Randomised controlled trial to compare surgical stabilisation of the lumbar spine with an intensive rehabilitation programme for patients with chronic low back pain: the MRC spine stabilisation trial
BMJ 2005; 330 doi: https://doi.org/10.1136/bmj.38441.620417.8F (Published 26 May 2005) Cite this as: BMJ 2005;330:1233All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Re: Fairbank et al
Dear Sir,
Fairbank and colleagues have completed a noteworthy study. However,
several study design and statistical issues warrant additional comment
because they may bias the results against surgery. By virtue of the
inclusion and exclusion criteria, the patients most likely to benefit from
surgery were excluded. As such, the enrolled patients are not
representative of the population as a whole, which limits the inferences
that can be drawn from this trial. In addition, the investigation was
conducted at centers only in the United Kingdom (U.K.), which, again,
limits the generalizability of the findings to other countries, like the
United States (U.S.), where clinical practice patterns and treatment
regimens may differ from the U.K.
Surgical studies with bone morphogenetic proteins or with minimal
invasive surgery typically find ODI changes after intervention in excess
of 20 points.2 This is more than double that reported by Fairbank for the
surgical (change 12.5 pts) and rehab groups (change 8.7 pts). With more
current spine technologies, it is possible that different conclusions
might have been reached.
In light of the intent-to-treat analysis (ITT) employed (data
analyzed according to treatment assigned, not treatment undergone), the
differential crossover rate is particularly troubling: 28% of PT patients
underwent surgery, whereas only 4% of surgical patients had PT. With
this level of noncompliance, Sheiner contends that an average causal
effect analysis (ACE) is more relevant than an ITT, particularly with
respect to therapeutic decision-making.3 An ACE would evaluate the
effectiveness of surgery after “adjusting” for the observed level of
crossover. For example, Robins et al., using an ACE, demonstrated that
their ITT failed due to crossover rather than the lack of intrinsic
efficacy.4
The validity and policy relevance of the Fairbank study warrants
further consideration in light of these potential biases.
Fairbank J, Frost H, Wilson-MacDonald J, et al. Randomised
controlled trial to compare surgical stabilization of the lumber spine
with an intensive rehabilitation programme for patients with chronic low
back pain: the MRC spine stabilization trial. BMJ,
doi:10.1136/bmj.38441.620417.8F (published 23 May 2005).
2Burkus J.K., Heim S.E., Gornet M.F., and Zdeblick T.A. Is INFUSE bone
graft superior to autograft bone? An integrated analysis of clinical
trials using the LT-CAGE lumbar tapered fusion device. J. Spinal Disord.
Tech. 16(2): 113-22, 2003.
3Sheiner LB. Is the intent-to-treat analysis always (ever) enough? Br J
Clin Pharmacol 2002; 54:203-211.
4Robins JM, Finkelstein DM. Correcting for noncompliance and dependent
censoring in an AIDS clinical trial with inverse probability of censoring
weighted (IPCW) log-rank tests. Biometrics 2000; 56:779-788.
Competing interests:
Consultant Medtronic
Competing interests: No competing interests
We read with great interest the recent contribution by Fairbank et al
(May 2005) regarding the surgical versus rehabilitation interventions for
chronic low back pain. As a center that successfully utilizes both
operative and non-operative techniques in the management of our patients,
we hoped to draw conclusions from the paper, which were similar to our own
philosophy. Unfortunately, the study has a number of shortcomings that
may lead to misinterpretation of the collected data and consequently
conclusions, which are unsubstantiated.
First, the entry criteria for the study essentially excluded all
patients for whom surgery was the initial treatment recommendation. The
exclusion of this subset of patients, likely having the highest
pretreatment Oswestry scores, introduces an unacceptable bias. By
removing these patients, who would conceivably benefit most from surgery
and show the greatest benefit from each treatment dollar spent, the
authors effectively predetermine the outcome of the study. It is not
surprising that the post-operative Oswestry scores barely reached
statistical significance.
Second, by employing the intention to treat method of analysis, the
study allows and fails to account for one-way crossover of patients. For
example, the 48 patients (28%) in the non-operative group who crossed over
to the surgical group should be considered failures of rehabilitation.
Obviously, if these patients required surgery within the two years of
study follow-up, they would have incurred the cumulative cost of both
treatment regimens. This could hardly be considered an appropriate
control group, nor a cost-effective means of patient care.
Third, the authors admit to an extended period of time for patient
accrual into the study (June 1996 to February 2002). In a study that
requires a prolonged enrollment time (over 60 months) to enroll patients,
we are concerned by the 20% loss to follow-up rate at 24 months. How can
one possibly make decisions regarding cost-effectiveness and overall
treatment efficacy, if one cannot verify adherence to the patient’s
prescribed treatment regimen and follow up over the standard two years?
Similarly, we are concerned about the length of enrollment. Could the
techniques and costs associated with any intervention possibly be
standardized over such a period of time?
In conclusion, we applaud the significant effort put for by Fairbank
et al, but also caution against drawing conclusions unsubstantiated by
their data.
Competing interests:
None declared
Competing interests: No competing interests
Fairbank et al(1) reported a randomised control led trial of surgical
stabilisation of the lumbar spine compared to intensive rehabilitation
programme in patients with chronic low back pain(1). The principle of
randomisation was based on a “grey zone” of uncertainty (on behalf of the
treating clinician) as to whether surgical stabilisation or an intensive
rehabilitation programme was the appropriate management i.e. that there
was clinical equipoise as to whether the patient should have surgery or
non-surgical treatment.
Many spinal surgeons including myself would not feel that this is an
area where clinical equipoise is possible. If a patient is thought to be
appropriate for an intensive rehabilitation programme then that is what
should occur. Surgery is contraindicated until all conservative
treatments have been exhausted, including an intensive rehabilitation
programme. A number of spinal surgeons were unwilling to joint this trial
because of that.
If a spinal surgeon is apparently in clinical equipoise as to
surgical/non-surgical management, then it is likely that a patient has
either (i) not had a full course of medical management or (ii) has less
severe disease. Either of these would bias the trial in favour of the
intensive rehabilitation programme. Another noteworthy point is that
despite the likelihood that the trial is biased in favour of non-surgical
treatment 28% of patients randomised for rehabilitation eventually had a
surgical procedure.
Koes(2) correctly concludes that there is limited evidence that
confirms the benefit of surgical stabilisation and that appropriate
patient selection is difficult. However the primary position is that
until a patient has had a full course of non-surgical treatment surgical
stabilisation is not an appropriate option.
Mr N V Todd MD FRCS
Consultant Neurosurgeon & Spinal Surgeon
Newcastle General Hospitals NHS Trust
1. Katz. J. Lumbar spinal fusion: surgical rates, costs, and
complications. Spine 1995;20:78S-83S
2. Koes BW. Surgery versus intensive rehabilitation programmes for
chronic low back pain
Competing interests:
None declared
Competing interests: No competing interests
Editor- Fairbank and colleagues (1) suggest that there is little
difference in outcome for patients with chronic back pain whether given
intensive rehabilitation or spinal fusion. The key limitation to this
study however is that the best candidates for surgery were excluded.
In this study, patients in the surgical group had a mean Oswestry
Disability Score of 46.5 and in the intensive rehabilitation group of
44.8. In clinical practice there is a group of patients with higher
disability scores than this (2,3) which in itself makes attendance at a
daily intensive rehabilitation programme difficult and at worst
impossible. I (CH) can speak from personal experience having suffered from
chronic back pain for two and a half years. Despite being highly
motivated, well informed and otherwise fit and healthy my symptoms did not
settle after the acute episode. I was left with an Oswestry Disability
Score of 63. At this level of disability attending a regular
rehabilitation programme is practically impossible. For me, after months
of various analgesics, physiotherapy, rehabilitation and a self directed
CBT approach to pain management - none of which gave me a significant
improvement in quality of life - surgery was the only option. Certainly
the operative risks were far outweighed by the potential benefits.
We need to be aware that there is a group of patients for which
surgery is the answer. If we do not recognise this we run the risk of
condemning patients with potentially curative problems to years of pain
and suffering. Worse still, these patients can become pigeonholed as
having a large psychological component to their problem. As human beings
could any of us live with years of debilitating pain without any
psychological effect?
In summary, although intensive rehabilitation looks promising for
some patients perhaps we should be looking at better means of selecting
those who will benefit from rehabilitation and those for whom surgery
really is the only option.
1 Fairbank J, Frost H, Wilson-MacDonald J, Yu L, Barker K, Collins R.
Randomised controlled trial to compare surgical stabilisation of the
lumbar spine with an intensive rehab programme for patients with chronic
low back pain; the MRC spine stabilisation trial. BMJ 2005; 330:1233-1238.
2 Guyer RD, McAfee PC, Hchschuler SH, Blumenthal SL, Fedder IL,
Ohnmeiss DD, Cunningham BW. Prospective randomized study of the Charite
artificial disc: Data from two investigative centres. Spine J. 2004; 4(6
suppl):252s-259s.
3 McAfee PC, Fedder IL, Saidy S, Shucosky EM, Cunningham BW. SB
Charite Disc Replacement; report of 60 prospective randomised cases in a
US centre. Spine Disorders Tech. 2003; 16 (4) 424-33.
Competing interests:
None declared
Competing interests: No competing interests
Fairbank and co-workers (1) are to be congratulated for carrying out
a difficult investigation of a sizeable cohort that helps to clarify
issues faced by most practitioners. Given the paucity of well-controlled
studies on surgery for low back pain and the authority of the Spine
Stabilisation Trial Group, one would expect their conclusions to influence
professional standards of care and clinical practice. However, this could
be a mixed blessing, as several limitations of the study point up the need
for expanded information on a broader range of low back sufferers.
Subjects in this cohort were adults with >12 month history of
chronic low back pain who were eligible only “… if the clinician and
patient were uncertain which of the study treatment strategies was best.”
Given such vague criteria, one must assume that large numbers of low back
sufferers were not included, such as those with mild pain and, almost
surely, the preponderance of those with important or progressive
neurological abnormalities. It is these latter, high-risk patients for
whom most practitioners find surgical referral to be indicated, and their
exclusion from the study leaves us with no new information to assess
surgical (vs. non-surgical) choices (1).
In the accompanying editorial comment, Koes (2) properly suggests
that an untreated control group should have been included, particularly
since such untreated low-back sufferers might have improved as well.
Similarly, another group should have been treated with both surgery and
physiotherapy, since elements of each modality might be beneficial in
different ways and thus might be additive. Although such studies are
difficult to carry out, this type of supplementary information is
essential to the clinician’s ability to make informed judgments. We are,
at this time, not able to discard surgery as a potential therapeutic
option for specific patients.
References:
1. Fairbank J, Frost H, Wilson-MacDonald J, Yu L-M, Barker K, Collins
R for the Spine Stabilisation Trial Group. Randomised controlled trial to
compare surgical stabilisation of the lumbar spine with an intensive
rehabilitation programme for patients with chronic low back pain: the MRC
spine stabilisation trial. BMJ 2005; 330:1233-9.
2. Koes BW. Surgery versus intensive rehabilitation programmes for
chronic low back pain. BMJ 2005;330:1220-1.
Competing interests:
None declared
Competing interests: No competing interests
As a pain patient with severe LBP I have fared better in a pain
clinic where the appraoch was more holistic-medication (minimal),
exercise, manipulation, diet advice, trigger point injections
occassionally and cognitive behavioural therapy.
I was previously under a pain clinic inn London in a big teaching
hospital and I was just given medication,, lectured about diet and
exercise and treated as if I was thick and told " I'll be the one who
makes the decisions" (pain consultant) and also when p[ain got worse and I
didn't wish to increase my MST consultant asked "what is wrong with MST?".
I went in search of a more holistic approach and 15 months later, my
medication has been reduced by 65%, pain levels much so reduced and far
more importantly this was in an NHS pain clinic with a consultant who
genuinely believed that I could and would get better without all the
medical interventions the previous consultant kept throwing at me.
From a primary care perspective I was not impressed that medication
was the best that could be offerred initially and no more-the knowledge
was lacking, my bck was not examined properly until I hit the second pain
consultant and I have huge concerns about doctors basic diagnostic skills
in musculo skeletal pain.
All doctors should be able to do a proper back examination-there is
no excuse except sloppiness or lack of interest in the presenting patient.
Proper examination leads to proper diagnosis and better outcomes in
pain management.
Competing interests:
None declared
Competing interests: No competing interests
You raise some interesting points . I gather you are not an
'orthopaedic' bed salesman!
This link might be of interest to you (Instinctive and
anthropological
approach to back pain which you and your family seem to have
worked out for yourself)
http://bmj.bmjjournals.com/cgi/content/full/321/7276/1616
A pamphlet called the back book publised by HMSO is available
in most health centres and has advice by Professor Wadell
(Orthopaedic Surgeon and author of the Back Pain Revolution and
a Dr of Psychlogy Chris Main)
There is also a back pamphlet published by ARC (arhritis research
council) which is available on line or from most health centres
which has the kind of advice that you propose .
The overmedicalisation of back pain is a big issue that Gordon
Wadell has addressed over many years . If it were as simple as
graded restoration of function and home massage (which I think
are along the right lines) most chronic disability would be a thing of
the past .
For most people this approach is correct but many people simply
will not follow this schema and there is a huge 'care' industry to
turn to if coping strategies are stretched.
Another author mentioned dualistic interpretations of pain and
suffering . Any spell working in a pain clinic will expose you to
numerous cases where pain is 'seen' and interpreted using the
'evidence' of the scanner or xray .
Repeated back operations are still being performed when the
factors either perpetuating sensitvity or hugely influencing it are
not being addressed (fear avoidance and the issues of prior
adverse life events )
Positively, there is evidence that education utilising up to date
pain physiology and motor re-education has helped small
numbers of intractable cases -- perhaps this work may expand in
other clinics too .
Ian Stevens
Competing interests:
None declared
Competing interests: No competing interests
I am a veteran of two excruciating bouts with lower back problems,
and the husband and father of two others who have had similar experiences,
I was most interested by the fact that the intensive rehabilitation
programme outlined in Fairbank and colleagues' excellent article did not
seem to require any significant medical or other professional knowledge.
Any intelligent person with some awareness of one's own body and with the
ability to keep an optimistic mind, should be able to design and carry out
an "intensive rehabilitation programme" on one's own.
In our family experience, all three of us solved the problem by
walking, walking and more walking. (Bicycling was not so effective,
perhaps because it exercises much fewer parts of the body.) A little
amateur back massage was also helpful in one incident. And as preventive
medicine, my wife and I threw the mattress and bedsprings away more than
fifteen years ago. A thin camping mat, no more than a centimeter in
thickness, on top of a hard wooden bed, is all we use. After some getting
used to, it is perfectly comfortable. Since sleeping on such a bed, the
problem has not returned except for one very brief and quickly cured
incident. None of us is a medical professional.
Rather than investing in surgery or rehabilitation programmes,
perhaps the NHS, sick funds, etc, should consider putting out a pamphlet
on how to keep your back healthy,to be distributed for free to the public.
Competing interests:
None declared
Competing interests: No competing interests
Were patients above 35yrs subjected to assesment and treatment of
osteoporosis before assigning them to either group in the study
undertaken?If not,they should have been since it might have resolved some
of the chronic low back pains!
Best Wishes
Milind.
Competing interests:
None declared
Competing interests: No competing interests
SF36 scores
I have been using SF36 in my pain clinic for initial assessment and
to evaluate the effect of treatments which are multimodal and
multidisciplinary. I have used norm based scores from the demo screen to
score the results.
It is not clear from the article whether the results reported for the
domain scores PF,RP,BP,GH, VT, SF, RE and MH scores were direct scores or
norm based scores. Please clarify.
Will it be possible to obtain the Norm based scores of the above
samples?
Competing interests:
None declared
Competing interests: No competing interests