Oh, my aching back

A few months ago, I had to undergo an L5-S1 spinal fusion. Before surgery I had a number of epidural injections to try to avoid the surgery and several studies to clearly identify the offending nerve root level. Like most patients, I said “yes” to my doctors’ suggestions at virtually every step of what to do next. I was in pain and wanted to get back to my normal life and work. Now that I am back on my feet, literally and figuratively, I pay very close attention when a request for a spine injection or a spine procedure crosses my desk. What is the evidence for effectiveness of the spinal procedures we recommend? Are we doing the right thing when we refer our patients to our interventional pain colleagues?

Epidural injections:

Epidural steroid injections (ESI) are usually done by the transforaminal approach. There have been a number of randomized trials and systemic reviews of ESI. The evidence is strong for short term (<6 week) relief and moderate for long term improvement when ESI is used to treat lumbar disc herniation causing radiculopathy. The evidence for ESI is moderate for short and long term relief of pain for cervical disc herniation causing radiculopathy. The evidence is limited for radicular pain in post-laminectomy patients and ESI has not been shown to be helpful for axial pain of the lumbar or cervical spine.

Diagnostic epidural injections can be also used to identify offending nerve roots in radicular pain patients (my diagnostic injections clearly showed L5 and not L4 to be the problem nerve root, which did fortunately minimize my eventual surgery). The good news is that if a selective lumbar epidural block (with or without steroid) is effective for one year, there is a very high chance that the patient will not need lumbar surgery for at least five years.

PHC uses InterQual criteria to determine medical necessity for ESI. The evidence supports cervical or lumbar ESI a patient over 18 years of age who has:

  • Documentation of radicular pain that is unresponsive to 30 days of conservative therapy (activity modification +/- NSAIDS) and
  • Imaging that documents nerve root compression at the appropriate level
  • Maximum of two levels per three months (two levels on same side or one level bilaterally)
  • The ESI may be repeated in two months up to a maximum of three injections per site in one year.

If imaging does not show nerve root impingement or the symptoms and imaging do not agree on the laterality or level, PHC will not approve epidural steroid injections for that patient. ESI for spinal stenosis without significant nerve root impingement is not medically indicated.

Facet Joint Injection and Median Branch Block:

Axial back pain is a common condition. Injection of the zygapophysial (facet) joints and median branch blocks have been well studied. There is moderate evidence for short and long term improvement in axial low back pain with intra-articular facet injection of anesthetics with steroids. The evidence is limited for pain relief for facet injection for neck pain. The evidence for short and long term pain relief by median branch block of the cervical, thoracic, and lumbar spine is moderate.

The evidence supports facet injection and median branch block for patients over 18 years who have:

  • Non-radicular pain consistent with facet pain for at least 12 weeks
  • MRI that excludes discitis or a spinal lesion as the cause of pain
  • No more than four levels will be approved
  • Injection may be repeated if the pain log shows >50% reduction in pain after the first injection
  • The maximum number of median branch blocks is two per year at the same level

Radiofrequency ablation (RFA) of the median branch nerve in the cervical or lumbar spine can be considered if there is 50% reduction in pain for at least four hours after two median branch blocks. RFA can be repeated at the same level once every 12 months.

Sacroiliac Injection:

The SI joints may be the cause of back pain in one-fourth of patients with low back pain. History and imaging findings are not reliable to establish the SI joint as the cause of back pain. The pain is more likely to be unilateral and non-radicular but may radiate into the buttock, thigh, or sacral area. The evidence for pain relief with intra-articular SI joint injection is limited for short and long term pain relief. InterQual criteria:

  • Failure of NSAIDS and activity modification for four weeks
  • Repeat injection may be considered for a 50% decrease in pain lasting at least eight weeks
  • There is a maximum of three SIJ injections in a 12 month period

Percutaneous Vertebroplasty:

Well controlled studies show no benefit over placebo for longstanding vertebral fractures with pain. The evidence is not as clear for acute fractures. Vertebroplasty may be approved on a case-by-case basis for:

  • Acute fracture (< three months) with debilitating pain after at least three weeks of conservative therapy
  • Vertebral hemangioma failing conservative therapy
  • Malignancy-related compression fracture without neurologic symptoms when pain interferes with activities of daily living.

Spinal Cord Stimulators:

Spinal cord stimulators (SCS) have been well studied. The evidence for SCS is strong for short term and moderate for long term pain relief for two spine situations: failed back surgery syndrome and complex regional pain syndrome. SCS is not covered by PHC for any other spine pain problem.

In all these spine interventions, PHC will require progress notes, imaging results when noted above, pain logs when appropriate, documentation of medications taken, and physical therapy dates prior to approving the intervention.

Evidence Summary:

Procedure Region Evidence
Short-term Long-term
Epidural Injection Lumbar disc with radiculopathy Strong Moderate
Cervical disc with radiculopathy Moderate Moderate
Post laminectomy radicular pain Limited
Axial back pain No benefit
Facet Joint injection Axial low back pain Moderate Moderate
Axial neck pain Limited Limited
Median branch block Cervical, thoracic, lumbar Moderate Moderate
Radiofrequency Ablation Cervical spine Strong Strong
Lumbar spine Strong Moderate
Sacroiliac injection SI joint Limited
Vertebroplasty Old compression fx (> 3 months) No benefit
Acute fx failing conservative tx Moderate
Spinal Cord Stimulator Failed back surgery syndrome Strong Moderate
Complex regional pain syndrome Strong Moderate

Jim Cotter, MD

Boswell, MV, et al. Interventional Techniques: Evidence-based Practice Guidelines in the Management of Chronic Spinal Pain. Pain Physician 2007;10:7-111