For the full versions of these articles see bmj.

com
Clinical Review

Diagnosis and management of

neuropathic pain
R Freynhagen,1 M I Bennett2

1
Department of Anaesthesiology, Neuropathic pain arises from damage, or pathological Box 1 | Examples of neuropathic pain syndromes
Critical Care Medicine, Pain change, in the peripheral or central nervous system. Peripheral nervous system focal and multifocal lesions
Therapy & Palliative Care, Pain It is usually a chronic condition that can be difficult
Center Lake Starnberg, Benedictus Post-herpetic neuralgia
Krankenhaus Tutzing, 82327 to treat because standard treatment with conventional Cranial neuralgias (such as trigeminal neuralgia,
Tutzing, Germany analgesics does not typically provide effective relief of glossopharyngeal neuralgia)
2
International Observatory on End pain. Patients with neuropathic pain commonly present Diabetic mononeuropathy
of Life Care, School of Health and to primary care professionals, but making a diagnosis Nerve entrapment syndromes
Medicine, Lancaster University,
Lancaster LA1 4YT may be difficult. Neuropathic pain is usually associated Plexopathy from malignancy or radiation
Correspondence to: R Freynhagen with substantially greater impairment of quality of life Phantom limb pain
r.freynhagen@krankenhaus- compared with other types of chronic pain, and the Post-traumatic neuralgia (such as nerve root compression,
tutzing.de post-thoracotomy)
disorder is a large cost burden on healthcare services.
Ischaemic neuropathy
Cite this as: BMJ 2009;339:b3002 In this review, we provide an overview of published
Peripheral nervous system generalised polyneuropathies
doi: 10.1136/bmj.b3002 evidence to help clinicians recognise and manage
Metabolic/nutritional—Diabetes mellitus, amyloid,
patients with neuropathic pain. pellagra, beriberi, multiple nutritional deficiency,
hypothyroidism
What is neuropathic pain and who gets it? Toxic—Alcohol, platinum, or taxane based chemotherapy,
A group of specialists of the International Association isoniazid, antiretroviral drugs
for the Study of Pain defines neuropathic pain as “pain Infective/autoimmune—HIV, acute inflammatory
arising as a direct consequence of a lesion or disease polyneuropathy (Guillain-Barré syndrome),
neuroborreliosis (Bannwarth’s syndrome)
affecting the somatosensory system.”1 In contrast to
Heriditary—Fabry’s disease
inflammatory or nociceptive pain, which is caused by
Malignancy—Carcinomatosis
actual tissue damage or potentially tissue damaging Others—Idiopathic small fibre neuropathy
stimuli, neuropathic pain is produced either by damage Central nervous system lesions
to, or pathological change in, the peripheral or central Spinal cord injury
nervous system, the system that normally signals pain. Prolapsed disc
As such, the term neuropathic pain represents a vary- Stroke (brain infarction, spinal infarction)
ing set of symptoms rather than a single diagnosis. Multiple sclerosis
Damage to the somatosensory system can provoke Parkinson’s disease
a range of responses; an absence of sensation and pain Surgical lesions (such as rhizotomy, cordotomy)
is probably a more common response than new onset Complex neuropathic disorders
of pain. Sensitisation of peripheral nerves with ectopic Complex regional pain syndrome types I and II

or spontaneous activity and hyperexcitability in the

Summary points modulatory pathways of the central nervous system
Neuropathic pain commonly presents in primary care and is
are possible mechanisms that could cause neuropathic
often unrecognised
Diagnosis is based on characteristic symptoms,
pain.2 3
altered sensation, and a clinical history that matches a Neuropathic pain can be caused by several different
neuroanatomical or dermatomal pattern disease processes, which can often overlap, but cur-
Less than half of patients achieve significant benefit with any rently there is no universally accepted classification for
single drug the disorder. However, four broad classes of diseases
Management includes making pain tolerable and are recognised based on aetiology and anatomy2: focal
maintaining emotional and physical functioning and multifocal lesions of the peripheral nervous sys-
Non-pharmacological approaches can be effective, but tem; generalised polyneuropathies of the peripheral
referral for specialist help is indicated if pain persists or
nervous system; lesions in the central nervous system;
remains uncontrolled
and complex neuropathic disorders (box 1).

BMJ | 15 august 2009 | Volume 339 391

CLINICAL REVIEW

painful or painful stimuli. In the clinic patients can com-

Box 2 | Common side effects of first line drugs for
neuropathic pain plain of dysaesthesias (unpleasant and strange sensations
in the skin, such as tingling and pins and needles), deep
Tricyclic antidepressants (amitriptyline, nortriptyline,
desipramine) seated gnawing pain, and abnormal thermal sensations
Drowsiness (burning, on fire, or ice cold). Less commonly, patients
Confusion complain of paroxysmal pains classically described as
Dry mouth shooting, stabbing, or electric shocks. Patients may com-
Orthostatic hypotension plain that the painful area is abnormally sensitive to any
Weight gain innocuous mechanical or thermal stimulus; sometimes
Urinary retention clothes brushing against the area or a cold draught of air
Screening electrocardiography is recommended before can be intensely painful.
beginning treatment in patients over 40 years Features of neuropathic pain may be evident within
α2-δ anticonvulsants (gabapentin and pregabalin) days of nerve damage or can take months to develop.
Drowsiness Sometimes, a minor insult to an area of previous injury
Dizziness, cognitive or gait impairment that had healed without problem can trigger neuropathic
Peripheral oedema pain.
Serotonin and norepinephrine reuptake inhibitors
(duloxetine, venlafaxine) How is neuropathic pain diagnosed?
Nausea
There is no standard diagnostic procedure for neuro-
Dizziness
pathic pain, so making a diagnosis is based on clinical
Dry mouth
judgment. The essential elements of this process are
Sexual dysfunction
to identify painful symptoms, altered sensation, and
Topical 5% lidocaine patch
Mild skin reactions (such as redness or swelling under
a clinical history that all match a neuroanatomical or
patch, erythema) dermatomal pattern.10 Screening methods for neu-
ropathic pain consist mostly of characteristic verbal
descriptors, though some have simple bedside tests in
How common is neuropathic pain? addition. Examples of the latter are the Leeds assess-
Estimates of neuropathic pain associated with specific ment of neuropathic symptoms and signs (LANSS)
aetiologies are well described. A 2008 study reported pain scale11 or the painDETECT questionnaire,5 which
age standardised incidence rates of 27.3 per 100 000 have an approximate accuracy of 80% compared with
person years for post-herpetic neuralgia, 26.7 for trigem- expert clinical judgment in identifying patients with
inal neuralgia, 26.7 for painful diabetic ­neuropathy, neuropathic pain. Screening methods, however, are not
and 0.8 for phantom limb pain.4 The prevalence of a substitute for good clinical assessment and are not
­neuropathic pain typically rises with age and severity intended to be diagnostic methods.12
of the ­underlying condition.5 6 Bedside examination is straightforward; the aim is
By contrast, the overall prevalence of neuropathic to identify altered sensation in the painful area, and so
pain in the general population is difficult to quantify, responses should be compared with a non-painful adja-
and exact data are lacking because of the large number cent or contralateral area. A painful response to lightly
of underlying causes and the lack of standardised meas- stroking the skin with a finger or cotton wool is a sign of
urement methods. Nevertheless, epidemiological sur- allodynia (pain caused by a stimulus that does not usu-
veys suggest that 6-8% of the general population report ally provoke pain), a common characteristic of neuro-
chronic pain with neuropathic characteristics.7 8 Neuro- pathic pain. Numbness (hypoalgesia) or an exaggerated
pathic back pain with radiating pain to the arm or leg, painful response (hyperalgesia) to pin prick testing with
and post-traumatic neuropathic pain (from accidental or
surgical injury) are probably the most common causes. A patient’s perspective—neuropathic facial pain resulting
Among every 1000 patients registered with a general from oropharyngeal cancer
practitioner, about 60-80 patients will have symptoms As far as I can remember my pain started during the second
of chronic neuropathic pain; in half of these patients, week of the radiotherapy and chemotherapy. In total I had
pain will require medication and regular support.9 six chemotherapy and 30 radiotherapy sessions over a
six week period. I found the radiotherapy was the most
painful. Not during treatment—it was the after effects. The
How does neuropathic pain present in clinical practice?
headache seemed to get worse on a daily basis.
Abnormal responses to nerve damage can account for
Trying to explain the pain is difficult. It is like hundreds of
many of the clinical characteristics of neuropathic pain.
needles inside my head. I ended up trying to relate it to
Symptoms can be unusual and are unlike more com- other pain I have suffered over my lifetime. For instance,
mon pain symptoms that patients will have previously ear infection at its worst, very bad migraine, tonsillitis. If
experienced (see the box “A patient’s perspective”). you could imagine all this pain in one blast it is about right,
Painful symptoms arising in an area of altered sensa- maybe even worse. At this time of my life I was lucky if I
tion (numbness or hyperexcitability) are the hallmark of managed to have an hour or two of undisturbed sleep. Now
neuropathic pain. 10 months down the line, I am finding it much easier to cope
Cardinal symptoms can be spontaneous pain (pain as my nerve ends are healing. The medication as a pain patch
has definitely been a great help with the nerve pain.
arising without stimulus) and abnormal responses to non-

392 BMJ | 15 august 2009 | Volume 339

 CLINICAL REVIEW

a monofilament or sharp object confirms an altered pin Tips for non-specialists

prick threshold. Finally, inability to distinguish warm • Make sure that the chronic pain type has been diagnosed
from cold objects suggests an altered thermal threshold. correctly: a neuropathic pain screening tool may help
A combination of characteristic painful symptoms in an (such as painDETECT, the LANSS pain scale)
area of altered sensation on bedside testing is usually • In general, start with monotherapy, but consider
enough to make a diagnosis of neuropathic pain. When combination therapy from early on (for example, start with
there is doubt, more detailed examination using quan- amitriptyline or duloxetine but then add gabapentin or
titative sensory testing is helpful though not commonly pregabalin, or vice versa)
available, and it is costly and time consuming.13 • Provide adequate time for any drug trial (two to eight
weeks; at least one to two weeks at maximum tolerated
Patients usually present with a spectrum of features
dosage)
(mixed pain) and it is more helpful to ask yourself the
• Address physical and emotional aspects, as well as pain,
question “does this pain have any neuropathic compo- by encouraging physical activity, improving poor sleep,
nent?”14 It is important to identify neuropathic compo- and treating depression and anxiety
nents because different classes of analgesic drugs are • A multidisciplinary approach to treatment is often the
usually required to manage this type of pain effectively. most successful; this includes seeking specialist help if
Recently, criteria for categorising neuropathic pain as your patient is not improving despite initial treatment
definite, probable, or possible have been proposed; this
would better indicate the reality of clinical practice than
a simple “nociceptive versus neuropathic” dichotomy.1 back pain in Germany estimated the annual cost to be
about €7.6bn (£6.4bn; $10.9bn).17
What are the personal and societal costs of
neuropathic pain? What is a sensible therapeutic approach for patients
Patients with chronic neuropathic pain usually report with neuropathic pain?
poorer physical and mental health compared with Combination therapy with drugs that result in additive
patients with other types of chronic pain, even when or synergistic effects to target different pain mecha-
adjusting for pain intensity.5 15 16 This association with nisms is a logical approach because less than half of
poor physical and mental health suggests that the all patients will achieve a significant benefit with any
nature, and not simply the intensity, of neuropathic single medical treatment for neuropathic pain.18 The
pain adversely affects quality of life and that successful evidence, however, to support the idea that combina-
management requires more than drug treatment alone. tion therapy is likely to be more efficacious and safer
The cost of neuropathic pain to society is not known, than each drug alone is less developed than that for
but recent economic modelling of neuropathic low single drugs (see the box “Tips for non-specialists”).18

Table 1 | Treatment recommendations for peripheral neuropathic pain adapted from recent guidelines and algorithms19-22
Dose range (mg/day) for Combined NNH for study withdrawal Combined NNT for 50% pain relief
Medication class/drug Recommended stage of treatment maintenance (range) (range)
Antidepressants
Tricyclics (nortriptyline, desipramine, First 25-150; secondary amine tricyclic 14.7 (10.2-25.2) 2.1/2.5/3.1 (1/8-3.7)
amitriptyline, imipramine) antidepressants are in favour
(nortriptyline, desipramine)
Duloxetine First or second 60-120 Relative risk not significant 4.1/5.2 (2.9-8.5)
Venlafaxine First or second 150-225 Relative risk not significant 4.6 (2.9-10.6)
Paroxetine, citalopram, bupropion Third Relative risk not significant 6.8 (3.4-441)
Anticonvulsants
Pregabalin First 150-600 11.7 (8.3-19.9) 4.2/4.9 (3.7-7.6)
Gabapentin First 1200-3600 17.8 (12-30) 4/4,4 (3.3-6.1)
Carbamazepine First (only for trigeminal neuralgia) 200-1200 21.7 (12.6-78.5) 2.0 (1.3-2.2)
Lamotrigine Second or third 200-400 (slow titration) Relative risk not significant 4.9 (3.5-8.1)
Oxcarbazepine Second (only for trigeminal 600-1800 (fewer safety concerns) Relative risk not significant NA
neuralgia)
Topiramate Third 200-400 6.3 (5-8) 7.4 (4.3-28)
Valproate Third 1000 Relative risk not significant 2.8 (2.1-4.2)
Opioids*
Oxycodone Second or third 10-120 Relative risk not significant 2.6 (1.9-4.1)
Morphine Second or third 15-300 Relative risk not significant 2.5 (1.9-3.4)
Tramadol Second or third 200-400 9 (6.0-17.5) 3.9/4.8 (2.6-26.9)
Methadone Second or third 15 NA NA
Miscellaneous
Topical lidocaine (patch 5%; gel) First or second (only for localised 1-3 patches/day applied for 12 h Relative risk not significant 4.4 (2.5-17.5)
areas of pain, focal neuropathy,
allodynia)
Cannabinoids Third 5-15 Relative risk not significant 9.5 (4.1-∞)
Topical capsaicin Third 11.5 (8.1-19.8) 6.7 (4.6-12)
NNH=number needed to harm on the basis of withdrawal from neuropathic pain studies owing to adverse effects.
NNT=number needed to treat on the basis of 50% pain relief from baseline.
*The combined NNH for study withdrawal (range) for opioids overall = 17.1 (10-66).

BMJ | 15 august 2009 | Volume 339 393

 CLINICAL REVIEW

Table 2 | Treatment recommendations for central neuropathic pain adapted from current evidence tine and venlafaxine)20‑22 are recommended as either
based literature19-22 26 first or second line treatment, as is the use of topical 5%
Medication class/drug Recommended stage of treatment lidocaine patches (for localised small areas of pain).19‑22
Antidepressants Meta-analyses show that tricyclic antidepressants are
Tricyclics (amitriptyline) First or second more efficacious and more cost effective than serotonin
Serotonin and norepinephrine reuptake inhibitors (duloxetine, First or second and norepinephrine reuptake inhibitors, although the
venlafaxine) latter have a better side effect profile and may therefore
Anticonvulsants
be more suitable for elderly patients or those with car-
Pregabalin First or second
diac disease (box 2).19 As new studies on serotonin and
Gabapentin First or second
Lamotrigine Second or third (in pain after stroke)
norepinephrine reuptake inhibitors emerge, these drugs
Valproate Third
may soon replace the use of tricyclic antidepressants.
Opioids* Topical lidocaine patches can be used preferentially for
Levorphanol localised small areas of peripheral neuropathic pain with
Miscellaneous mechanical allodynia (such as in post-herpetic neuralgia)
Cannabinoids Second (in multiple sclerosis) or for focal neuropathy, but not for patients with central
Mexiletine Third neuropathic pain. As systemic side effects are extremely
*Second or third (no specification). rare with topical treatments, they are safe particularly for
elderly patients. Data about impact on quality of life and
What are the recommendations from clinical guidelines comorbidities are only available for newer drugs such as
and algorithms? gabapentin, ­pregabalin, and duloxetine, which have all
Unfortunately, there is no absolute consistency from the shown positive effects.9 16 19
four most recently published guidelines and algorithms
regarding the assessment of first, second, and third line Second line approaches
drug treatments for neuropathic pain, although consist- Randomised controlled trials have shown that controlled
ency exists for the classes of medication.19‑22 There are release opioid analgesics, such as ­morphine, oxycodone,
several limitations of this evidence. The evidence is and tramadol, are effective in neuropathic pain.18‑22 There
focused on a limited range of peripheral neuropathic is no evidence that one opioid is any more effective than
pain disorders (in particular, post-herpetic neuralgia another, or less effective than other drug classes for neu-
and painful diabetic peripheral neuropathy) rather than ropathic pain, however, due to potential safety concerns
­central neuropathic pain (box 1); past studies have usu- such as tolerance, addiction, cognitive impairment, these
ally examined monotherapy rather than sequential or drugs are usually recommended as second or third-line
parallel combinations; priorities in guidelines are not treatments.
derived from head to head studies but from comparisons Opioids can be considered as a first-line approach in
of separate and heterogenous controlled trials; and the selected clinical circumstances, such as intractable pain,
duration of studies is usually limited to six to eight weeks, episodic exacerbations of severe pain, acute neuropathic
which tells us little about managing chronic neuropathic pain, and neuropathic cancer pain.20 The treatment of
pain that lasts for many years. Table 1 shows data for trigeminal neuralgia has distinct recommendations: car-
different classes of treatments. bamazepine (stronger evidence) or oxcarbazepine (better
tolerability) should be offered as first-line treatment for
First line approaches patients with this disorder, but are otherwise not rec-
A sensible first line approach is to use an anticonvulsant ommended for the management of neuropathic pain.23
that binds on the α2-δ subunit of presynaptic, voltage- There are encouraging results from studies regarding
gated calcium channels (gabapentin or pregabalin) cannabinoids, but their future role in the treatment of
or a tricyclic antidepressant. Comparisons of tricyclic neuropathic pain has still to be determined.24
antidepressants did not show any significant difference
between them, but tertiary amine tricyclic antidepres- Third line approaches
sants (amitriptyline, imipramine, clomipramine) are usu- Even with well established drugs, effectiveness is
ally associated with more severe side effects than with ­unpredictable, dosing can be complicated, analgesic
other tricyclic antidepressants, and secondary amine tri- onset is delayed, and dose limiting adverse effects are
cyclic antidepressants (nortriptyline and desipramine) common. So patients who are refractory to any first or
are better tolerated (box 2). The newer mixed serotonin second line treatment may benefit from other drugs, but
and norepinephrine reuptake inhibitors (such as duloxe- recommendations are based on much weaker evidence
or only expert clinical opinion. There is insufficient sup-
Sources and selection criteria port for the use of non-steroidal anti-inflammatory drugs
We searched PubMed and Medline for articles whose titles included the keywords “neuropathic such as aspirin, diclofenac, naproxen, or ibuprofen.
pain” with the limits “meta-analysis, review and randomised controlled trial”, and we restricted
the search to articles published in English in the previous five years. We individually reviewed Central neuropathic pain
the titles of the resulting articles to identify major themes. We identified all systematic reviews The treatment of central pain is even more controversial.
in the Cochrane Controlled Trials Register with the same terms. Finally we checked recent The evidence base is limited, and responders typically
recommendations and published clinical guidelines from different international pain associations
experience only partial pain relief at tolerable doses.
and societies and reviewed contemporary textbooks and personal files.
Based on recent evidence,25 26 the first line approach that

394 BMJ | 15 august 2009 | Volume 339

 CLINICAL REVIEW

1 Treede RD, Jensen TS, Campbell JN, Cruccu G, Dostrovsky JO, Griffin JW, et
Additional educational resources al. Neuropathic pain: redefinition and a grading system for clinical and
• Special Interest Group on Neuropathic Pain of the IASP (NeuPSIG) (www.neupsig.org)—Leading research purposes. Neurology 2008;70:1630-5.
2 Baron R. Mechanisms of disease: neuropathic pain: a clinical
professional forum for science, practice, and education in the field of neuropathic pain; news, perspective. Nat Clin Pract Neurol 2006;2:95-106.
guidelines on diagnosis and treatment; international educational meetings and symposiums 3 Scholz J, Woolf CJ. The neuropathic pain triad: neurons, immune cells
• German Research Network on Neuropathic Pain (DFNS) (www.neuro.med.tu-muenchen.de/ and glia. Nat Neurosci 2007;10:1361-8.
4 Hall GC, Carroll D, McQuay H. Primary care incidence and treatment of
dfns/e_index.html)—Provides excellent information about neuropathic pain and ongoing four neuropathic pain conditions: a descriptive study, 2002-2005. BMC
research (the central Integrative Network Project of the DFNS is a large neuropathic pain data Fam Pract 2008;9:26.
bank funded by the German Federal Ministry of Education and Research) 5 Freynhagen R, Baron R, Gockel U, Tölle TR. painDETECT: a new screening
• Neuropathic Pain Network (NPN) (www.neuropathicpainnetwork.org/english/coalition/index. questionnaire to identify neuropathic components in patients with back
pain. Curr Med Res Opin 2006;22:1911-20.
asp)—Coalition of organisations and patient support groups actively supporting people with 6 Gustorff B, Dorner T, Likar R, Grisold W, Lawrence K, Schwarz F, et al.
neuropathic pain Prevalence of self-reported neuropathic pain and impact on quality
• International Association for the Study of Pain (IASP) (www.iasp-pain.org)—Large of life: a prospective representative survey. Acta Anaesthesiol Scand
2008;52:132-6.
multidisciplinary organisation focused specifically on pain research and treatment 7 Torrance N, Smith BH, Bennett MI, Lee AJ. The epidemiology of chronic
• Oxford Pain Internet Site (Bandolier) (www.medicine.ox.ac.uk/bandolier/booth/ pain of predominantly neuropathic origin. Results from a general
painpag/#Chronic)—Based on the principles of evidence based medicine, this site has pulled population study. J Pain 2006;7:281-9.
8 Bouhassira D, Lanteri-Minet M, Attal N, Laurent B, Touboul C. Prevalence
together systematic reviews with pain as an outcome of chronic pain with neuropathic characteristics in the general
population. Pain 2008;136:380-7.
9 McDermott AM, Toelle TR, Rowbotham DJ, Schaefer CP, Dukes EM. The
is recommended for peripheral neuropathic pain condi- burden of neuropathic pain: results from a cross-sectional survey. Eur J
tions can be used for central pain; α2-δ anticonvulsants, Pain 2006;10:127-35.
tricyclic antidepressants, and serotonin and norepine- 10 Cruccu G, Anand P, Attal N, Garcia-Larrea L, Haanpää M, Jørum E, et
al. EFNS guidelines on neuropathic pain assessment. Eur J Neurol
phrine reuptake inhibitors are considered to be first or 2004;11:153-62.
second choice. Table 2 gives data for therapeutic options 11 Bennett M. The LANSS pain scale: the Leeds assessment of neuropathic
symptoms and signs. Pain 2001;92:147-57.
in detail. 12 Bennett MI, Attal N, Backonja MM, Baron R, Bouhassira D, Freynhagen R,
et al. Using screening tools to identify neuropathic pain (topical review).
Pain 2007;127:199-203.
Is there a role for non-pharmacological treatments? 13 Rolke R, Baron R, Maier C, Tölle TR, Treede RD, Beyer A, Binder A, et
Surgical and chemical sympathectomy have been used al. Quantitative sensory testing in the German Research Network on
Neuropathic Pain (DFNS): standardized protocol and reference values.
to treat neuropathic pain, but the evidence base is weak Pain 2006;123:231-43.
and complications may be substantial.27 Microvascular 14 Bennett MI, Smith BH, Torrance N, Lee AJ. Can pain be more or less
decompression is a recognised treatment for some types neuropathic? Comparison of symptom assessment tools with ratings of
certainty by clinicians. Pain 2006;122:289-94.
of trigeminal neuralgia. The use of neurodestructive pro- 15 Smith BH, Torrance N, Bennett MI, Lee A. Health and quality of life
cedures to treat painful disorders that are due to nerve associated with chronic pain of predominantly neuropathic origin in the
community. Clin J Pain 2007;23:143-9.
damage is misguided and can lead to even more pain. 16 Jensen MP, Chodroff MJ, Dworkin RH. The impact of neuropathic pain
Recent systematic reviews have shown that neuro­ on health-related quality of life: review and implications. Neurol
2007;68:1178-82.
stimulation, including the use of transcutaneous elec- 17 Schmidt CO, Schweikert B, Wenig CM, Schmidt U, Gockel U, Freynhagen
trical nerve stimulation, electroacupuncture, and R, et al. Modelling the prevalence and cost of back pain with neuropathic
components in the general population. Eur J Pain 2009; Feb 5. (Epub
repetitive transcranial magnetic stimulation, were all ahead of print)
found to be better than placebo, though these recom- 18 Gilron I, Max MB. Combination pharmacotherapy for neuropathic
mendations were not based on high grade evidence.28 29 pain: current evidence and future directions. Expert Rev Neurother
2005;5:823-30.
The evidence for spinal cord stimulation supports its 19 Finnerup NB, Otto M, McQuay HJ, Jensen TS, Sindrup SH. Algorithm
use in patients with refractory neuropathic back and for neuropathic pain treatment: an evidence based proposal. Pain
2005;118:289-305.
leg pain (particularly in failed back surgery syndrome) 20 Dworkin RH, O’Connor AB, Backonja M, Farrar JT, Finnerup NB, Jensen
and complex regional pain syndrome.30 Although the TS. Pharmacologic management of neuropathic pain: evidence-based
recommendations. Pain 2007;132:237-51.
role of traditional acupuncture in neuropathic pain is 21 Moulin DE, Clark AJ, Gilron I, Ware MA, Watson CP, Sessle BJ, et al.
not supported by current evidence, it is popular among Pharmacological management of chronic neuropathic pain—consensus
statement and guidelines from the Canadian Pain Society. Pain Res
patients, and as it is relatively harmless it is often used Manag 2007;12:13-21.
in addition to drug treatment. 22 Attal N, Cruccu G, Haanpää M, Hansson P, Jensen TS, Nurmikko T, et al.
EFNS guidelines on pharmacological treatment of neuropathic pain. Eur J
Neurol 2006;13:1153-69.
Questions for future research 23 Cruccu G, Gronseth G, Alksne J, Argoff C, Brainin M, Burchiel K, et al.
Important areas for future research include develop- AAN-EFNS guidelines on trigeminal neuralgia management. Eur J Neurol
2008;15:1013-28.
ing a specific diagnostic method for neuropathic pain; 24 Iskedjian M, Bereza B, Gordon A, Piwko C, Einarson TR. Meta-analysis
identifying associations between symptoms, signs, and of cannabis based treatments for neuropathic and multiple sclerosis-
related pain. Curr Med Res Opin 2007;23:17-24.
pathology to guide mechanism based treatment strate- 25 Finnerup NB. A review of central neuropathic pain states. Curr Opin
gies; comparing combination treatments with mono- Anaesthesiol 2008;21:586-9.
26 Tzellos TG, Papazisis G, Amaniti E, Kouvelas D. Efficacy of pregabalin
therapy; and conducting pharmacogenomic studies to and gabapentin for neuropathic pain in spinal-cord injury: an evidence-
guide prescribing. based evaluation of the literature. Eur J Clin Pharmacol 2008;64:851-8.
Contributors: RF and MIB contributed equally to the article. Both authors 27 Mailis A, Furlan A. Sympathectomy for neuropathic pain. Cochrane
Database Syst Rev 2003;(2):CD002918.
read and approved the final manuscript and are guarantors. 28 Cruccu G, Aziz TZ, Garcia-Larrea L, Hansson P, Jensen TS, Lefaucheur J-P,
Competing interests: In the past three years RF has received research et al. EFNS guidelines on neurostimulation therapy for neuropathic pain.
support and consulting or speaking fees from Grünenthal, Janssen-Cilag, Eur J Neurol 2007;14:952-70.
Lilly/Boehringer, Mundipharma, Organon, Pfizer, and Schwarz Pharma. MB 29 Pittler MH, Ernst E. Complementary therapies for neuropathic and
neuralgic pain: systematic review. Clin J Pain 2008;24:731-3.
has received honorariums, consultancy fees, and research grants from Pfizer, 30 Boswell MV, Trescot AM, Datta S, Schultz DM, Hansen HC, Abdi S, et al.
NAPP, and Cephalon. Interventional techniques: evidence-based practice guidelines in the
Provenance and peer review: Commissioned; externally peer reviewed. management of chronic spinal pain. Pain Physician 2007;10:7-111.

BMJ | 15 august 2009 | Volume 339 395