Article Text

Download PDFPDF
An update on the drug treatment of neuropathic pain. Part 2: antiepileptics and other drugs


In An update on the drug treatment of neuropathic pain. Part 1: antidepressants,1 we highlighted the problems associated with the management of neuropathic pain and reviewed the use of antidepressants. This month we discuss the use of antiepileptic drugs and other analgesics.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Antiepileptic drugs

Antiepileptic drugs have been used in the treatment of neuropathic pain since the 1960s.2 The mechanism by which they exert their analgesic effect in neuropathic pain is uncertain, although it is probably related to a reduction in the high-frequency firing of neurones.24

Gabapentin and pregabalin

Gabapentin and pregabalin have analgesic, anticonvulsant and anxiolytic properties, and are licensed in the UK for the treatment of neuropathic pain.5,6


Gabapentin was investigated in a systematic review of 29 studies (involving 3,571 patients) in the treatment of chronic pain of various causes; 78% of the patients were suffering from post-herpetic neuralgia, painful diabetic neuropathy, or mixed neuropathic pain.7 According to the Initiative on Methods, Measurement and Pain Assessment in Clinical Trials (IMMPACT) definition of at least moderate benefit (see Box 1),1 gabapentin was more effective than placebo in 14 of these studies (2,831 participants), with 43% of the participants improving with gabapentin compared with 26% improving with placebo. Using the IMMPACT definition of substantial benefit, gabapentin was more effective than placebo in 13 studies (2,627 participants), with 31% of participants improving with gabapentin compared with 17% improving with placebo. The number needed to treat (NNT) to achieve this benefit was 7 (95% CI 6 to 9).7

Box 1: IMMPACT subjective measures of improvement in pain:1

Assessment using a 0–10 scale:

≥30% pain relief over baseline = moderate improvement

≥50% pain relief over baseline = substantial improvement

Patient Global Impression of Change:

Patients rate improvement with treatment during a clinical trial on a 7-point scale ranging from ‘very much improved’ to ‘very much worse’, with ‘no change’ as the mid-point. ‘Much improved’ is regarded as a moderate benefit and ‘very much improved’ as substantial benefit.

A systematic review of pregabalin2 identified 19 placebo-controlled trials (7,003 patients) of its use in various conditions, including post-herpetic neuralgia, painful diabetic neuropathy, and central neuropathic pain, but also fibromyalgia (where the pain is not always neuropathic).8 At doses of 300mg/day, 450mg/day and 600mg/day, pregabalin was found to be an effective analgesic compared with placebo. However, a dose of 150mg/day was found to be ineffective. The NNT for substantial benefit over baseline for 600mg/day of pregabalin compared with placebo was 4 (95% CI 3 to 5) for post-herpetic neuralgia and 5 (95% CI 4 to 7) for painful diabetic neuropathy.2

Direct head-to-head comparisons between gabapentin and pregabalin are lacking.

Unwanted effects

The unwanted effects caused by gabapentin and pregabalin are similar. Both cause sedation, which may be helpful for some patients if sleep is a problem; beneficial effects on anxiety may similarly occur in some patients. However, sedation is also the most common dose-limiting unwanted effect; it can be minimised by starting with a low dose and increasing gradually.3

The systematic review of gabapentin found 17 studies (3,022 patients) that reported “withdrawals resulting from adverse events”.7 Such withdrawals occurred in 12% of patients on gabapentin (at doses ≥1200mg/day) and in 8% of patients on placebo; (RR 1.4, 95% CI 1.1 to 1.7). The number needed to harm (NNH) for gabapentin was 32 (95% CI 19 to 100). Serious adverse events (4%) were no more common than with placebo.

The same review found 11 studies (2,356 patients) that reported patients experiencing at least one unwanted effect.7 Adverse events occurred significantly more often in patients taking gabapentin than in those on placebo, with 66% in the gabapentin groups having at least one adverse event. The most common were dizziness (experienced by 21%), somnolence (16%), peripheral oedema (8%), and gait disturbance (9%). There were insufficient data for comparisons with other active treatments.

The systematic review of pregabalin found that at 600mg/day, daily somnolence typically occurred in 15–25% of patients, dizziness in 27–46%, and treatment withdrawal in 18–28% of patients.2 The percentage of patients who experienced at least one unwanted effect was not influenced by dose. The incidence of serious adverse events was similarly not affected by dose and was not significantly different from those taking placebo. In the trials conducted in patients with painful diabetic neuropathy, the NNH for withdrawal from treatment was 9 (95% CI 7 to 12) with 600mg/day and 16 (95% CI 9.9 to 37) with 300mg/day.

Guidelines and clinical use

The National Institute for Health and Clinical Excellence (NICE) guideline considers that there is moderate- to high-quality evidence for the use of gabapentin and pregabalin, which have both shown efficacy across various study populations and in various conditions causing neuropathic pain.9 However, the guideline recommends that pregabalin is a better treatment option than gabapentin because it has a lower NNT than gabapentin, a simpler dosing and titration regimen (see Box 2), the adverse-effect profiles of the two agents are similar, and health economic analysis suggests that pregabalin is more cost-effective than gabapentin. If adequate results are not obtained with pregabalin, the recommendation is that another drug class (generally an antidepressant) should be tried, rather than another antiepileptic drug.

Box 2: Gabapentin and pregabalin dosing instructions5,6,10

Gabapentin is initiated at a dose of 300mg on day one, 300mg twice a day on day two and 300mg three times a day on day three. Alternatively, the starting dose is 900mg/day given as three equally divided doses. Thereafter, based on individual patient response and tolerability, the dose can be further increased in 300mg/day increments every 2–3 days up to a maximum dose of 3,600mg/day. The minimum time to reach a dose of 1,800mg/day is 1 week, to reach 2,400mg/day is a total of 2 weeks, and to reach 3,600mg/day is a total of 3 weeks.

Pregabalin can be started at a dose of 150mg per day given as two or three divided doses. Based on individual patient response and tolerability, the dose may be increased to 300mg per day after an interval of 3–7 days, and if needed, to a maximum dose of 600mg per day after an additional 7-day interval.

In 2010, DTB highlighted the financial implications to the NHS resulting from NICE's recommendation on the use of pregabalin for neuropathic pain. In an editorial we questioned whether the data underpinning the guidance were “clear-cut and robust and the assumptions reasonable”.11 We noted that “for the same level of investment, the NHS faces the choice of treating more patients with a cheaper but (on indirect evidence) slightly less effective drug, or fewer patients with a more expensive, seemingly more effective one”. In addition, the original health technology appraisal containing the health economic analysis has not been accepted for publication and is not available on the HTA website*. NICE has recently announced that it is consulting on the guidance and has published a statement on its website (see Box 3).

Box 3: NICE statement on pregabalin9

NICE is aware that there have been concerns about the associated costs that pregabalin may bring to the NHS as one of the first-line treatment options for adults with neuropathic pain. Therefore, NICE will fully update its clinical guideline on the pharmacological management of neuropathic pain in non-specialist settings in order to address ongoing uncertainties regarding the cost-effectiveness of some of the recommended treatment options. There are no known concerns regarding the safety or clinical efficacy of the recommended drugs.

The European Federation of Neurological Societies (EFNS) guidelines place both gabapentin and pregabalin as first-line treatment options for diabetic peripheral neuropathy, post-herpetic neuralgia and central neuropathic pain.12 The International Association for the Study of Pain (IASP) guidelines also recommend both gabapentin and pregabalin as first-line therapies.13


Historically, the most commonly used antiepileptic drug for neuropathic pain has been carbamazepine.9 Although the evidence is limited, it continues to be widely used in the management of trigeminal neuralgia, for which use it is licensed in the UK.


A systematic review of carbamazepine in the treatment of acute and chronic pain identified 15 studies involving 629 patients.4 Of these studies, 14 investigated chronic neuropathic pain, although five were noted to have poor reporting and 10 involved fewer than 50 patients. Outcome reporting was inconsistent.

In eight of the studies, patients with trigeminal neuralgia made up part or all of the subject group. However, many of the studies are old (five published in the 1960s, and only one study was published in the past 10 years) and some were very small. Efficacy for carbamazepine was claimed in all of these studies. An analysis of the two randomised, controlled studies that compared carbamazepine versus placebo (98 patients) showed a relative benefit of the drug compared with placebo of 6.0 (95% CI 2.8 to 13) and NNT of 2 (95% CI 1 to 2).

Unwanted effects

The most common unwanted effects of carbamazepine are impaired cognition, drowsiness and ataxia. These can be especially troublesome in older patients. Blood dyscrasias and drug rashes also occur relatively frequently.4 The systematic review found that 66% of patients taking carbamazepine experienced at least one adverse event (compared with 27% of the placebo groups) and that the NNH was 3 (95% CI 2 to 4).4 Withdrawals due to adverse event occurred in 4% of patients taking carbamazepine. Serious adverse events were not reported consistently and could not be evaluated.

Guidelines and clinical use

The NICE guideline notes that the evidence for the use of carbamazepine in the treatment of neuropathic pain is very limited and dated, and therefore it is not recommended for general use in the treatment of neuropathic pain.9 However, it is also acknowledged that this drug has been the routine treatment for trigeminal neuralgia in clinical practice since the 1960s and that clinical experience and anecdotal evidence suggest that it may be effective in this condition. Moreover, trigeminal neuralgia is an extremely painful condition for which there is as yet no good-quality evidence to inform more specific treatment recommendations for analgesia. Therefore, the guidance states that carbamazepine may have a role in treating trigeminal neuralgia and that it is to be expected that current clinical practice will continue.

The EFNS guidelines recommend carbamazepine as a first-line treatment for trigeminal neuralgia.12


It has been widely believed that neuropathic pain is unresponsive to opioid analgesia. However, evidence has grown that opioids can be effective against neuropathic pain in some patients, and their use in this indication has increased over recent years.14


A Cochrane review of opioids in neuropathic pain identified 23 trials, all of which were short term (with 14 trials being for less than 24 hours).15 These studies gave mixed results with respect to analgesic efficacy, although cross-study quantitative analysis of data was difficult owing to heterogeneity of the trials, including variable measurements of pain outcome. Other reviews of the literature have similarly found that studies of short-term use of opioids in neuropathic pain provided only equivocal evidence of efficacy,14,16 but one of these concluded that opioids can reduce the intensity of dynamic mechanical allodynia and perhaps of cold-induced allodynia.14

The other nine trials assessed in the Cochrane review were of longer duration (8–70 days) and involved 460 patients with various types of neuropathic pain, including post-herpetic neuralgia, painful diabetic neuropathy and phantom limb pain.15 All of these trials showed efficacy of opioids for pain relief in spontaneous neuropathic pain. Meta-analysis of seven of these trials, which had suitable data, revealed a mean reduction in pain intensity after treatment of 13 on a visual scale from zero to 100. The review noted that this 13-point reduction compared “not unfavourably” with other commonly used agents for neuropathic pain. In general, the studies used fixed, low to moderate doses of opioids. The agents studied in these nine trials included morphine, oxycodone and methadone.15

Another systematic review of opioids for chronic, non-cancer pain identified 15 randomised placebo-controlled trials, 11 of which (involving 1,025 patients) compared oral opioids with placebo for periods ranging from 4 days to 8 weeks.17 The short-term efficacy of opioids was found to be good in both neuropathic and musculoskeletal pain. The mean decrease in pain intensity in most of the 15 studies in the review was at least 30% with opioids and was similar for neuropathic pain and musculoskeletal pain. However, only a minority of patients in these studies continued on to long-term management with opioids, and the small number of selected patients and the short follow-up periods preclude conclusions about problems such as tolerance and addiction with long-term use. ref-type="bibr" rid="DTB139C17">17

Unwanted effects

The unwanted effects of opioids are well known, and they are often poorly tolerated. The most common unwanted effects in the nine studies discussed above were nausea, constipation, sedation, dizziness and vomiting. The NNH for discontinuation of opioid treatment due to unwanted effects was 17 (95% CI 9 to 100). ref-type="bibr" rid="DTB139C15">15 Constipation in particular is likely to be an ongoing problem that does not diminish with long-term use, and a suitable laxative regimen may be required. ref-type="bibr" rid="DTB139C18">18 Opioids can also increase confusion and cognitive problems and the risk of falls in older patients. ref-type="bibr" rid="DTB139C13">13

Opioids carry the possibility of dependence in long-term use, the risk being up to 50% in studies of non-neuropathic chronic pain. ref-type="bibr" rid="DTB139C13">13 Those with a personal or family history of substance abuse and those with a major psychiatric disorder are at particular risk. The possibility of dependence, and its management if it occurs, are therefore issues that may need to be addressed in patients using opioids for long-term relief of neuropathic pain.

Guidelines and clinical use

Short-acting preparations have a quick onset of action, and in clinical practice they are often used to provide relief for short-term exacerbations or during the titration phase of another medication before the maintenance dose is reached. ref-type="bibr" rid="DTB139C13">13

The NICE clinical guideline acknowledges that the evidence for the use and efficacy of opioids such as morphine and oxycodone is limited, that the likelihood of withdrawal from treatment because of unwanted effects is high, and that there are concerns about dependence in long-term use. ref-type="bibr" rid="DTB139C9">9 The guideline discusses the lack of experience of many clinicians in non-specialist settings in prescribing and managing long-term opioid use. The guideline recommends that morphine and oxycodone should not be started in non-specialist settings without an assessment by a specialist pain service or a condition-specific service, although they suggest that treatment can be safely continued in a non-specialist setting provided that there is a multidisciplinary care plan, local shared care agreements and careful management of adverse effects.

The EFNS guidelines recommend opioids as a second- or third-line treatment for painful diabetic neuropathy, post-herpetic neuralgia and central neuropathic pain. ref-type="bibr" rid="DTB139C12">12

The IASP guidelines recommend opioids as second-line treatments when first-line therapy has not produced an adequate response, and as first-line treatments when immediate pain relief is required and as short-term therapy for acute neuropathic pain. ref-type="bibr" rid="DTB139C13">13


Tramadol is an agonist at opioid mu receptors and an inhibitor of serotonin and noradrenaline uptake. It is licensed in the UK for use in moderate to severe pain.


A systematic review identified seven trials of tramadol in the treatment of neuropathic pain. ref-type="bibr" rid="DTB139C19">19 Five of these were placebo-controlled, and tramadol was superior to placebo in all of them. A meta-analysis of the data from three of these trials (303 patients) found the NNT with tramadol compared with placebo to achieve ≥50% pain relief was 4 (95% CI 3 to 6). The other trials compared tramadol with clomipramine or morphine, but there were insufficient data to draw conclusions of the comparative effect of tramadol from these studies.

Unwanted effects

The most common side effects of tramadol are sedation, nausea, constipation, orthostatic hypotension, and a decrease in the seizure threshold. Sweating, dry mouth and dizziness can also be problematic. ref-type="bibr" rid="DTB139C19">19 There is a risk of serotonin syndrome when it is used in combination with other serotonergic drugs, including SNRIs and SSRIs. ref-type="bibr" rid="DTB139C13">13 The NNH to cause discontinuation of tramadol in the systematic review analysis was 8 (95% CI 5 to 20). ref-type="bibr" rid="DTB139C19">19

Guidelines and clinical use

The NICE clinical guideline notes that tramadol has a lower rate of withdrawal than morphine because of unwanted effects, and that the risk of dependence is also less with tramadol. ref-type="bibr" rid="DTB139C9">9 The guideline recommends tramadol as a third-line treatment for neuropathic pain in non-specialist settings, either as monotherapy or as a rescue analgesic in combination with a second-line treatment.

The EFNS guidelines recommend tramadol as a second- or third-line treatment for painful diabetic neuropathy and central neuropathic pain. ref-type="bibr" rid="DTB139C12">12 It can be considered a first-line option for acute exacerbations of painful diabetic neuropathy.

The IASP guidelines recommend tramadol as a second-line treatment, although it may be considered a first-line treatment in the same situations as opioids are considered first-line – that is, when immediate pain relief is wanted and as short-term therapy for acute neuropathic pain. ref-type="bibr" rid="DTB139C13">13

Topical treatments

Topical capsaicin

In the UK, Axsain cream (0.075%) is licensed for post-herpetic neuralgia, and Qutenza patches (8%) are licensed for painful peripheral neuropathy in non-diabetic adults.

A systematic review of topical capsaicin in the treatment of chronic neuropathic pain identified six placebo-controlled trials (involving 656 patients). ref-type="bibr" rid="DTB139C20">20 When data were pooled, it was found that at 8 weeks 60% of patients responded to topical capsaicin compared with 42% to placebo (relative risk 1.4, 95% CI 1.2 to 1.7; NNT = 6 [95% CI 4 to 10]). The review concluded that although capsaicin had at best moderate efficacy, it may nevertheless be useful as an adjunctive therapy or monotherapy in patients who are either unresponsive to other treatments or intolerant of them.

A more recent review was unable to draw any conclusions about the efficacy of low strength capsaicin (<1%) and concluded that the information “suggests that low-concentration topical capsaicin is without meaningful effect beyond that found in placebo creams; given the potential for bias from small study size, this makes it unlikely that low-concentration topical capsaicin has any meaningful use in clinical practice”. ref-type="bibr" rid="DTB139C21">21

In view of the limited evidence available, the NICE guideline does not recommend the use of topical capsaicin in the non-specialist setting, ref-type="bibr" rid="DTB139C9">9 but notes that, on the basis of clinical experience, it is likely that some patients with localised neuropathic pain may benefit from topical capsaicin.

The Scottish Medicines Consortium has approved Qutenza for restricted use within NHS Scotland for the treatment of post-herpetic neuralgia in patients who have not achieved adequate pain relief from, or who have not tolerated conventional first- and second-line treatments. Treatment should be under the supervision of a specialist in pain management. ref-type="bibr" rid="DTB139C22">22

The EFNS guidelines recommend topical capsaicin as a second- or third-line treatment for post-herpetic neuralgia. ref-type="bibr" rid="DTB139C12">12

Topical lidocaine

Lidocaine is a local anaesthetic agent that acts by reducing the neuronal membrane permeability to sodium, which results in blockage of propagation of nerve impulses. Lidocaine also suppresses spontaneous impulse generation from dorsal root ganglia, which may be useful in post-herpetic neuralgia, since this is the location of the dormant virus.

A lidocaine skin patch is licensed for use in both the UK and the USA for the treatment of post-herpetic neuralgia.

A systematic review identified three studies (involving 314 patients) of the use of topical lidocaine as an analgesic in post-herpetic neuralgia. ref-type="bibr" rid="DTB139C23">23 The reviewers noted that this is a small number of study participants compared with that available for other treatments for post-herpetic neuralgia. On the basis of these trials, the review concluded that there was insufficient evidence to recommend topical lidocaine as a first-line agent in the treatment of post-herpetic neuralgia.

The NICE guideline states that there is a lack of evidence, especially from placebo-controlled trials, and does not recommend the use of topical lidocaine as first-line or second-line treatment across all neuropathic pain conditions in non-specialist settings. ref-type="bibr" rid="DTB139C9">9 However, as with topical capsaicin, the guideline acknowledges that it may be of some benefit to a subset of patients with localised neuropathic pain who are unable to use oral medications, and that it may be especially useful as a rescue analgesic while referral to a specialist pain service is awaited.

The EFNS guidelines recommend lidocaine plasters as a first-line treatment for post-herpetic neuralgia, especially in older patients. ref-type="bibr" rid="DTB139C12">12

The IASP guidelines consider topical lidocaine as a first-line option for the treatment of localised neuropathic pain. ref-type="bibr" rid="DTB139C13">13

Combination treatment

There are relatively few studies of combination therapy and the resulting evidence is oftens of low quality. A systematic review has noted that the small number of available studies for any one specific combination, together with other study factors such as limited size and duration of trials, preclude the recommendation of any specific drug combination for neuropathic pain. ref-type="bibr" rid="DTB139C24">24 Similarly, on the basis of the studies of combination therapy that are analysed in the NICE guideline, ref-type="bibr" rid="DTB139C9">9 no formal recommendations are made for the use of specific drug combinations in non-specialist settings.

However, on the basis of current practice and the experience of patients and carers, the guideline notes that certain drug combinations may be useful options in situations where monotherapy has given unsatisfactory results or where switching or stopping a drug is inappropriate for some reason. Three combinations are mentioned for consideration in such circumstances while referral to a specialist service is awaited: ref-type="bibr" rid="DTB139C9">9

  • amitriptyline (or nortriptyline or imipramine) in combination with pregabalin;

  • duloxetine in combination with pregabalin; and

  • tramadol in combination with a second-line treatment.

The EFNS guidelines state that combination therapy with tricyclic antidepressant plus gabapentin or gabapentin plus an opioid seems to be useful in some patients. ref-type="bibr" rid="DTB139C12">12

The IASP notes that studies of combination therapies are lacking and should be a priority for future research. ref-type="bibr" rid="DTB139C13">13


Table 1:

Cost of selected drugs


For the management of chronic neuropathic pain there is limited availability of high-quality clinical trials of sufficient duration. The lack of head-to-head clinical trials means that many recommendations are based on indirect comparisons and health economic analyses that are only as good as the data on which they are based. Current guidance recommends that a tricyclic antidepressant or either of the antiepileptic drugs gabapentin or pregabalin should be first-line choices for the pharmacological treatment of neuropathic pain of most causes. In the absence of overwhelming evidence of superiority of one drug over another it would seem sensible to use the lowest cost agents first. Opioids should be regarded as second- or third-line treatment options. Carbamazepine remains an option for first-line treatment of trigeminal neuralgia.

There is in general little evidence, or inconclusive evidence, for the use of other drugs, which are not usually recommended for initiation in non-specialist settings, although they may be clinically useful in some patients who have not responded to first- or second-line therapies. It is also not generally recommended that combination therapy should be initiated in non-specialist settings, although there may be some circumstances in which it may be warranted, for instance while a specialist referral is awaited.

Despite there being available several new drugs with proven efficacy in the treatment of neuropathic pain, further studies are warranted, particularly into the efficacy and indications for combination therapies and for the best ways to achieve long-term relief in this often difficult-to-treat clinical condition.