Best Practice & Research Clinical Rheumatology 24 (2010) 241252

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Rheumatology
journal homepage: www.elsevierhealth.com/berh

Sciatica
Jean-Pierre Valat, MD, PhD a, *, Stephane Genevay, MD b, Marc Marty, MD c,
Sylvie Rozenberg, MD d, Bart Koes, PhD e
a

Rhumatologie, Universite Franois Rabelais, faculte de medecine de Tours, Tours cedex, 10 boulevard Tonnelle, BP 3223,
37032 Tours cedex, France
b
Division of Rheumatology, University Hospitals of Geneva, Switzerland
c
pital Henri Mondor, Universite Paris 12, AssistancePubliqueHo
pitaux de Paris, Creteil, France
Rhumatologie, Ho
d
Rhumatologie, CHU Pitie-Salpetrie`re, Paris, France
e
Department of General Practice, Erasmus University Medical Center, Rotterdam, the Netherlands

Sciatica is a symptom rather than a specic diagnosis. Available

evidence from basic science and clinical research indicates that
both inammation and compression are important in order for the
nerve root to be symptomatic. Tumour necrosis factor-alpha (TNFa) is a key mediator in animal models, but its exact contribution in
human radiculopathy is still a matter of debate. Sciatica is mainly
diagnosed by history taking and physical examination. In general,
the clinical course of acute sciatica is considered to be favourable.
In the rst 68 weeks, there is consensus that treatment of sciatica
should be conservative. We review and comment on the levels of
evidence of the efcacy of patient information, advice to stay
active, physical therapy analgesics, non-steroidal anti-inammatory drugs (NSAIDs), epidural corticosteroid injections and transforaminal peri-radicular injections of corticosteroid. There is good
evidence that discectomy is effective in the short term. but, in the
long term, it is not more effective than prolonged conservative
care. Shared decision making with regard to surgery is necessary in
the absence of severe progressive neurological symptoms.
Although the term sciatica is simple and easy to use, it is, in fact, an
archaic and confusing term [1]. For most researchers and clinicians, it refers to a radiculopathy, involving one of the lower
extremities, and related to disc herniation (DH). As such, the term
sciatica is too restrictive as nerve roots from L1 to L4 may also be
involved in the same process. However, even more confusing is the
fact that patients, and many clinicians alike, use sciatica to describe

* Corresponding author. Tel.: 33 2 47 47 59 17; Fax: 33 2 47 47 46 39.

E-mail address: valat@med.univ-tours.fr (J.-P. Valat).
1521-6942/$ see front matter 2009 Published by Elsevier Ltd.
doi:10.1016/j.berh.2009.11.005

242

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

any pain arising from the lower back and radiating down to the leg.
The majority of the time, this painful sensation is referred pain
from the lower back and is neither related to DH nor does it result
from nerve-root compression. Although differentiating the radicular pain from the referred pain may be challenging for the clinician, it is of primary importance. This is because the epidemiology,
clinical course and, most importantly, therapeutic interventions
are different for these two conditions. It should, however, be
emphasised that the quality of the available evidence is rather
limited due to a considerable heterogeneity in the study populations included in the trials. This makes generalisation of ndings across studies, and to routine clinical practice, a challenge [2].
Prevalence estimates of radicular pain related to DH also vary
considerably between studies, which is, in part, due to differences
in the denitions used [3].
A recent review showed that the prevalence of sciatic symptoms is
rather variable, with values ranging from 1.6% to 43% [3]. If stricter
denitions of sciatica were used, for example, in terms of pain
distribution and/or pain duration, lower prevalence rates were
reported. Studies in working populations with physically
demanding jobs consistently report higher rates of sciatica
compared with studies in the general population.
2009 Published by Elsevier Ltd.

Pathophysiology
For decades, sciatica was considered to be merely the result of nerve-root compression by DH.
However, clinical observations contradicting this purely mechanical hypothesis have been published
for over 50 years and are summarised in Table 1. Other hypotheses with regard to the mechanism(s) of
sciatica have been extensively studied in animal and human experiments and are described below.
The role of inammation and tumour necrosis factor-alpha Animal models: The presence of an
inamed neural element in DH was rst described in the early 1950s [4]. However, it was only in the
early 90s that Saal et al [5]. reported elevated levels of phospholipids A2 (PLA2) a key enzyme in the
cascade of inammation in lumbar DH in symptomatic patients. Although a subsequent study did not
support any direct implication of PLA2[6] in radiculopathy, animal-model experiments show that the
nucleus pulposus (NP), the core part of the intervertebral disc (IVD) and the main component of DH,
presents inammatory properties [7,8]. Deposition of NP on nerve roots induces pain, electrical
dysfunction of the nerve and histological modications [911]. Similarities between the latter (i.e.,
axonal changes, characteristic myelin injury, increased vascular permeability and intravascular coagulation) and the effect of an endoneural tumour necrosis factor-alpha (TNF-a) injection were rst
reported by Olmarker [12]. The role of TNF-a, already demonstrated in these publications, was later
conrmed in a study in which the effects of NP on the nerve root were reproduced by similarly
depositing a small quantity of TNF-a extracted from the NP on the nerve [13]. Lastly, the inhibitory
effect of specic TNF-a-blocking agents, commonly used to treat patients with inammatory diseases,
was reported in several animal models of radiculopathy [1418]. However, it is noteworthy that TNF
knockout (KO) mice still develop histological modications and clinical symptoms, although to a lesser
degree, than wild-type (WT) mice [19]. The involvement of other inammatory mediators (interleukin
(IL)-1, IL-6, prostaglandin E2 and nitric oxide)[20,21] as well as the hypothesis of a local auto-immune
reaction [22], have not been studied as extensively. Furthermore, the role of mechanical compression
has not been totally discarded as it may potentiate the effect of inammation [23]. Thus, when light
compression is applied over a long period of time on the nerve root, it induces the production of
inammatory cytokines [24].
Human experiments: Levels of several inammatory cytokines (TNF-a, IL-1 and IL-6) are increased
in degenerative human IVD compared to normal IVD [25,26]. Within DH, only highly sensitive methods
enable the detection of TNF-a [27,28], whereas experiments using immunoassays (enzyme-linked

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

243

Table 1
Observations refuting a purely mechanical hypothesis.
B
B
B
B
B
B
B
B

Presence of disc herniation in asymptomatic subjects

Pressure on normal nerve roots does not cause pain
Pressure on adjacent level nerve roots does not cause pain
Severe symptoms without evidence of nerve root compression
Symptom severity does not correlate with the size of the disc herniation
Outcome might be favourable with conservative treatment
Outcome might be favourable despite persistence of the disc herniation
Discectomy has only moderate long-term success

immunosorbent assay (ELISA)) have reported negative results [2931], suggesting that TNF-a is present
at extremely low levels. In a recent study comparing the levels of TNF-a in patients with radicular pain
with levels in patients with low back pain, increased levels of TNF-a were found in the peri-radicular fat
surrounding the inamed nerve root, but not in the DH itself [32]. This experiment does not permit the
identication of the actual source of production of TNF-a, and more studies are needed to explore the
respective roles of granulation tissue [33], Schwann cells[34] and adipocytes [35].
Highly positive results from early open-label clinical trials studying TNF inhibitors (iniximab[36]
and etanercept[37]) in patients with radicular pain due to DH were not conrmed by a rst randomised
trial with a single infusion of iniximab [38]. However, a second randomised, placebo-controlled study
with two subcutaneous injections of adalimumab reported a slight but signicant effect on leg pain and
an important decrease in the need for surgery at 6 months [39]. Positive results were also reported in
a placebo-controlled doseresponse study on the foraminal injection of very low doses of etanercept
[40], opening new prospects for the future development of an efcient therapeutic procedure for these
patients.
Clinical features
Sciatica is a symptom rather than a specic diagnosis. The most important symptom is leg pain
radiating below the knee and into the foot and toes. Clinical ndings of neurological decit, such as
muscle weakness and reex changes, may also be present. Inw90% of cases, sciatica is caused by
a herniated disc with nerve-root compression, but lumbar canal or foraminal stenosis and (less often)
tumours or cysts are other possible causes.

Diagnosis
Sciatica is mainly diagnosed by history taking and physical examination. By denition, patients have
pain radiating into the leg. Sciatica is characterised by radiating pain that follows a dermatomal pattern
and patients may report sensory symptoms. The diagnostic value of clinical history and physical
examination in sciatica, however, has not been well studied [41]. No history items or physical examination tests have both high sensitivity and high specicity. The straight-leg-raising test (or Lasegues
sign) is the most commonly used physical test. The sensitivity of the straight-leg-raising test is estimated to be 91%, specicity 26%.4[2] The crossed straight-leg-raising test has a pooled specicity of
88%, but a sensitivity of only 29% [42]. Overall, if a patient reports the typical radiating pain in one leg
combined with a positive result on one or more neurological tests indicating nerve-root tension or
neurological decit, the diagnosis of sciatica appears justied. Signs and symptoms that help to
distinguish between sciatica and non-specic low-back pain are unilateral leg pain greater than lowback pain, pain radiating to foot or toes, numbness and paraesthesia in the same distribution, straightleg-raising test induces more leg pain and localised neurology, which is limited to one nerve root.
In acute sciatica, diagnostic imaging may only be indicated if there are indications that the symptoms may be caused by underlying pathology (infections, malignancies) other than DH. In patients with
severe symptoms who fail to respond follwoing 68 weeks of non-surgical treatment, imaging might
be useful to identify if a herniated disc with nerve root compression is present. Surgery may only be
indicated if imaging ndings correspond well with the clinical symptoms. DH is highly prevalent

244

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

Table 2
Levels of evidence for efcacy of non-surgical treatments.
Modalities
Non pharmacological modalities
Information to patients
Bed rest
Physical therapy with or without advice to stay active
Acupuncture, spinal manipulation, corsets, tractions
Percutaneous Electrical Nerve Stimulation (PENS), Transcutaneous
Electrical Nerve Stimulation (TENS)
Pharmacological modalities
Analgesic, non-steroidal anti-inammatory drugs, muscle relaxants, drugs
for neuropathic pain (antidepressant, anti-epileptic)
Epidural corticosteroid injection
Transforaminal corticosteroid injection

Level of evidence for efcacy

Not known
Not known for sciatica but strong evidence
that to stay active is better than bed rest
Low
Not known
Low

Not known
Low (short-term)
Low (short-term)

(2036%) in people without symptoms who do not have sciatica [43]. At present, there is no evidence
that one type of imaging method (magnetic resonance imaging (MRI), computed tomography (CT), etc.)
is more accurate than others [44]. X-rays are not recommended for the diagnosis of lumbar DH, because
they cannot visualise discs [44].
Course and prognosis
In general, the clinical course of acute sciatica is favourable. Most pain and related disabilities
resolve within 2 weeks. In a randomised trial that compared non-steroidal anti-inammatory drugs
(NSAIDs) with placebo for acute sciatica in primary care, 60% of the patients recovered within 3 months
and 70% within 12 months [45]. Nearly 50% of patients with acute sciatica included in placebo groups in
randomised trials of non-surgical interventions reported improvement within 10 days and about 75%
reported improvement after 4 weeks [46]. Without surgery, 80% of the patients recover within 8 weeks
and 95% within 1 year [47]. At present, there is no clear insight into factors associated with a poor or
favourable outcome.
Treatment options
Having an idea of the precise efcacy of each treatment option is very important as it provides the
patient the ability to make the best-informed choice.
Non-surgical treatments
There is consensus that treatment of sciatica in the rst 68 weeks should be conservative. In the
presence of red ags, specic causes (e.g., infection, tumour, osteoporotic fractures) may be present and
further investigations, such as imaging and laboratory diagnostics, are indicated. Opioid-resistant pain,
motor decit or cauda equina syndrome are conditions that justify referral to a specialist to discuss
emergency surgery. Except in these cases, imaging and laboratory diagnostic tests are not useful and
optimal management requires a combination of non-pharmacological and pharmacological modalities.
Non-pharmacological modalities (Table 2)
Adequately informing patients about sciatica (causes of sciatica, no need to perform diagnostic
imaging, expected prognosis and possibility of recovery without surgery, etc.) plays an important role
in improving patient satisfaction and likely recovery [48]. Patients should be provided ample information to avoid misunderstandings about the condition. However, no published data have yet investigated the effect of information and education of patients suffering from sciatica.

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

245

In the absence of evidence showing benets [4951], rest and bed rest cannot be recommended, and
such advice has been found to have a harmful effect in acute low-back pain [51]. At present, there is
limited evidence to support the use of physical therapy in treating sciatica. In a recent systematic review
[52], efcacy of physical therapy was reported in four trials and none demonstrated a benet in
comparison with no treatment or other conservative treatment. One randomised clinical trial published
after this article was written supports the use of physical therapy (exercise therapy plus giving information and advice about sciatica), especially for patients reporting severe disability at baseline [53].
No evidence supports the use of acupuncture, spinal manipulation and corsets in cases of
sciatica[52] and these modalities of management cannot be recommended. As there is conicting
evidence concerning traction in several comparisons [54], traction is also not currently recommended.
Transcutaneous electrical nerve stimulation (TENS) is a treatment that uses low-voltage electrical
currents. Percutaneous electrical nerve stimulation (PENS) is similar to TENS, but, with PENS, the
electrical stimulation is passed through the skin into the soft tissue using probes similar to acupuncture
needles. One randomised, single-blind, crossover study [55], comprising 64 patients assessed efcacy
of TENS and PENS versus sham treatment in the management of sciatica. TENS and PENS were
signicantly more effective than sham treatment in decreasing radicular pain. PENS was signicantly
more effective than TENS in improving physical activity and quality of sleep. In case of chronic sciatica,
TENS or PENS could be tested and used if they provide a relief of pain [49], but this recommendation is
based on a single small study only.
Pharmacological modalities (Table 2)
There is no evidence to support the efcacy of analgesics in sciatica [52]. Nevertheless, it is
reasonable to use World Health Organisation (WHO) level I and II analgesics (paracetamol and weak
opioids) for pain relief as long as necessary [48]. In cases of severe pain, strong opioids can be used for
a short period. Patients should be informed about the risks and benets associated with opioid therapy.
A recent Cochrane review found that the efcacy of NSAIDs for patients with sciatica has not been
established [56]. Even compared to placebo, the favourable effects of NSAIDs could not be demonstrated in patients with sciatica. Nevertheless, in clinical practice, they are often tried when insufcient
pain relief has been obtained with paracetamol alone. When NSAIDs are prescribed, long-term use
should be avoided as much as possible.
No evidence supports the efcacy of muscle relaxants for radicular pain [57], although they are
sometimes prescribed for a short period of time, either alone or in association with analgesics or
NSAIDs, for patients with associated low-back pain.
Epidural corticosteroid injections have been widely used in clinical practice for many years. Carette
et al. reviewed the results of 13 randomised controlled studies conducted prior to 1997 [58]. The trials
reported conicting results with ve supporting the use of local corticosteroid and eight not supporting their use for sciatica. Overall, no effect on pain intensity was detectable beyond the rst month
and none of the studies indicated a decrease in the need for surgery. Three additional randomised
controlled trials have been conducted subsequently [5961]. These trials indicate that epidural corticosteroid injections may provide signicant pain relief for 36 weeks, but have no effect on disability,
use of surgery or time to return to work.
A recent review that included a larger number of trials concluded that there was no evidence of
positive short-term effects of epidural corticosteroid other than pain, and that the long-term effects
were unknown [52]. Therefore, epidural corticosteroid injections could be recommended as a means of
reducing pain in the short term although no long-term effects can be expected. Giving more than three
injections does not seem to confer additional effects and are not recommended. Some technical
characteristics, such as the injected volume, may inuence results and should be investigated. The
corticosteroid is usually injected via the interspinous route, although injections may be performed into
the rst sacral foramen or the sacrococcygeal hiatus. No comparative data support the use of injection
under radiological guidance.
Transforaminal periradicular injections (TPIs) are usually performed through the lateral foraminal
approach, under radiological guidance, after injection of a contrast agent. Four randomised controlled
trials have been conducted using different study designs [6265], and TPI was found to be effective in

246

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

three of them. One study showed a reduction in the need for surgery [63]. Of the three studies that
compared epidural injections to TPI [6668], two found that the outcomes were better with the TPI.
Severe complications, such as spinal cord infarctions, have been reported with TPIs [69], which may
limit their usefulness. No evidence supports the use of intradiscal corticosteroid injections.
In cases of chronic sciatica, antidepressants can have an effect on back pain [70], but the effect on
radicular pain is unclear. The use of antidepressants in sciatica is not supported by evidence from
randomised controlled trials [7173]. As sciatica could be considered to be a neuropathic pain, some
drugs indicated for neuropathic pain (e.g., carbamazepine, gabapentin, clonazepam) could be used to
treat sciatica. However, at present, no published evidence supports effectiveness of these drugs for
sciatica. The current challenge is to clearly dene criteria of neuropathic pain because there is no
universal agreement about delineation and classication of this group of disorders [74].
Surgery
The rate of spinal surgery has steadily increased in recent decades; it has, however, shown a great
variation between, and also within, countries. Absolute indications for surgery include altered bladder
function, progressive muscle weakness and opioid-resistant pain intensity. These indications are rare.
The usual indication is to provide more rapid relief of pain and improvement of disability in patients
whose recovery is slow. The results of randomised controlled trials comparing surgical to conservative
treatments are summarised in Table 3.
More than 25 years ago, Weber[75] reported the outcome of 280 patients with L5 or S1 sciatica who
had been consecutively admitted to a neurology department during 1970 and 1971. One hundred and
twenty-six patients with persisting symptoms following 14 days of inpatient conservative treatment
were randomised to either surgery or further conservative treatment consisting of 6 weeks of rehabilitation. Patients were followed up by questionnaire during the rst year, then re-examined at 1, 4
and 10 years. Amongst the patients assigned to surgery, one did not receive an operation while,
amongst the patients assigned to conservative care, 17 (25%) were operated on during the rst year
after a mean of 7.5 months. At 1 year, the surgical group had better results than the conservative group
(p 0.0015) with intention-to-treat and as-treated analyses. At 4 years, a non-statistically signicant
trend for a more favourable effect of surgery was noted. The number of relapses was not different and,
at 10 years, there were no signicant differences in outcome between the two groups. The results of
this trial suggested that surgery was associated with better outcomes at 1 year, but the differences were
less pronounced in the results after 4 years.
The study by Osterman and colleagues in 2006[76] compared the results of microdiscectomy and
conservative management. Patients with sciatica who were referred to an orthopaedic consultation
were randomised to one of two groups: the microdiscectomy group or the controlled group with
physiotherapeutic instruction. The primary outcome measure was leg-pain intensity and the secondary
outcome measures included back pain and work ability (the visual analogue scale, VAS), quality-of-life
measure, disability score, depression, satisfaction with treatment and global assessment. Evaluations
were performed at 6 weeks, 3 and 6 months and 1 and 2 years. Twenty-eight patients were randomised
in each group. Eleven patients in the control group (40%) crossed over to surgery. In the intentionto-treat analysis, at each follow-up, the surgical group fared better, but the differences between groups
were statistically signicant only for leg-pain intensity at 6 weeks and for satisfaction with treatment at
6 weeks, 6 months and 2 years. The as-treated analysis showed no statistically signicant differences
between the two groups. In terms of results, the surgical patients with more than median leg pain
(40 mm) did not differ from conservatively treated patients or patients with less leg pain. In a subgroup
analysis, discectomy was superior to conservative treatment when the herniation was at the L4L5
level.
The Spine Outcome Research Trial (SPORT) was conducted at 13 multidisciplinary spine practices in
the US [77]. Patients were randomised to two groups: surgery (standard open discectomy) or nonoperative care (usual care, at least physical therapy, home exercises, NSAIDS, etc.).gggbbb The primary
study outcome measures were Short Form 36 (SF-36) bodily pain and physical function scales and
Oswestry Disability Index (ODI) at 6 weeks, 3 months, 6 months, 1 year and 2 years. The secondary
measures were patient self-reported improvement, work status, satisfaction with current symptoms

Author (year)

Number patients
per group

Inclusion criteria

Duration
of sciatica

Non adherence

Main Results (Intention to treat analysis)

Weber
(1983) [75]

60 surgery
66 conservative

3.5 weeks

Surgery: 1 no operation
Conservative: 17/66 (65%)
had operation during 1st year

1 and 4 years:
surgery > conservative
on patients subjective statement

Osterman
(2006) [76]

28 surgery
28 conservative

L5 or S1 sciatica,
persisting symptoms after
14 days conservative
treatment in hospital
Sciatica 612 weeks

11 weeks

6 weeks: surgery > conservative treatment for

leg pain but benet not sustained

Weinstein
(2006) [77]

232 surgery
240 conservative

Surgery: all had operation

Conservative: 11/28 (39%)
had operation
Surgery: 92/232 (40%) no
operation Conservative:
107/240 (45%) had operation

Peul
(2007) [81]

141 surgery
142 conservative

Surgery: 16/141 (39%) no

operation Conservative:
55/142 (39%) had operation

Relief of pain faster for surgery

Probability of recovery after 1 year: 95% in each group

Radicular pain > 6 weeks,

positive nerve root tension
sign or neurologic decit
and DH on MRI or CT
Severe sciatica 612
weeks with
conrmed DH

80% < 6
months

9 weeks

Treatment effect favoured surgery at 3 months, 1 and 2

years but not statistically signicant

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

Table 3
Randomized controlled studies comparing surgical to conservative treatment for sciatica due to lumbar disc herniation.

247

248

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

and with care and sciatica bothersome index. Patients were enrolled in either the randomised trial or
a concurrent observational cohort [78]. Only the trial results are presented in this article. Four hundred
and seventy-two patients were included in the randomised trial. A high rate of non-adherence to
treatment group was recorded: amongst the 232 patients assigned to receive surgery, 140 (60%)
underwent surgery at 2 years; amongst the 240 assigned to non-operative care, 107 (45%) underwent
surgery at 2 years. Characteristics of crossover patients were statistically different from patients who
did not cross over (those more likely to cross over to receive surgery tended to have lower incomes,
worse baseline symptoms, more baseline disability on the ODI and were more likely to rate their
symptoms as getting worse at enrolment than the other patients receiving non-operative treatment).
In the intent-to-treat analysis, there were strong improvements in both treatment groups; for each
measure and at each follow-up point (3 months and 1 and 2 years) the treatment effect favoured
surgery but was not statistically signicant for the primary outcomes at any follow-up points. For the
secondary outcomes of sciatica discomfort and self-rated progress, there was greater improvement
with surgery (p 0.003 and p 0.04, respectively). The as-treated analysis showed strong statistically
signicant advantages for surgery at all follow-up times.
Four-year follow-up results for the SPORT trial were published in 2008 [79]. The non-operative
group improved signicantly, and this improvement persisted throughout the 4-year period. In the
intent-to-treat analysis of the randomised cohort, all measures over 4 years favoured surgery, but there
were no statistically signicant treatment effects in any of the primary outcome measures at any time
interval. Treatment effects for the primary outcomes in the combined as-treated analysis were
maintained and statistically signicant in favour of surgery out to 4 years. Except for work status, all
secondary measures retained a signicant benet for surgery at 4 years. The differences in outcome
between treatment groups remained relatively constant after the rst year. Return to work appeared to
be independent of the treatment received, and did not follow improvement in pain, function or
satisfaction with treatment. The rate of reoperation was 10% at 4 years, with 50% listed as recurrent
herniations at the same level.
Previous studies have reported that lumbar discectomy is less successful for relief of back pain than
leg pain. It is interesting to look at the results on back pain in the SPORT study [80]. Back pain improved
both in surgical and non-operative patients, but surgical patients improved signicantly more. Patients
who underwent surgery were more likely to report no back pain compared with the non-operative
patients at each follow-up period (28% vs. 12% at 3 months and 25.5% vs. 17.6% at 2 years).
A more recent trial[81] questioned the effect of early surgery in patients with 612 weeks of severe
sciatica. This multicentre, prospective, randomised trial included surgical candidates with MRIconrmed DH, incapacitating sciatica with a positive nerve-root tension sign or a corresponding
neurological decit. Patients were randomised to either early surgery or usual care. If sciatica persisted
for 6 months following randomisation, microdiscectomy was offered and, if there was increasing leg
pain or progressive neurological decits, this led to surgery earlier than 6 months. The outcome
measures were the Roland Disability Questionnaire for Sciatica (RDQ), VAS for leg pain and the Likert
scale of global recovery at 2, 4, 8, 12, 26, 38 and 52 weeks. Recovery was dened as complete or nearly
complete disappearance of symptoms as measured on a seven-point Likert scale. Of the 283 eligible
patients, 141 were assigned to receive early surgical treatment (16 recovered prior to surgery) and 142
to conservative treatment. Fifty-ve (39%) patients in the conservatively treated group underwent
surgery during the rst year (median period: 14.6 weeks) for intractable pain. In the intent-to-treat
analysis, disability and back pain were not statistically different over 52 weeks in the two groups. For
leg pain, an early effect appeared to favour surgery from 2 to 26 weeks, but the effect was nearly equal
at 1 year. Early surgery had a positive effect on the speed of recovery during the rst 36 weeks
(p < 0.001) but the difference decreased over time with similar recovery of about 95% noted for both
groups after 52 weeks. The median time of recovery was 4 weeks for early surgery and 12 weeks for
prolonged conservative treatment. The subgroup of 55 patients with persistent sciatica and conservative treatment followed by surgery had a similar improvement in these scores at 1 year as compared
with patients allocated to early surgery. The short-term benet of early surgery continued to narrow
between 6 and 24 months. Patient satisfaction decreased slightly between 1 and 2 years for both
groups. Outcomes did not change during the second year, when 20% of all patients reported an
unsatisfactory outcome. In the conservatively treated group, baseline factors predicting surgery were:

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

249

higher pain intensity and functional limitation [82]. For a 20-mm increase in pain on the VAS score,
there was an increased chance of undergoing surgery (mutually adjusted odds ratio: 1.7 (95% condence interval (CI): 1.12.7)) and, for a three-point deterioration in RDQ score, the increased risk was
1.8 (95% CI: 1.22.9). This was complemented by the patients preferences.
From these randomised controlled studies, we can conclude that surgery provides faster relief for
patients with sciatica, whereas the results of surgery and conservative care are similar at 1 year and
beyond. A rate of 510% of patients needing surgery has been observed although conservative treatments can be prolonged without harmful effects for patients [81]. Recent trials have demonstrated the
difculty of performing randomised controlled trials investigating surgical treatment due to the high
rate of non-adherence to group allocation.
In practice, and in accordance with most authors [42], following 68 weeks of symptoms that have
failed to respond to conservative treatment, if a DH is found to be the cause of sciatica then patients
may become surgical candidates. Surgical discectomy can be scheduled only for carefully selected
patients with sciatica due to lumbar disc prolapse. When no surgery can be performed (contraindication to surgery, refusal of patient, etc.) and pain becomes chronic, if the diagnosis of DH is conrmed,
minimally invasive techniques (percutaneous treatment) could be proposed to patients, although the
efcacy of these treatments remains unclear and chymopapain chemonucleolysis is no longer available
in many countries [83].
Conclusion
Our understanding of the pathophysiology of radicular pain due to DH has greatly improved over
the past years. Some intriguing points still remain to be explored. First, TNF is a key mediator in many
animal models; however, TNF-KO mice still develop mild radiculopathy. Second, TNF inhibitors do not
appear to be as effective in humans as they are in animal models. There is a need to further study the
differences between animal and human radiculopathy due to DH and also to design a comprehensive
approach to test TNF inhibitors in humans.
The clinical course of sciatica is considered to be favourable in most patients, probably because of the
favourable natural course of sciatica. Most of the oft-used treatments have at present only limited
(physical therapy, epidural corticosteroid injections, transforaminal periradicular injections of corticosteroid, transcutaneous electrical nerve stimulation, etc.) or no evidence (patient information, analgesic,
NSAIDS, muscle relaxants, antidepressant, anticonvulsant, etc.) supporting their efcacy, and better trials
are needed. Currently available conservative treatments may decrease pain without modifying the longterm clinical course of sciatica. Based upon four randomised controlled studies, discectomy is effective in
the short term but the results are similar to conservative care at 1 year and beyond.
Like Deyo [84], we can conclude that for patients with persistent sciatica (>68 weeks), and
without major neurological decit, there seems to be a reasonable choice between surgical and prolonged non-surgical treatment. Preference for treatment may be inuenced by the severity of the
symptoms, patients willingness to wait for spontaneous healing, and their aversion to surgical risk.

Practice points
 Inammation and compression are important components of radiculopathy due to lumbar
DH.
 TNF-a is a key mediator in animal models but its exact contribution in human radiculopathy
due to DH is still a matter of debate.
 Provide the patient with adequate information about sciatica (causes, no need to perform
diagnostic imaging, expected prognosis and possibility of recovery without surgery, etc.).
 Provide advice to stay active and continue daily activities.
 Surgery provides faster relief than conservative treatment for patients with greater than 68
weeks of sciatica.
 Results between surgery and conservative care are similar at 1 year and beyond

250

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

Research agenda
 Clarify the denition of sciatica.
 Investigate other inammatory mediators involved in the development of radiculopathy in
TNF-KO mice.
 Explore differences between animal models and human radiculopathy due to DH.
 Design dose-ranging studies and randomised control trials to better characterise the efcacy
of TNF inhibitors.
 Investigate the effect of patient education on the clinical course of sciatica treated with
conservative treatment.
 Develop and investigate new treatments according to a better understanding of pain
(mechanical, biochemical, neuropathic components, etc.).
 Evaluate the efcacy of most of the available conservative treatments.
 Evaluate the efcacy of mini-invasive surgery.
 Develop and investigate TNF-a inhibitors and new drugs for neuropathic pain.
 Investigate the inuence of patient preferences (e.g., surgical versus non-surgical
approaches) on treatment outcomes.

Conict of interest
J.-P. Valat has been an expert for Pzer and for Grunenthal.
S. Genevay has received clinical research grants from Pzer and Abbott.
B. Koes has been involved in several clinical studies and systematic reviews related to sciatica.
M. Marty has been an expert for Pzer, Sano, and Grunenthal.
S. Rozenberg has been an expert for Grunenthal and Sano.
References
[1] Fairbank JC. Sciatic: an archaic term. BMJ 2007;335:112.
[2] Genevay S, Atlas SJ, Katz JN. Variation in eligibility criteria from studies of radiculopathy due to a herniated disc and of
neurogenic claudication due to lumbar spinal stenosis: A structured literature review. Spine in press.
[3] Konstantinou K, Dunn KM. Sciatica: review of epidemiological studies and prevalence estimates. Spine 2008;33:246472.
[4] Lindahl O, Rexed B. Histologic changes in spinal nerve roots of operated cases of sciatica. Acta Orthop Scand 1951;20:
21525.
[5] Saal JS, Franson RC, Dobrow R, et al. High levels of inammatory phospholipase A2 activity in lumbar disc herniations.
Spine 1990;15:6748.
[6] Gronblad M, Virri J, Ronkko S, et al. A controlled biochemical and immunohistochemical study of human synovial-type
(group II) phospholipase A2 and inammatory cells in macroscopically normal, degenerated, and herniated human lumbar
disc tissues. Spine 1996;21:25318.
[7] McCarron RF, Wimpee MW, Hudkins PG, et al. The inammatory effect of nucleus pulposus. A possible element in the
pathogenesis of low-back pain. Spine 1987;12:7604.
[8] Olmarker K, Blomquist J, Stromberg J, et al. Inammatogenic properties of nucleus pulposus. Spine 1995;20:6659.
[9] Byrod G, Rydevik B, Nordborg C, et al. Early effects of nucleus pulposus application on spinal nerve root morphology and
function. Eur Spine J 1998;7:4459.
[10] Olmarker K, Nordborg C, Larsson K, et al. Ultrastructural changes in spinal nerve roots induced by autologous nucleus
pulposus. Spine 1996;21:4114.
[11] Olmarker K, Rydevik B, Nordborg C. Autologous nucleus pulposus induces neurophysiologic and histologic changes in
porcine cauda equina nerve roots. Spine 1993;18:142532.
[12] Olmarker K, Larsson K. Tumor necrosis factor alpha and nucleus-pulposus-induced nerve root injury. Spine 1998;23:253844.
[13] Igarashi T, Kikuchi S, Shubayev V, et al. Volvo award winner in basic science studies: exogenous tumor necrosis factoralpha mimics nucleus pulposus-induced neuropathology. Molecular, histologic, and behavioral comparisons in rats. Spine
2000;25:297580. 2000.
[14] Homma Y, Brull SJ, Zhang JM. A comparison of chronic pain behavior following local application of tumor necrosis factor
alpha to the normal and mechanically compressed lumbar ganglia in the rat. Pain 2002;95:23946.
[15] Murata Y, Onda A, Rydevik B, et al. Selective inhibition of tumor necrosis factor-alpha prevents nucleus pulposus-induced
histologic changes in the dorsal root ganglion. Spine 2004;29:247784.
[16] Olmarker K, Nutu M, Storkson R. Changes in spontaneous behavior in rats exposed to experimental disc herniation are
blocked by selective TNF-alpha inhibition. Spine 2003;28:163541.

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

251

[17] Sasaki N, Kikuchi S, Konno S, et al. Anti-TNF-alpha antibody reduces pain-behavioral changes induced by epidural
application of nucleus pulposus in a rat model depending on the timing of administration. Spine 2007;32:4136.
[18] Schafers M, Lee DH, Brors D, et al. Increased sensitivity of injured and adjacent uninjured rat primary sensory neurons to
exogenous tumor necrosis factor-alpha after spinal nerve ligation. J Neurosci 2003;23:302838.
[19] Yamashita M, Ohtori S, Koshi T, et al. Tumor necrosis factor-alpha in the nucleus pulposus mediates radicular pain, but not
increase of inammatory peptide, associated with nerve damage in mice. Spine 2008;33:183642.
[20] Kawakami M, Tamaki T, Hayashi N, et al. Possible mechanism of painful radiculopathy in lumbar disc herniation. Clin
Orthop Relat Res 1998;351:24151.
[21] Ozaktay AC, Cavanaugh JM, Asik I, et al. Dorsal root sensitivity to interleukin-1 beta, interleukin-6 and tumor necrosis
factor in rats. Eur Spine J 2002;11:46775.
[22] Geiss A, Larsson K, Junevik K, et al. Autologous nucleus pulposus primes T cells to develop into interleukin-4-producing
effector cells: an experimental study on the autoimmune properties of nucleus pulposus. J Orthop Res 2009;27:97103.
[23] Olmarker K, Myers RR. Pathogenesis of sciatic pain: role of herniated nucleus pulposus and deformation of spinal nerve
root and dorsal root ganglion. Pain 1998;78:99105.
[24] Kobayashi S, Baba H, Uchida K, et al. Effect of mechanical compression on the lumbar nerve root: localization and changes
of intraradicular inammatory cytokines, nitric oxide, and cyclooxygenase. Spine 2005;30:1699705.
[25] Le Maitre CL, Freemont AJ, Hoyland JA. The role of interleukin-1 in the pathogenesis of human intervertebral disc
degeneration. Arthritis Res Ther 2005;7:R73245.
[26] Weiler C, Nerlich AG, Bachmeier BE, et al. Expression and distribution of tumor necrosis factor alpha in human lumbar
intervertebral discs: a study in surgical specimen and autopsy controls. Spine 2005;30:4453.
[27] Ahn SH, Cho YW, Ahn MW, et al. mRNA expression of cytokines and chemokines in herniated lumbar intervertebral discs.
Spine 2002;27:9117.
[28] Takahashi H, Suguro T, Okazima Y, et al. Inammatory cytokines in the herniated disc of the lumbar spine. Spine 1996;21:
21824.
[29] Burke JG, Watson RW, McCormack D, et al. Intervertebral discs which cause low back pain secrete high levels of proinammatory mediators. J Bone Joint Surg Br 2002;84:196201.
[30] Kang JD, Georgescu HI, McIntyre-Larkin L, et al. Herniated lumbar intervertebral discs spontaneously produce matrix
metalloproteinases, nitric oxide, interleukin-6, and prostaglandin E2. Spine 1996;21:2717.
[31] Nygaard OP, Kloster R, Dullerud R, et al. No association between peridural scar and outcome after lumbar microdiscectomy. Acta Neurochir (Wien) 1997;139:1095100.
[32] Genevay S, Finckh A, Payer M, et al. Elevated levels of tumor necrosis factor-alpha in periradicular fat tissue in patients
with radiculopathy from herniated disc. Spine 2008;33:20416.
[33] Virri J, Gronblad M, Seitsalo S, et al. Comparison of the prevalence of inammatory cells in subtypes of disc herniations and
associations with straight leg raising. Spine 2001;26:23115.
[34] Qin Y, Cheng C, Wang H, et al. TNF-alpha as an autocrine mediator and its role in the activation of Schwann cells. Neurochem Res 2008;33:107784.
[35] Rosen ED, Spiegelman BM. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 2006;444:84753.
[36] Karppinen J, Korhonen T, Malmivaara A, et al. Tumor necrosis factor-alpha monoclonal antibody, iniximab, used to
manage severe sciatica. Spine 2003;28:7503.
[37] Genevay S, Stingelin S, Gabay C. Efcacy of etanercept in the treatment of acute, severe sciatica: a pilot study. Ann Rheum
Dis 2004 Sep;63(9):11203.
[38] Korhonen T, Karppinen J, Paimela L, et al. The treatment of disc herniation-induced sciatica with iniximab: results of
a randomized, controlled, 3-month follow-up study. Spine 2005;30:27248.
[39] Genevay S, Viatte S, Finckh A et al. Adalimumab in the Treatment of Acute Severe Sciatica, a Randomized Double Blind
Placebo Controlled Study. Arthritis Rheum (in press).
[40] Cohen SP, Bogduk N, Dragovich A, et al. Randomized, double-blind, placebo-controlled, dose-response, and preclinical
safety study of transforaminal epidural etanercept for the treatment of sciatica. Anesthesiology 2009;110:111626.
[41] Vroomen PCAJ, MCTFM de Krom, Knottnerus JA. Diagnostic value of history and physical examination in patients suspected of sciatica due to disc herniation: a systematic review. J Neurol 1999;246:899906.
[42] Deville WLJM, Windt DAWM, van der Dzaferagic A, et al. The test of Lasegue: systematic review of the accuracy in
diagnosing herniated discs. Spine 2000;25:11407.
[43] Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al. Magnetic resonance imaging of the lumbar spine in people without
back pain. N Engl J Med 1994;331:6973.
[44] Jarvik JG, Deyo RA. Diagnostic evaluation of low back pain with emphasis on imaging. Ann Intern Med 2002;137:58697.
[45] Weber H, Holme I, Amlie E. The natural course of acute sciatica with nerve root symptoms in a double blind placebocontrolled trial of evaluating the effect of piroxicam (NSAID). Spine 1993;18:14338.
[46] Vroomen PC, de Krom MC, Slofstra PD, et al. Conservative treatment of sciatica: a systematic review. J Spinal Disord 2000;
13:4639.
[47] Legrand E, Bouvard B, Audran M, et al. Sciatica from disk herniation: medical treatment or surgery? Joint Bone Spine 2007;
74(6):5305.
[48] Koes BW, van Tulder MW, Peul WC. Diagnosis and treatment of sciatica. BMJ 2007;334:13137.
[49] Vroomen PC, de Krom MC, Wilmink JT, et al. Lack of effectiveness of bed rest for sciatica. N Engl J Med 1999;340:41823.
[50] Hofstee DJ, Gijtenbeek JMM, Hoogland PH, et al. Weisteinde sciatica trial: A randomized controlled study of bed rest and
physiotherapy for acute sciatica. J Neurosurg Spine 2002;96:459.
[51] Hagen KB, Jamtvedt G, Hilde G, et al. The updated Cochrane review of bed rest for low back pain and sciatica. Spine 2005;
30:5426.
[52] Luijsterburg PA, Verhagen AP, Ostelo RW, et al. Effectiveness of conservative treatments for the lumbosacral radicular
syndrome: a systematic review. Eur Spine J 2007;16:88199.
[53] Luijsterburg PA, Verhagen AP, Ostelo RW, et al. Physical therapy plus general practionnerscare versus generals care alone
for sciatica: a randomised clinical trial with a 12-month follow up. Eur Spine Journal 2008;17:50917.

252

J.-P. Valat et al. / Best Practice & Research Clinical Rheumatology 24 (2010) 241252

[54] Clarke JA, van Tulder MW, Blomberg SEI, et al. Traction for low-back pain with or without sciatica. Cochrane Database of
Systematic Reviews 2007. Issue 2. Art. No.: CD003010.DOI: 10.1002/14651858.CD003010.pub4.
[55] Ghoname EA, White PF, Ahmed HE, et al. Percutaneous electrical nerve stimulation: an alternative to TENS in the
management of sciatica. Pain 1999;83:1939.
[56] Roelofs PDDM, Deyo RA, Koes BW, et al. Non-steroidal antiinammatory drugs for low back pain. Cochrane Database of
Systematic Reviews 2008;. doi:10.1002/14651858.CD000396.pub3. Issue 1. Art. No.: CD000396.
[57] van Tulder MW, Touray T, Furlan AD, et al. Muscle relaxants for nonspecic low back pain: a systematic review within the
framework of the cochrane collaboration. Spine 2003;28:197892.
[58] Carette S, Leclaire R, Marcoux S. Epidural corticosteroid injections for sciatica due to herniatednucleus pulposus. N Engl
J Med 1997;336:163440.
[59] Valat JP, Gireaudeau B, Rozenberg S. Epidural corticoid injections for sciatica: a randomized, double blind, controlled trial.
Ann Rheum Dis 2003;62:63943.
[60] Arden NK, Price C, Reading I. A multicentre randomized controlled trial of epidural corticosteroid injections for sciatica:
the WEST study. Rheumatology (Oxford) 2005;44:1399406.
[61] Wilson-MacDonald J, Burt G, Grifn D, et al. Epidural steroid injection for nerve root compression. A randomised,
controlled trial. Bone Joint Surg Br 2005;87:3525.
[62] Riew KD, Yin Y, Gilula L, et al. The effect of nerve-root injections on the need for operative treatment of lumbar radicular
pain. J Bone Joint Surg 2000;82A:158993.
[63] Karppinen J, Malmivaara A, Kurunlahti M, et al. Periradicular inltration for sciatica. A randomized controlled trial. Spine
2001;26:105967.
[64] Vad VB, Bhat AL, Lutz GE, et al. Transforaminal epidural steroid injections in lumbosacral radiculopathy. Spine 2002;27:116.
[65] Ng L, Chaudhary N, Sell P. The efcacy of corticosteroids in periradicular inltration for chronic radicular pain.
A randomized, double-blind, controlled trial. Spine 2005;30:85762.
[66] Kolsi I, Delecrin J, Berthelot JM, et al. Efcacy of nerve root versus interspinious injections of glucocorticoids in the
treatment of disk-related sciatica. A pilot, prospective, randomized, double-blind study. Jone Bone Spine 2000;67:1138.
[67] Thomas E, Cyteval C, Abiad L. Efcacy of transforaminal versus interspinious corticosteroid injection in discal radiculalgia
a prospective, randomised, double-blind study. Clin Rheumatol 2003;22:299304.
[68] Shaufele MK, Hatch L, Jones W. Interlaminar versus transforaminal epdural injections r the treatment of symptomatic
lumbar intervertebral disc herniations. Pain Physician 2006;9:3616.
[69] Wybier M. Transforaminal epidural corticosteroid injections and spinal cord infarction. Jone Bone Spine 2008;75:5235.
[70] Perrot S, Javier RM, Marty M, et al. Is there any evidence to support the use of anti-depressants in painful rheumatological
conditions? Systematic review of pharmacological and clinical studies. Rheumatology (Oxford) 2008;47:111723.
[71] Atkinson JH, Slater MA, Williams RA, et al. A placebo-contolled randomized clinical trial of nortryptiline for chronic low
back pain. Pain 1998;76:28796.
[72] Khoromi S, Patsalides A, Parada S, et al. Topiramate in chronic lumbar radicular pain. J Pain 2005;6:82936.
[73] Khoromi S, Cui L, Nackers L. Morphine, nortriptyline and their combination vs placebo in patients with chronic lumbar
root pain. Pain 2007;130:6675.
[74] Bennett MI, Smith BH, Torrance N, et al. Can pain be more or less neuropathic? comparison of symptom assessment tools
with ratings of certianty by clinicians. Pain 2006;122:28994.
[75] Weber H. Lumbar disc herniation. A controlled, prospective study with ten years of observation. Spine 1983;8:13140.
sterman H, Seitsalo S, Karpinnen J. Effectiveness of microdiscectomy for lumbar disc herniation. Spine 2006;31:240914.
[76] O
[77] Weinstein JN, Tosteson TD, Lurie JD, et al. Surgical vs nonoperative treatment for lumbar disk herniation. The Spine
Outcome Research Trial (SPORT): a randomized trial. JAMA 2006;296:244150.
[78] Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical vs nonoperative treatment for lumbar disk herniation. The Spine
Outcome Research Trial (SPORT). Observational cohort. JAMA 2006;296:24519.
[79] Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical versus non operative treatment for lumbar disc herniation: four-year
results for the Spine Patient Outcomes Research Trial (SPORT). Spine 2008 Dec 1;33(25):2789800.
[80] Pearson AM, Blood EA, Frymoyer JW, et al. SPORT lumbar intervertebral disk herniation and back pain. Does treatment,
location or morphology matter? Spine 2008;33:42835.
[81] Peul WC, Van Houvelingen HC, van den Hout WB, et al. Surgery versus prolonged conservative treatment for sciatica.
N Engl J Med 2007;356:224556.
[82] Peul WC, Brand R, Thomeer RT, et al. Improving prediction of inevitable surgery during non-surgical treatment of
sciatica. Pain 2008;138:5716.
[83] Gibson JNA, Waddell G. Surgical interventions for lumbar disc prolapse. Cochrane Database of Systematic Reviews 2007;.
doi:10.1002/14651858.CD001350.pub4. Issue 2. Art. No.: CD001350.
[84] Deyo RA. Back surgery. Who needs it? NEJM 2007;356:223943.