Research Review By Dr. Rob Rodine©

Date Posted:

Mar. 2009

Study Title:

Imaging strategies for low-back pain: systematic review and meta-analysis


Chou R, Fu R, Carrino JA, Deyo RA

Author's Affiliations:

Oregon Health and Science University, Portland OR

Publication Information:

Lancet 2009, 373: 463-472.

Background Information:

In 1994 the Agency for Healthcare Policy and Research recommended against the use of radiography during the first month of an acute low back pain (LBP) episode. This recommendation was made in an attempt to curb the rising number of clinically unhelpful radiographs that were being taken in the low back pain population, who generally did not exhibit red flags in their histories.

Often, clinicians will order radiography in an attempt to set patients at ease and meet expectations of a thorough clinical intake. Other times, imaging is ordered for the purposes of financial incentive or in an attempt to make an anatomical diagnosis. However, the astute clinician must be mindful of the inherent limitations in using radiography as a routine procedure for low back patients.

With any intervention or diagnostic procedure there is also the possibility to cause harm. For radiography, this harm is represented in the exposure to ionizing radiation. Therefore, the potential harm must be worth the information gained. This worth can be quantified in the final outcome of the patient’s case, which is the purpose of this paper.

In this study, a systematic review and meta-analysis was conducted investigating the relationship between the use of immediate and routine imaging and clinical outcome when compared to cases without immediate and routine imaging, in patients with acute low back pain.

Pertinent Results:

A total of 479 citations were identified and 13 were included for detail and quality assessment. Of these 6 trials were included. Within these 6 trials, data was available on a total of 1804 patients.

Three studies examined immediate lumbar radiography versus usual care without radiography. One study examined immediate lumbar radiography versus a brief educational intervention, and the future use of radiography if there was no improvement seen at 3 weeks. One trial examined immediate MRI or CT versus usual care without advanced imaging. In this study, patients primarily presented with chronic low back pain (82%) and were referred to a surgeon.

The final study examined patients with acute low back pain that underwent MRI examination and were then randomized to either receive their imaging results within 48 hours, or only if clinically applicable. While one trial excluded cases of sciatica and another did not report on this description, the remaining trials demonstrated a proportion of sciatica ranging from 24-44%.

Five of the six included trials were deemed to be of high quality (4+/8), whereas the remaining trial was deemed to be of lesser quality (2/8). Five of the six trials did not describe the blinded outcome assessment and four of six trials did not describe the randomization procedures adequately.

Outcomes Assessment:
  • when the outcome measures of pain and function were used, no differences were found within short and long term follow-up times for those patients that received immediate lumbar radiography and those that received usual clinical care (it should be noted however that a slightly positive trend was noted for those patients not receiving immediate radiography, though non-statistically significant)
  • using the measure of weighted mean differences, those patients receiving usual clinical care demonstrated slightly favorable short term follow-up pain scores (using the VAS)
  • no differences were noted when the Short Form 36 bodily pain questionnaire was used
  • long term follow-up pain scores demonstrated no differences with the VAS or the SF-36
Function was measured in trials using the Roland Morris Disability Questionnaire, the SF-36 and the Aberdeen low back pain score.
  • only one of the six included trials reported pain and functional measurements at extended (defined as 2 years) follow-up – it was demonstrated that neither the immediate use of MRI or CT imaging benefited patients during this timeline when compared to usual clinical care
  • when the measures of quality of life, mental health and overall patient reported improvement were used, immediate imaging versus usual clinical care groups did not differ
Of important note, 4 trials obtained either radiography in all patients, or recorded clinical diagnosis through a minimum of 6 months follow-up and found no cases of serious disease or missed pathology as a result of usual clinical care.

Clinical Application & Conclusions:

The authors of this study conclude that in patients not presenting with red flag characteristics, and therefore unlikely to be at risk for serious underlying disease, did not demonstrate improved clinical outcomes with the addition of immediate imaging to usual clinical care practices. The authors also cite several studies to support that lumbar imaging is associated with potentially harmful ionizing radiation exposure, increases costs and does not alter diagnosis or treatment plan decision making, in the absence of serious underlying disease.

As there is deemed to be little benefit with routine imaging for low back pain, a risk of associated harm, increased costs and no alternation in management strategies, routine imaging should be avoided. The authors do stress however that these results are most applicable to patients within a primary contact health care setting that are presenting with acute or sub-acute low back pain. It is recommended that further research be conducted in order to determine imaging strategies for chronic low back pain.

Despite limitations this study helps to direct clinicians away from routine lumbar imaging in the absence of serious disease indicators. Routinely imaging LBP patients without red flags does not lead to improved clinical outcome, and the harm outweighs the benefit. Therefore, the astute clinician will always ask themselves “how does my assessment strategy alter my management strategy?”

Study Methods:

A thorough literature search was conducted attempting to identify all literature from 1966 to 2008 using the Medline database and the Cochrane database. Reference lists were also reviewed for pertinent citations. All languages were included.

Only randomized controlled trials were included. Trials had to examine the clinical outcome in patients exposed to routine and immediate imaging versus care without immediate imaging. This comparison was only made in cases that were absent of serious pathology indications (red flags).

Outcomes of interest included: pain, function, mental health, quality of life, patient satisfaction and overall patient reported improvement.

Citations were screened for inclusion by two independent reviewers who then reached a consensus on disagreements. Two independent reviewers completed quality assessment and data abstraction as well. Quality assessment was conducted based on the modified criteria used by the Cochrane Back Review Group. Studies were assessed as high quality if they met 4/8 criteria. Symptom duration was categorized as acute (<4 weeks), subacute (4-12 weeks) and chronic (>12 weeks). Outcomes were categorized as short term (<3 months), long term (6 months – 1 year), and extended (>1 year).

Study Strengths / Weaknesses:

These conclusions are not without limitation. There were a limited number of studies included, significantly decreasing the power of the results. In addition, studies demonstrated heterogeneity which minimized the authors’ ability to appropriately pool data and generate stronger conclusions.

Studies were also diverse clinically, not just methodologically. In addition, included patients presented with various complaint durations, modality type and assessment strategy and utilized various clinical outcome measures.

Additional References:

  1. Di Iorio et al. A survey of primary care physician practice patterns and adherence to acute low back problem guidelines. Arch Fam Med 2000; 9: 1015-21.
  2. Ammendolia C et al. Adherence to radiography guidelines for low back pain: a survey of chiropractic schools worldwide. JMPT 2008; 31: 412-418.
  3. Ammendolia C et al. Do chiropractors adhere to guidelines for back radiographs? A study of chiropractic teaching clinics in Canada. Spine 2007; 32: 2509-2514.