with EB guidelines, a significant minority was not, especially when it came to the number and frequency of injections and the length of conservative treatment prior to the intervention. There also was a significant number of patients (20%) who underwent an LBI procedure despite the fact that the evidence for those procedures was graded as only fair. To some extent, these variances can be attributed to variances in guideline recommendations, and the appropriate application of guidelines often is not clear-cut for a number of reasons including:

  • the conclusions of guideline authors often are equivocal, owing to variations in the strength of the scientific evidence used to perform guideline development, and variations in the recommendations of different guidelines;
  • numerous studies have shown that clinicians often do not follow guideline recommendations and that it can take years for guidelines to make their way into clinical practice17-19;
  • practitioners are concerned that over-reliance on guidelines can result in a “cookbook” approach to medicine, restricting their ability to make decisions that are informed by the needs and preferences of the individual patient and the experience of the health care providers.20


Guidelines can be helpful, but further information is needed when guidelines are equivocal and key issues associated with LBI are not resolved. Research to facilitate the development of more definitive EB guidelines will help to bring clarity to the best practices for our patients.


There is recognized controversy with regard to several aspects of LBI, including indications/contraindications for the procedure, use and duration of conservative therapy, and number and frequency of injections. The AAAHC Institute study has captured variations in practice associated with these issues. Additional evidence is needed to better understand these issues and ensure appropriate use of LBI.

Last updated on: May 19, 2015
close X