Low back pain: which is the best way forward?

Mr A, a fit and healthy 62 year old, presented with a 10 year history of intermittent non-specific low back pain. He had been able to control his symptoms with several self help measures, including weight control, regular aerobic activity, a firm mattress, back mobilisation exercises, and a naturally positive attitude. He had an active lifestyle and travelled extensively, but his episodes of acute pain had become increasingly severe and frequent and were starting to interfere with his quality of life. Radiological examination showed degenerative changes in the lumbar spine, predominantly at L4 and L5. A review of Mr A’s medical history and a clinical examination did not give rise to any suspicion that he might have a serious spinal abnormality or nerve root problems.

The clinical question

We considered the following clinical question. Which of the treatments available locally showed the best evidence of limiting acute attacks of low back pain in a patient with a history of chronic back pain? Our specific aim in Mr A’s case was to obtain reliable information as quickly as possible so that we could immediately advise him of the best treatment option.

The search

Our previous experience of reviewing published reports on chronic low back pain had shown that there are many types of treatment, both complementary and conventional. Since our experience is with conventional treatments, we confined our search to these. Four main treatments available locally were relevant to Mr A’s condition: physiotherapy and exercise programmes; spinal manipulation; spinal steroid injections; and transcutaneous electrical nerve stimulation. We decided to research the evidence for these treatments and to look for any other evidence based treatments that might be relevant.

Previous searching of Medline and Embase had shown a profusion of reports on treating low back pain. On closer examination, however, many of these lacked scientific rigour—there were questions over aspects such as randomisation, outcome measures, and duration of follow up. Furthermore, although searching these databases might provide reviews of much of the relevant evidence, the process is relatively time consuming.

Best Evidence¹ is a good starting point for quick identification of quality controlled evidence. Information can be accessed within minutes (or seconds, depending on your keyboard skills), so it is ideally suited for clinical use. Best Evidence is a compilation on CDRom of evidence based reports on articles from the ACP Journal Club and Evidence-Based Medicine. These publications give examples of current best evidence and provide expert commentaries.² They aim to select scientifically sound research studies that are relevant clinically and provide definitive evidence. Studies are selected according to predetermined methodological and clinical criteria to ensure their validity.³ Although searching is speedy, the coverage of journals is selective and information is updated only once a year.

Searching for information on Best Evidence was relatively quick and easy. A full text search using “chronic low back pain” in the selected topic group “therapeutics and prevention” provided nine titles, eight of which seemed to be of direct relevance. As Best Evidence provides a full report, all eight documents were quickly scanned, initially focusing on the authors’ conclusions and expert commentaries, and seven were examined in greater detail.

We also accessed the Cochrane Library to check whether other evidenced based reports were available. The Cochrane Library is a quarterly updated CD Rom compiled by the Cochrane Collaboration, an international organisation that aims to provide quality appraised summaries and systematic reviews of current best practice. If your particular area of interest is covered, the Cochrane Library represents a valuable and time saving resource. It comprises four sections (box).

Components of the Cochrane Library

• Cochrane Database of Systematic Reviews—assimilates the findings of scientifically rigorous and broad based trials that use appropriate outcome measures for clinical decision making.

• Database of Abstracts of Reviews of Effectiveness—is a collection of systematic reviews assembled by the York Centre for Reviews and Dissemination

• Cochrane Collaboration Trials Registry—provides a more complete list of randomised controlled trials than any of the bibliographic databases

• Cochrane Review Methodology Database—lists protocols of reviews currently in progress and includes contact points for research coordinators, a summary of their study design, and an expected date of publication

Because the Cochrane Library is currently small, a comparatively simple search strategy is sufficient. Using the advanced search option, the medical subject headings (MeSH) terms and text words “back-pain” and “low-back pain” were entered. These terms were then combined with “therapy.” This strategy provided 17 hits under “back-pain” (text word), 143 hits under “low-back pain” (MeSH), and 440 hits under “back-pain” (MeSH). Although these numbers seemed overwhelming, on closer scrutiny it was evident that a hit was recorded every time the relevant word was found, even in references, citations, and bibliographies. Once the search results were displayed, only one complete review was pertinent to the question, together with 12 potentially interesting abstracts in the Database of Abstracts of Reviews of Effectivenenss. One summary provided an evaluation of therapeutic evaluations for low back pain.⁴ In addition, there were 130 hits in the Cochrane Collaboration Trials Registry. These covered injection therapy (anaesthetics, steroids, or other drugs), manipulation, osteopathy, and chiropractic. It took less than five minutes to perform the search (and become familiar with the software).

Because Best Evidence had been updated eight months previously, we decided to search the websites of ACP Journal Club (incorporating Evidence-Based Medicine; www.acponline.org.) and Bandolier (www.jr2.ox.ac.uk/bandolier) for recent updates. Bandolier did not provide any information that was directly relevant, but it did give a lively and amusing commentary with good advice from the Clinical Standards Advisory Group’s guidelines.⁵ The ACP Journal Club search did not yield additional relevant studies.

The evidence

One document in the Cochrane Library described several new treatments such as laser therapy and injections of collagen proliferant substances.⁴ Though these seem promising, little definitive evidence on their use is currently available. Furthermore, it seemed that attendance at a “back school” alone had little useful effect on pain relief or disability.

Transcutaneous electrical nerve stimulation

The Cochrane Database of Systematic Reviews included a review showing that transcutaneous electrical nerve stimulation was better than placebo for relieving pain and restoring function in patients with low back pain that had lasted more than eight weeks.⁶,⁷ However, the beneficial effects were not appreciable, and not all patients showed a benefit.

Exercise programmes

A large study comparing the effects of McKenzie-type exercises (repetitive movements of flexion, extension, and side gliding) with education without exercises at a back school showed that exercises were more effective than placebo even after five years.⁸,⁹ Of greater interest was a recent study evaluating the effectiveness of a graded exercise programme.¹⁰,¹¹ This showed that education combined with physical reconditioning and exercise could help prevent recurrences and improve overall levels of functioning.

Manipulation therapy

A meta-analysis of primary studies of spinal manipulation showed an increased likelihood of recovery after 2-3 weeks of treatment for patients with acute and subacute low back pain.¹²,¹³ However, the benefits were lost within a few weeks, consistent with the course of the disease. Unfortunately, there was insufficient evidence regarding the efficacy of manipulation in helping patients with chronic low back pain.

Spinal steroid injections

One large controlled study showed that patients given an injection into the facet joint had slightly less pain and better functional ability after six months than patients who were given a placebo.¹⁴,¹⁵ Although no patient reported any adverse effects, few patients reported any sustained improvement.

Applying the evidence

The next step was to determine whether these findings could be applied to Mr A, since recommendations from population based studies often have to be adapted to meet the circumstances of individual patients. However, for patients suffering from chronic and painful conditions any intervention that offers any possibility of success is likely to be considered, even when long term benefits have not been proved. For this reason all the possible options and expected outcomes were considered.

From discussion with Mr A, it emerged that previous use of a transcutaneous electrical nerve stimulation machine had not been beneficial. Regular chiropractic and manipulation therapy was not an option because Mr A could not afford this. Although steroid injections had previously brought good relief from pain, these do not offer any long term solution. After full consideration, Mr A chose a more active rehabilitation programme using an incremental aerobic exercise and fitness programme. He was also advised that recurrences were likely but could be treated in the same manner.

Conclusion

The evidence showed what we had suspected, that general awareness and education about back care combined with a regular exercise programme is probably the best option at present. One useful discovery we made was the Cochrane Review Methodology Database, which highlighted all reviews currently in progress. This is a convenient way of tracking the latest updates in any field of interest, and is certainly something we will use in the future.