3.7 Management of nerve root pain (syn: sciatica, radicularpain)

True nerve root pain has a lifetime prevalence of less than 5%. Lumbar-sacral nerve root pain radiates in the distribution of one or more dermatomes and may be associated with neurological deficits. Although back pain may coexist, the severity of the leg pain predominates. The most common cause is disc prolapse (also called herniation). The majority are at L4/L5 or L5/S1. However, other pathologies which have the potential to compress or irritate a nerve root include tumour, infection, or neuritis. It has become clear in the last few years that disc degeneration and the proportion of degenerate discs that is herniated is driven by genetic factors. Genetics explains 60% of the variance, with minimal contributions from environmental factors such as smoking and trauma. This means that family history should always be explored. There is evidence that mechanical compression of a nerve root results in sensory and/or motor deficits. The onset, duration, and extent of the pressure have been shown to be important. However, compression alone does not cause pain. Some degree of inflammation/irritation must exist for nerve to refer pain. The pathophysiology of nerve root pain is not fully understood. Research has explored the chemical effect of the nucleus pulposus coming into contact with a nerve root, in particular the dorsal root ganglion. A large number of inflammatory and signalling substances, including cytokines (especially tumour necrosis factor, TNF), nitric oxide, and nerve growth factor have been suggested to play a role. There may be an immunological element as well, as the nucleus is normally avascular. It is likely that mechanical and biochemical factors act together, altering intraneural permeability, intraneural blood flow, and nutrition, resulting in degeneration of nerve fibres and reduced nerve conduction.

Root pain caused by disc prolapse is usually preceded by back pain. The interval varies from a few minutes to years. Leg pain usually predominates in a dermatomal distribution. Pain is exacerbated by coughing and sneezing, walking, and standing. Night pain is common. Pain may be relieved by lying and flexing the spine. Root pain (sciatica) can be extremely severe, and associated with numbness and paraesthesia. Weakness may be experienced with poor push-off, flapping gait, and foot drop. The appearance of foot drop is sometimes associated with pain improvement. Perineal numbness, and associated bladder or bowel symptoms are important indicators of cauda equina syndrome but the latter may be associated with pain severity and analgesic intake.

Examination should include an assessment of sensation to both light touch and pinprick (the end of a paper-clip is effective). Straight-leg raising is restricted. I prefer to flex the hip and knee first, and then gradually and gently extent the knee. This allows assessment of hip pain, and is unfamiliar to experienced patients with illness behaviour. Weakness should be sought particularly in extension of the great toe and foot, foot eversion, and knee extension. Ankle and knee reflexes must be recorded.

Imaging is best done by magnetic resonance (MR). If this is unavailable or contraindicated, computed tomography or myelography may be used. Disc herniations are seen in up to 25% of normal scans, so it is essential to interpret the scan in light of the clinical picture. Large discs tend to respond better to treatment than small ones. Attention should be paid to the vertebral canal diameter. Small canals give more trouble than large ones. Look out for other contributing factors to nerve pain such as lateral recess and exit foramenal stenosis (Table 3.7.1).

Nerve root pain caused by disc herniation usually gets better. Unfortunately the time scale is unpredictable. Clinical practice, observational data, and now some useful randomized controlled trials have given reasonably clear guidance on management. No non-operative treatment comes out better than natural history. However, explanation and analgesia, possibly combined with muscle relaxants, are important. Epidurals (either spinal or caudal) do not alter natural history, though the short-term relief of root pain may be convenient for some. Surgery may be indicated from 8 weeks from onset, unless there is evidence of cauda equina involvement or progressive neurological deficit, when surgery is indicated straight away. This is informed by a number of trials, including the SPORT study, and, most recently Peul and colleagues, who compared an early with delayed surgical strategy. All of these favour early surgery over non-operative care in the short/medium term. The patient has to weigh up the risks (persisting pain, nerve damage, cauda equina damage (rare but devastating), dural tears, and recurrence) against (90%) benefit of relief of nerve root pain but not back pain. Earlier studies give a recurrence rate of 10% in 10 years, but Peul and colleagues found 20% at 2 years. There is evidence from one trial that traditional disc space curettage produces inferior results to simple fragment removal. It should be emphasized that these trials are done on patients with clear-cut indications. In clinical practice this not always the case, and there are many studies showing poor outcome when the McColloch criteria are not met (see Box 3.7.2). He wrote in the first edition of this book the following:

The prerequisite for any surgical procedure is a precise diagnosis in a patient who is positively motivated to recover from surgery. Not only must the clinical syndrome be obvious (Box 3.7.2), but there must also be a clearly defined, unequivocal lesion on imaging that corresponds, anatomically, with the clinical root level. Where MR imaging is readily available, it has become the imaging modality of first choice.

The young patient with a disc herniation is a special problem. Because of the high proteoglycan content of the discal material (responsive to chymopapain) and the prevalence of disc protrusions rather than disc extrusions in this age group, I recommend that the optimum treatment is chemonucleolysis rather than surgical intervention. Unfortunately chyomopapain is not currently available, although there are hopes that it may return to the market. Surgery can be effective but the recurrence rate is significantly higher than with chymopapain.

The patient who has successful relief of sciatica after disc excision has a 5–20% chance of a recurrent disc rupture causing sciatica. Unfortunately, most recurrent disc herniations are at the same level and same side, and scar tissue from the previous surgery introduces a new element in diagnosis and treatment. Scarring ‘tacks’ down the dura to the back of the disc space so that a smaller amount of herniated nuclear material is capable of producing a significant amount of pain and neurological deficit in the relatively immobile nerve root. Because of the immobility of the scarred dura, transdural ruptures, although rare, can occur. Almost invariably the scar between the dura and the disc restricts migration of the recurrent disc fragments, i.e. on exploration they are routinely found opposite the disc space.

Determining the anatomic level by clinical assessment can be difficult because some neurological changes are residual from the prior lumbar disc herniation, and the dura may not simply be immobilized, it may be distorted from the scar tissue, leading to root involvement lower than usual for the level of the lumbar disc herniation (e.g. a recurrent L4–L5 lumbar disc herniation may affect a number of sacral roots).

Investigation can be difficult to interpret because of the scar tissue. It may be helpful to enhance the MR examination with gadolinium. The surgery is also difficult and prone to complications such as missed lesions, dural tears, and neurological damage. A basic principle in any repeat surgery is to gain as wide an exposure as possible.

Surgical discectomy and/or decompression to relieve sciatica is an operation with clear-cut indications and is usually followed by a high level of success. Microsurgical techniques facilitate the exercises but will not improve an overall success rate of 80–90%. In the end, the decision to operate on a patient with a lumbar disc herniation and/or lateral zone stenosis is largely dependent on patient preference and rarely on necessity. In the short term, however, surgery will result in less pain and improved function compared with conservative care.