Study Questions Additional Benefit of Physical Therapy for Low-back Pain
By Simon French BAppSc(Chiro), MPH.

A recent Australian study published in the prestigious journal The Lancet has showed there is no benefit from additional physical therapy after care from a general practitioner (GP) for people with acute low-back pain (LBP) (Hancock 2007). The study addresses a question a GP would ask: “if a patient consults me with acute non-specific low-back pain, and I provide good quality care, do I then need to recommend non-steroidal anti-inflammatories (NSAIDs) or refer them for physical therapy?”.

This important study is a challenging one for manual therapists who treat people with acute non-specific LBP. It was a very well designed and conducted randomised controlled trial (RCT). Participants (240 in total) with acute low-back pain who had seen their GP and had been given advice and paracetamol were randomly allocated to one of four groups: 1) non-steroidal anti-inflammatories (NSAIDS) in the form of diclofenac (50 mg twice daily) and placebo physiotherapy; 2) physiotherapy and placebo drug; 3) NSAIDs and physiotherapy (n=60); or 4) double placebo. The investigators concluded that people with acute non-specific LBP who attend a GP providing guideline recommended treatment get no additional benefit from attending a physiotherapist providing manual therapy nor from taking NSAIDs.

So what generalisations can be drawn from this one, relatively small study, to the whole of the acute non-specific LBP population?

This study only included people with pain for less than six weeks, and the mean duration of pain for these people was nine days. For people with pain duration longer than this, the study may not be relevant.

All patients who presented to the GPs in the study were referred to a study physiotherapist for further treatment. Perhaps there are only some patients who need manual therapy in the early stages of acute non-specific LBP. This study adds in some way to the growing body of evidence that lumping all patients with acute non-specific LBP into one diagnostic category and giving them all one particular therapy is not necessarily appropriate. However, there are currently no reliable methods to determine which patients will, and which will not, respond to different types of treatment for acute non-specific LBP (Billis et al 2007). The future challenge for researchers and clinicians is to show which of these patients will and will not respond favourably to certain types of physical therapy.

The physiotherapy treatment provided in the study was at the discretion of the treating practitioner, guided by a treatment algorithm. Only 5% of the participants received high velocity spinal manipulative therapy (SMT), the remainder received low velocity mobilisation. Although there is no clear evidence that high velocity SMT is more effective than low velocity SMT (Assendelft 2004), this study did not test SMT effectiveness per se. It would have been interesting to see how the participants had responded if they had been referred to a chiropractor or osteopath who provided high velocity SMT.

It is important to recognise that this study does not conclude that GP care is more effective than manual therapy, as has been reported in some media. This study did not examine the GP treatment recommended in most evidence-based guidelines because all participants received this (advice and paracetamol) and there was no comparator group to this treatment. The comparisons made were between NSAIDs, physiotherapy and placebo. Surprisingly, to date the combination of GP advice and recommending analgesics has not actually been tested in a RCT (Liddle 2007). It would have been interesting in this study to have seen a “GP only” group who were not referred to a physiotherapist and see how they responded. Perhaps there is something gained from being referred to a physiotherapist, even if the treatment itself is a placebo.

The generalisability of this study is also worth questioning, ie what happens in the real world of care for people with acute low-back pain? If someone with acute non-specific LBP presents to a GP in Australia, will they receive evidence-based care? The evidence suggests otherwise and best practice is not guaranteed (Buchbinder R & Jolley D 2007). Also, all patients in this study recovered by 12 weeks. This seems a very good result for this cohort of patients. A longer follow-up may have indicated which of the different treatment groups had recurrences and which did not. A recent systematic review of all prognostic studies did not show a result as good as this for acute non-specific LBP patients (Pengel et al 2003). Hence one needs to be cautious to relate what has occurred in this current study to the “real world” of LBP treatment.

If today’s manual therapists are to implement evidence into daily practice, the main massage from this well conducted study is that a GP providing evidence-based care can provide all that is needed for this group of patients. Discerning chiropractors and osteopaths would be wise to be aware of this study’s strengths and limitations and consider it when providing care to people who present with acute low-back pain.

References:
Billis EV, McCarthy CJ, Oldham JA (2007). Subclassification of low back pain: a cross-country comparison. Eur Spine J 16(7): 865-79.
Buchbinder R, Jolley D (2007). Improvements in general practitioner beliefs and stated management of back pain persist 4.5 years after the cessation of a public health media campaign. Spine 32(5): E156-62.
Hancock MJ, Maher CG, Latimer J, McLachlan AJ, Cooper CW, Day RO, Spindler MF, McAuley JH (2007). Assessment of diclofenac or spinal manipulative therapy, or both, in addition to recommended first-line treatment for acute low back pain: a randomised controlled trial. Lancet 370(9599): 1638-43.
Liddle S, Gracey J, Baxter G (2007). Advice for the management of low back pain: A systematic review of randomised controlled trials. Manual Therapy 12(4): 310-27.
Pengel LH, Herbert RD, Maher CG, Refshauge KM (2003). Acute low back pain: systematic review of its prognosis. BMJ 327(7410): 323-5.

(Editors note: The abstract for this paper is the lead item in the “Abstract” section)