Can surgery end low back pain?
Professor Charles Greenough assesses the latest evidence
on diagnostic and surgical techniques for simple back pain
Mechanical back pain, discogenic back pain or simple back pain are essentially interchangeable terms. They refer to low back pain which is generally aggravated by activities and relieved by rest. Two principal patterns may be recognised.
'Toothache of the back' An aching pain in the lumbar region that commonly radiates to the buttocks and posterior thighs and can less commonly radiate into the leg and even foot. The old adage that pain felt below the knee must be due to nerve root compression is wholly inaccurate.
These patients are stiff in the morning, tend to experience more pain as the day progresses, get worse with activity, can't sit or stand still and wake at night when they turn over.
Giving way Acute onset of severe pain and spasm often precipitated by trivial movements. These attacks last for a day or two and gradually settle. In between, sufferers can be reasonably symptom free or can exhibit symptoms in the pattern of toothache of the back.
On clinical examination patients display few features of value in diagnosis or prognosis. Lumbar spasm is variable from day to day. The range of spinal movement is difficult to reproduce without sophisticated technology and is significantly influenced by psychological distress.
Of greater value are the 'red flags' listed in table 1. A number of studies has attempted to clarify the sensitivity and specificity of these symptoms and signs both in metastatic disease and spinal infection (table 2). With a combination of some factors a high sensitivity in metastatic disease can be achieved.
Adverse prognostic factors for recovery from acute low back pain have also been identified, the so-called 'yellow flags' (table 3). These adverse factors lend further support to the concept of early active management of these patients.
X-rays are non-contributory and are not indicated. Recent research has suggested that in patients with red flags a limited protocol MRI is to be preferred.
But MRI has the disadvantage of demonstrating a wide range of degenerative features that have very little diagnostic value: 25 per cent of men in the third decade demonstrate dehydration of the L5/S1 disc, rising to 50 per cent in the fourth and fifth decades. Asymptomatic disc prolapse is seen in up to 30 per cent of normal volunteers.
The high-intensity zone, a bright lesion within the annulus on the T2 weighted image, was initially thought to be correlated with mechanical back pain, but more recent studies have failed to confirm this. The association between high-intensity zone and the reproduction of pain on discography has not been consistent.
Unfortunately, many normal appearances associated with ageing are reported as abnormalities, causing worry and distress to patients. As there are no MRI changes demonstrably associated with simple mechanical back pain, MRI is of no value and may even be harmful.
This is a technique where a radio-opaque medium is injected into the intervertebral disc under pressure.
A positive discogram demonstrates the presence of an abnormal morphology together with an exact or similar reproduction of the patient's usual pain.
Correlation between discography and MRI findings is by no means exact, while complications include infective discitis. There is a little evidence that discography is predictive.
However, the history of patients with positive discograms appears to suggest improvement in the majority.
More recently, attention has turned to the lumbar musculature and its possible role in low back pain.
Studies recording electromyogram signals from surface electrodes have demonstrated differences between asymptomatic volunteers and patients complaining of chronic low back pain.
The technique can discriminate between chronic pain patients and asymptomatic volunteers with specificity of 75 per cent and a sensitivity of 65 per cent.
EMG measurements are also found to predict future pain even in subjects with no past history of back pain.
These and similar studies indicate that the lumbo-spinal musculature have a significant role to play in the prevention of low back pain and lend some physiologically based support to the modern active management of low back pain.
The prognosis of low back pain and the results of many interventions are heavily dependent on patient-related factors. Psychological distress is particularly important in this regard.
It may be that such distress renders a patient less able to cope with symptoms of back pain and present to their health care practitioner a clinical picture based partly on failure of coping strategies.
Health care practitioners are often poor in diagnosing psychological distress on clinical grounds, particularly in cases of depression, and so a formal evaluation of distress is of great value.
The presence of a compensation claim is statistically associated with a reduced outcome. Previous failed surgical attempts are also associated with a reduced chance of success.
Measurement of outcome
Four areas of outcome are often reported in the literature:
lDisability measured by various instruments
lAspects of participation in society, such as employment
lQuality of life.
However, there is often no good agreement between the results obtained by these various measures. Patient satisfaction is obviously of primary importance to the patients themselves, but the factors on which the patient is making such judgment are often unclear.
Those paying for health care will be more interested in cost-effectiveness, representing the cost of a measurable improvement in ability.
Purchasers and society at large will also be interested in cost-benefit, which is calculated using quality-of-life measures, return to employment or other similar variables.
The conservative management of low back pain has been thoroughly examined. Emphasis is placed on active management with reassurance, maintenance of activities and an early return to work.
Many national guidelines have been produced worldwide and these are remarkable for their consistency. In the UK, the Clinical Standards Advisory Group has produced an excellent synthesis of management1.
In addition to conservative
measures, a number of different types of surgical intervention has also been used in the management of mechanical back pain.
A number of techniques have been used to try to alter the biomechanics or innervation of the intervertebral disc tissues, of which intradiscal electrothermal therapy (IDET) is probably most well-known. In this procedure a catheter is placed under image intensifier control within the intervertebral disc immediately adjacent to the posterior annulus. Using radio-frequency current this catheter is then heated to a specified temperature for a set period of time.
Promoters of the technique report significant reduction in pain and disability. The mechanism of action has been postulated to be a mixture of denervation, increasing stiffness of annular tissues and healing of posterior high-intensity zones. However, a number of basic science studies has cast doubt on the postulated mechanisms.
For example, temperatures in-vitro more than a few millimetres from the catheter itself have been found to be inadequate to induce necrosis of nerve endings. Review by the Cochrane Collaboration concluded: 'There was limited evidence suggesting that intra-discal radio-frequency thermo- coagulation was not effective for discogenic low back pain.'
Disc nucleus replacement
A number of devices has been described for implanting prosthetic material into the nuclear cavity of the intervertebral disc. A wide variety of materials has been used experimentally and a smaller number of these has been used more widely.
In 2003 the results of a prosthetic disc nucleus device were reported in 48 patients with one-year follow-up. In this series 78 per cent of patients achieved a good or excellent result with reduction of the Oswestry Disability Index from 59 pre-operatively to 18 at follow-up. Four re-operations were required in this group for migration of the prosthesis.
But at the present time disc nucleus replacement remains an experimental technique that has yet to be subjected to formal assessment in a randomised controlled trial. Until such trials are available the procedure should probably be restricted to formal studies.
A number of techniques has been developed to restrict movement at an intervertebral motion segment without attempting to abolish it completely. Two of the more widely known examples are the Graf ligament and the Dynesis device. Both of these rely on screws passed down the pedicles of the vertebrae into the vertebral body.
In the Graf system, polyester bands are then attached between the screws under tension. The Dynesis system connects the screws with a composite structure comprising both a tension band and a slightly compressible material.
Overall, for flexible stabilisation there is modest (one randomised controlled trial) evidence of effectiveness. But complications are not negligible and questions remain over the longevity of these devices, which rely entirely on satisfactory long-term integrity of the screw-bone interface and of the implanted materials. Significant revision rates have been observed between four and seven years.
It has been postulated that the flexible stabilisation systems may provide a period of symptom relief following which fusion may be undertaken if required.
But it has not been demonstrated that fusion in patients with failed flexible stabilisation carries the same chance of a successful result as fusion in a previously unoperated spine.
Certainly the literature suggests fusion undertaken in patients with failed previous fusion surgery carries significantly worse results.
Over the last two decades increasing interest has been expressed in total prosthetic disc replacement where the intervertebral disc is excised with the retention of only the lateral and posterior outer annulus. A total disc replacement prosthesis is then implanted in the disc space.
A number of such implants have been designed. Potential advantages of total disc replacement are the absence of bone graft donor site, retention of motion and possible reduction of degenerative changes at adjacent levels and immediate mobilisation post-operatively.
Potential disadvantages have been concerns over end plate erosion, displacement of the disc prosthesis, the effect of any wear debris produced and the potentially difficult revision in cases of failure.
Total disc replacement requires an anterior approach that can be undertaken only by fully trained and experienced spinal surgeons. In addition to the risk of laceration to the major vessels, failure of erection and retrograde ejaculation have been reported.
The technique is still to be proven by clinical trials but appears to be at least as effective as fusion surgery.
Arthrodesis of the spine has been undertaken for more than a century and internal fixation of the spine was first introduced in 1891.
Since then many hundreds of publications have considered the results of lumbar spinal fusion in mechanical back pain. Only recently, however, has any scientific evidence been published in the form of randomised controlled trials.
In summary, there is firm evidence for the effectiveness of lumbar spine fusion for mechanical low back pain. But the effectiveness is relatively small, approximately 11 or 12 points on the Oswestry Disability Index, and no differences have been observed between fusion surgery and well-structured functional rehabilitation programmes.
The technique of fusion surgery appears not to be contributory, although the possibility that anterior surgery alone may yield better results has not been tested.
Back pain is a symptom. A definitive diagnosis is not usually made, the pathology is poorly understood, imaging is not diagnostic and the prognostic factors are primarily psycho-socio-economic. Surgery is indicated only in a very few patients.
The key to successful results is patient selection, which must involve a holistic assessment including psychological distress. It is vital that both patient and health care professional have a realistic view of the results that may be achieved.
Charles Greenough is consultant in orthopaedics at Middlesbrough General Hospital
lMRI is of little diagnostic value for mechanical back pain and may even be harmful
lEMG is a useful investigative tool and indicates importance of lumbo-spinal musculature
lThere is little evidence that intradiscal electrothermal therapy (IDET) is effective
lDisc nucleus replacement remains an experimental technique untested by formal assessment
lThe effective of fusion surgery is proven but small
3. Yellow flags for low back pain
lBedrest >2 days
lPassive physical therapies
lDelay in return to work
1.Clinical Standards Advisory Group. Low Back Pain. HMSO. London, 1994