Methadone for neuropathic pain in adults

Abstract

Background

This review replaces an earlier review, "Methadone for chronic non‐cancer pain in adults". This review serves to update the original and includes only studies of neuropathic pain. Methadone belongs to a class of analgesics known as opioids, that are considered the cornerstone of therapy for moderate‐to‐severe postsurgical pain and pain due to life‐threatening illnesses; however, their use in neuropathic pain is controversial. Methadone has many characteristics that differentiate it from other opioids, which suggests that it may have a different efficacy and safety profile.

Objectives

To assess the analgesic efficacy and adverse events of methadone for chronic neuropathic pain in adults.

Search methods

We searched the following databases: CENTRAL (CRSO), MEDLINE (Ovid), and Embase (Ovid), and two clinical trial registries. We also searched the reference lists of retrieved articles. The date of the most recent search was 30 November 2016.

Selection criteria

We included randomised, double‐blind studies of two weeks’ duration or longer, comparing methadone (in any dose, administered by any route, and in any formulation) with placebo or another active treatment in chronic neuropathic pain.

Data collection and analysis

We used standard methodological procedures expected by Cochrane. Two review authors independently considered trials for inclusion in the review, assessed risk of bias, and extracted data. There were insufficient data to perform pooled analyses. We assessed the overall quality of the evidence for each outcome using GRADE and created a 'Summary of findings' table.

Main results

We included three studies, involving 105 participants. All were cross‐over studies, one involving 19 participants with diverse neuropathic pain syndromes, the other two involving 86 participants with postherpetic neuralgia. Study phases ranged from 20 days to approximately eight weeks. All administered methadone orally, in doses ranging from 10 mg to 80 mg daily. Comparators were primarily placebo, but one study also included morphine and tricyclic antidepressants.

The included studies had several limitations related to risk of bias, particularly incomplete reporting, selective outcome reporting, and small sample sizes.

There were very limited data for our primary outcomes of participants with at least 30% or at least 50% pain relief. Two studies reported that 11/29 participants receiving methadone achieved 30% pain relief versus 7/29 participants receiving placebo. Only one study presented data in a manner that allowed us to calculate the number of participants with at least 50% pain relief. None of the 19 participants achieved a 50% reduction in pain intensity, either when receiving methadone or when receiving placebo. No study provided data for our other primary outcomes of Patient Global Impression of Change scale (PGIC) much or very much improved (equivalent to at least 30% pain relief) and PGIC very much improved (equivalent to at least 50% pain relief).

For secondary efficacy outcomes, one study reported maximum and mean pain intensity and pain relief, and reported statistically significant improvements versus placebo for all outcomes with 20 mg daily doses of methadone, but not with 10 mg daily doses. The second study reported differences in pain reduction between methadone (n = 26) and morphine (n = 38) and found morphine to be statistically superior. The third study reported the number of responders (variously defined) for several pain and functional outcomes and found methadone to be statistically superior to placebo for the outcomes of categorical pain intensity and evoked pain. In the two studies that reported data, 0/29 participants withdrew due to lack of efficacy, whereas 4/29 participants withdrew due to adverse events while taking methadone versus 3/29 while taking placebo.

One study reported incidences for several individual adverse events, but found a statistically significant increased incidence for methadone over placebo for only one event, dizziness. The other studies did not report data in a manner that enabled us to analyze adverse events. There were no serious adverse events or deaths reported.

We assessed the quality of the evidence as very low for all efficacy and safety outcomes using GRADE, primarily because of the heterogeneity of study designs and populations, short durations, cross‐over methodology, and few participants and events.

Authors' conclusions

The three studies provide very limited, very low quality evidence of the efficacy and safety of methadone for chronic neuropathic pain, and there were too few data for pooled analysis of efficacy or harm, or to have confidence in the results of the individual studies. No conclusions can be made regarding differences in efficacy or safety between methadone and placebo, other opioids, or other treatments.

Plain language summary

Methadone for neuropathic pain in adults

Bottom line
There is no good evidence to support or reject the suggestion that methadone works in any neuropathic pain condition.

Background
Neuropathic pain is pain coming from a damaged nervous system. It is different from pain messages that are carried along healthy nerves from damaged tissue (e.g. from a fall or cut, or an arthritic knee). Neuropathic pain is often treated by different medicines (drugs) from those used for pain from damaged tissue, which we often think of as painkillers. There are different types of neuropathic pain, with different causes. Some medicines that are used to treat depression or epilepsy can be very effective in some people with neuropathic pain by altering the signal that is carried along nerves that transmit painful stimuli (something that results in a change in how the body works). Sometimes opioid painkillers are used to treat neuropathic pain. Opioid painkillers are drugs such as morphine. Morphine is derived from plants, but many opioids are also made by chemical synthesis rather than being extracted from plants. Methadone is one of these synthetic opioids. Methadone has many characteristics that make it different from other opioids, which may influence its effectiveness or the side effects that patients experience.

Study characteristics
In November 2016, we searched for clinical trials where methadone was used to treat neuropathic pain in adults. We found three small studies, enrolling 105 participants, that met our requirements for the review. The studies were all quite different in their design: the methods of two studies reflected how frequently methadone is prescribed in practice, in that participants received it twice or three times daily. One trial had a more experimental design. All three trials had two phases. The lengths of the studies varied, from 20 days to around eight weeks for each phase. The studies were similar in that all administered low doses of methadone, which may or may not reflect the doses typically prescribed in clinical practice.

Key findings
Two studies looked at how many participants got at least 30% pain relief. Eleven of 29 participants receiving methadone achieved 30% pain relief versus seven of 29 receiving placebo. In one study, none of the 19 participants achieved a 50% reduction in pain intensity, either when receiving methadone or when receiving placebo (a sugar pill). These reductions in pain intensity have been shown to be important to patients. In addition, one study found improvements in average and maximum pain intensity and pain relief when comparing methadone with placebo.

In the two studies that reported dropouts from the study, none of 29 participants dropped out because they thought methadone or the placebo was not helping their pain; whereas four of 29 dropped out because of side effects while taking methadone and three of 29 while taking placebo.

One study reported how many participants had specific side effects, and found increased dizziness with methadone compared to placebo. There were no serious side effects or deaths reported. There was so little information from these studies that we concluded there was no convincing evidence to support or reject a meaningful benefit for methadone versus placebo or any other treatment.

Quality of the evidence
We rated the quality of the evidence as very low because there were only three small studies with different designs, and with few participants and events. In addition, the studies were probably not long enough to show how well methadone would work (or how safe it would be) over a longer time period. Very low quality evidence means that we are very uncertain about the results.

Summary of findings

Summary of findings 1. Methadone compared with placebo for neuropathic pain.

Methadone compared with placebo for neuropathic pain
Patient or population: adults with chronic neuropathic pain Settings: community Intervention: methadone, orally, various doses Comparison: placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative risk (95% CI)	No of studies, participants	Quality of the evidence (GRADE)	Comments
	Probable outcome with methadone	Probable outcome with placebo
Moderate benefit: at least 30% reduction in pain, or PGIC much or very much improved	At least 30% reduction in pain intensity 11/29	At least 30% reduction in pain intensity 7/29	Not calculated	2 studies, 29 participants, 18 events	⊕⊝⊝⊝ Very low	Downgraded 3 times: 2 heterogeneous, short duration, cross‐over studies; mixed diagnoses; few participants and events.
Substantial benefit: at least 50% reduction in pain, or PGIC much improved	At least 50% reduction in pain intensity 0/19	At least 50% reduction in pain intensity 0/19	Not calculated	1 study, 19 participants, 0 events	⊕⊝⊝⊝ Very low	Downgraded 3 times: single, short duration, cross‐over study; few participants, mixed diagnoses, and 0 events.
Withdrawals due to lack of efficacy	0/29	0/29	Not calculated	2 studies, 29 participants, 0 events	⊕⊝⊝⊝ Very low	Downgraded 3 times: 2 heterogeneous, short duration, cross‐over studies; mixed diagnoses; data incompletely presented; few participants and 0 events.
Withdrawals due toadverse event withdrawal	4/29	3/29	Not calculated	2 studies, 29 participants, 7 events	⊕⊝⊝⊝ Very low	Downgraded 3 times: 2 heterogeneous, short duration, cross‐over studies; mixed diagnoses; data incompletely presented; few participants and events.
Participants experiencing anyserious adverse events	No data	No data	‐	‐	⊕⊝⊝⊝ Very low	‐
Deaths	No data	No data	‐	‐	⊕⊝⊝⊝ Very low	‐
CI: confidence interval.
GRADE Working Group grades of evidence High quality: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate quality: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of effect, but there is a possibility that it is substantially different. Low quality: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low quality: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Background

This review replaces an earlier review, "Methadone for chronic non‐cancer pain in adults" (Haroutiunian 2012). This review serves to update the original and includes only studies of neuropathic pain.

This review is based on a template for reviews of drugs used to relieve neuropathic pain. The aim is for all reviews to use the same methods, based on new criteria for what constitutes reliable evidence in chronic pain (Moore 2010a; Moore 2012; Appendix 1).

Description of the condition

The 2011 International Association for the Study of Pain definition of neuropathic pain is "pain caused by a lesion or disease of the somatosensory system" (Jensen 2011), and based on a definition agreed at an earlier consensus meeting (Treede 2008). Neuropathic pain is a consequence of a pathological maladaptive response of the nervous system to 'damage' from a wide variety of potential causes. It is characterised by pain in the absence of a noxious stimulus and may be spontaneous (continuous or paroxysmal) in its temporal characteristics or be evoked by sensory stimuli (dynamic mechanical allodynia where pain is evoked by light touch of the skin). Neuropathic pain is associated with a variety of sensory loss (numbness) and sensory gain (allodynia) clinical phenomena, the exact pattern of which varies between people and disease, perhaps reflecting different pain mechanisms operating in individual people and therefore potentially predictive of response to treatment (Demant 2014; Helfert 2015; von Hehn 2012). Preclinical research hypothesises a bewildering array of possible pain mechanisms that may operate in people with neuropathic pain, which largely reflect pathophysiological responses in both the central and peripheral nervous systems, including neuronal interactions with immune cells (Baron 2012; Calvo 2012; von Hehn 2012). Overall, the treatment gains in neuropathic pain, to even the most effective of available drugs, are modest (Finnerup 2015; Moore 2013a), and a robust classification of neuropathic pain is not yet available (Finnerup 2013).

Neuropathic pain is usually divided according to the cause of nerve injury. There may be many causes, but some common causes of neuropathic pain include diabetes (painful diabetic neuropathy (PDN)), shingles (postherpetic neuralgia (PHN)), amputation (stump and phantom limb pain), neuropathic pain after surgery or trauma, stroke or spinal cord injury, trigeminal neuralgia, and HIV infection. Sometimes the cause is unknown.

Many people with neuropathic pain conditions are significantly disabled with moderate or severe pain for many years. Chronic pain conditions comprised five of the 11 top‐ranking conditions for years lived with disability in 2010 (Vos 2012), and are responsible for considerable loss of quality of life and employment, and increased healthcare costs (Moore 2014a). One US study found the healthcare costs were three‐fold higher for people with neuropathic pain than matched control participants (Berger 2004). One UK study and one German study showed a two‐ to three‐fold higher level of use of healthcare services in people with neuropathic pain than people without (Berger 2009; Berger 2012). For example, studies of PHN have demonstrated large loss of quality of life and substantial costs (Scott 2006; van Hoek 2009).

In systematic reviews, the overall prevalence of neuropathic pain in the general population is reported to be between 7% and 10% (van Hecke 2014), and about 7% in one systematic review of studies published since 2000 (Moore 2014a). In individual countries, prevalence rates have been reported as 3.3% in Austria (Gustorff 2008), 6.9% in France (Bouhassira 2008), and up to 8% in the UK (Torrance 2006). Some forms of neuropathic pain, such as PDN and postsurgical chronic pain (which is often neuropathic in origin), are increasing (Hall 2008; Moulin 2014). The prevalence of PHN is likely to fall if vaccination against the herpes virus becomes widespread.

Estimates of incidence vary between individual studies for particular origins of neuropathic pain, often because of small numbers of cases. In primary care in the UK, between 2002 and 2005, the incidences (per 100,000 person‐years' observation) were 28 (95% confidence interval (CI) 27 to 30) for PHN, 27 (95% CI 26 to 29) for trigeminal neuralgia, 0.8 (95% CI 0.6 to 1.1) for phantom limb pain, and 21 (95% CI 20 to 22) for PDN (Hall 2008). Other studies have estimated an incidence of 4 in 100,000 per year for trigeminal neuralgia (Katusic 1991; Rappaport 1994), and 12.6 per 100,000 person‐years for trigeminal neuralgia and 3.9 per 100,000 person‐years for PHN in one study of facial pain in the Netherlands (Koopman 2009). One systematic review of chronic pain demonstrated that some neuropathic pain conditions, such as PDN, can be more common than other neuropathic pain conditions, with prevalence rates up to 400 per 100,000 person‐years (McQuay 2007).

Neuropathic pain is difficult to treat effectively, with only a minority of people experiencing a clinically relevant benefit from any one intervention (Kalso 2013; Moore 2013a). A multidisciplinary approach is now advocated, combining pharmacological interventions with physical or cognitive (or both) interventions. The evidence for interventional management is very weak, or non‐existent (Dworkin 2013). Conventional analgesics such as paracetamol and nonsteroidal anti‐inflammatory drugs (NSAID) are not thought to be effective, but without evidence to support or refute that view (Moore 2015a). Some people may derive some benefit from a topical lidocaine patch or low‐concentration topical capsaicin, although evidence about benefits is uncertain (Derry 2012; Derry 2014). High‐concentration topical capsaicin may benefit some people with PHN (Derry 2017). Treatment is often by so‐called 'unconventional analgesics' (pain modulators) such as antidepressants (duloxetine and amitriptyline; Lunn 2014; Moore 2014b; Moore 2015b; Sultan 2008), or antiepileptic drugs (gabapentin or pregabalin; Moore 2009; Moore 2014c; Wiffen 2013). Evidence for efficacy of opioids is unconvincing, given the limited, generally low‐quality data (Gaskell 2016; Stannard 2016).

The proportion of people who achieve worthwhile pain relief (typically at least 50% pain intensity reduction; Moore 2013b) is small, generally only 10% to 25% more than with placebo, with numbers needed to treat for an additional beneficial outcome (NNTB) usually between 4 and 10 (Kalso 2013; Moore 2013a). Neuropathic pain is not particularly different from other chronic pain conditions in that only a small proportion of trial participants have a good response to treatment (Moore 2013a).

The current National Institute for Health and Care Excellence (NICE) guidance for the pharmacological management of neuropathic pain suggests offering a choice of amitriptyline, duloxetine, gabapentin, or pregabalin as initial treatment for neuropathic pain (with the exception of trigeminal neuralgia), with switching if the first, second, or third drugs tried are not effective or not tolerated (NICE 2013). This concurs with other guidance (Finnerup 2015).

Description of the intervention

Methadone belongs to the class of drugs known as opioids. It is a synthetic opioid, in the structural subclass of diphenylpropylamines, and was developed in the 1930s. While traditionally used for maintenance and detoxification of people with heroin and other opioid dependencies, it has undergone a resurgence as an analgesic, as it is inexpensive and has distinct pharmacological properties that may confer advantages over other opioids (Trafton 2009).

Methadone is a potent agonist at the mu‐opioid and delta‐opioid receptors. Most clinical practice and research in most countries uses a racaemic mixture of two isomers, levorotatory (L) methadone and dextrorotatory (D) methadone, although in Germany L‐methadone alone is used (Bruera 2002). L‐methadone is 8 to 50 times more potent than D‐methadone in humans and is believed to be almost entirely responsible for its analgesic properties (Fainsinger 1993).

Methadone is available as a lipophilic hydrochloride salt and is available as formulations for oral, rectal, and parenteral administration (Fainsinger 1993; Ripamonti 1997). It is well absorbed by all routes. Oral administration is followed by rapid gastrointestinal absorption with measurable plasma levels at 30 minutes. The peak plasma levels after an oral dose occur at four hours and begin to decline 24 hours after dosing. Oral bioavailability is high, generally over 85% (mean ± standard deviation: 79% ± 21). The recommended dose to be given parenterally is between 50% and 80% of the oral dose (Davis 2001; Gannon 1997). Although local toxicity associated with subcutaneous administration has been reported (Bruera 1991), in many cases this is manageable (Mathew 1999).

How the intervention might work

Opioids provide analgesia by binding to opioid receptors of the mu and kappa class and blocking the release of neurotransmitters such as substance P. Opioid receptors are expressed both centrally and peripherally (during the inflammatory response in injured tissue). Methadone shares many of the analgesic and unwanted effects typical of other opioids. However, it has pharmacokinetic and pharmacodynamic properties that distinguish it from other opioids (Trafton 2009).

1. Unlike many other opioids, methadone has extensive oral bioavailability. Most opioids have less than 40% oral bioavailability (Trafton 2009).

2. Methadone is metabolised in the liver via the cytochrome P‐450 system, and is excreted via the kidneys and intestines. Dosage adjustment is not required in renal or hepatic insufficiency, or in haemodialysis. Additionally, methadone does not appear to produce active, potentially toxic metabolites. Methadone has a long, biphasic elimination half‐life. It may take up to 10 days to reach steady‐state serum levels. It is inherently long acting and is significantly less expensive than opioids that are pharmaceutically manipulated into controlled‐release formulations (Trafton 2009). Its slow onset and offset is also thought to confer methadone a lower risk of addiction in comparison with other opioids (Kreek 2010).

3. N‐methyl‐D‐aspartate (NMDA) receptor antagonism. Activation of the NMDA receptor by excitatory amino acids, such as glutamate, has been implicated in the development of neuropathic pain and appears to have a role in the development of opioid tolerance and opioid‐induced hyperalgesia. While the ability of methadone to block the NMDA receptor has been demonstrated in animal models, it is unclear if this has clinical relevance at normal doses. It is postulated that this property may lend methadone an advantage over other opioids when treating neuropathic pain, with less need for dosage escalations (Trafton 2009).

These distinct properties suggest that methadone may have a different efficacy and safety profile than other commonly prescribed opioids.

Why it is important to do this review

Opioids are considered the cornerstone of therapy for moderate‐to‐severe acute pain or pain of similar intensity due to life‐threatening illnesses, but their long‐term use in non‐cancer pain is controversial. Meta‐analyses have provided efficacy data on opioids for the treatment of neuropathic pain (McNicol 2013). In the US, the therapeutic use of opioids has risen significantly since the early 2000s, with 259 million prescriptions written for opioids in 2012 alone (CDC 2014). This has been accompanied by an increase in the rate of deaths due to opioid overdose. The annual mortality rate due to methadone overdose increased by six times in the decade up to 2009 (CDC 2012). Almost one third of the 15,500 deaths due to prescription opioids in the US in 2009 involved methadone, even though only 2% of opioid prescriptions (around four million) were written for this drug. This disproportionate rate has in part been attributed to diversion of the drug (i.e. deaths have occurred in people using methadone for non‐medical purposes). However, one large cohort study of participants receiving methadone for chronic non‐cancer pain estimated that people receiving methadone had a 46% increased risk of out‐of‐hospital mortality than people receiving an alternative opioid, morphine, suggesting that increased rates of death may be due to pharmacological differences between methadone and other opioids (Ray 2015). In addition to its complex pharmacokinetic profile (outlined in How the intervention might work), methadone causes prolongation of the QT interval on electrocardiogram (ECG) at high doses. QT interval is the interval measured from the beginning of the QRS complex to the end of the T wave on ECG, measuring the time between depolarisation and repolarisation (or recovery) of the heart ventricles. Methadone binds in vitro to the cardiac HERG potassium ion channel (K_v11.1 potassium channel coded by human Ether‐à‐go‐go Related Gene) and has been shown to prolong cardiac depolarisation in a dose‐dependent manner. People with prolonged QT interval are at risk of developing torsades de pointes, a potentially life‐threatening ventricular tachyarrhythmia. Data indicate that methadone may be responsible for sudden cardiac death, even in concentrations that are considered therapeutic for most people (Chugh 2008; Krantz 2009). Intravenous methadone is associated with greater QTc interval prolongation than the oral preparation. This risk may also be increased with concurrent use of other QT interval‐prolonging medications.

The increase in prescribing of methadone in recent years and the accompanying increase in fatalities associated with its use constitutes a public health concern. Conversely, the potentially beneficial properties of methadone, such as NMDA receptor antagonism, may be responsible for increased effectiveness in certain people with neuropathic pain.

The standards used to assess evidence in chronic pain trials have changed substantially since about 2010, with particular attention on trial duration, withdrawals, and statistical imputation following withdrawal, all of which can substantially alter estimates of efficacy. The most important change is the move from using mean pain scores, or mean change in pain scores, to using the number of people who have a large decrease in pain (by at least 50%); this level of pain relief correlates with improvements in comorbid symptoms, function, and quality of life. These standards are set out in the PaPaS Author and Referee Guidance for pain studies of the Cochrane Pain, Palliative and Supportive Care Group (PaPaS 2012), and reflect what people with chronic pain want from treatment (Moore 2013a).

This Cochrane Review assessed evidence using methods that make both statistical and clinical sense, using developing criteria for what constitutes reliable evidence in chronic pain (Moore 2010a). Trials included and analyzed met a minimum of reporting quality (blinding, randomization), validity (duration, dose and timing, diagnosis, outcomes), and size (ideally at least 400 participants in a comparison in which the number needed to treat is 4 or above (Moore 1998)). This approach sets high standards and marks a departure from how reviews were conducted previously.

This rigorous approach is particularly important for opioids in chronic non‐cancer pain. Opioids in clinical trials of non‐cancer pain are associated with very high withdrawal rates of up to 60% over about 12 weeks (Moore 2010b). Many withdrawals occur within the first few weeks, when patients experience pain relief but cannot tolerate the drug. The common practice of using the last observed results carried forward to the end of the trial many weeks later (last observation carried forward (LOCF)) can produce results based largely on people who are no longer in the trial, and who in clinical practice could not achieve pain relief because they could not take the tablets. The newer standards, outlined in Appendix 1, would not allow this LOCF effect and will produce very different results. For example, one large analysis of pooled data from trials in osteoarthritis and chronic low back pain conducted over about 12 weeks judged oxycodone to be effective whereas an analysis of the same data using the new patient‐centred standards showed oxycodone to be significantly worse than placebo (Lange 2010).

One previous Cochrane Review demonstrated the limitations of our knowledge about opioids in neuropathic pain, except in short‐duration studies of 24 hours or less (McNicol 2013), and a review specific to methadone is timely.

Objectives

To assess the analgesic efficacy and adverse events of methadone for chronic neuropathic pain in adults.

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials (RCTs) with double‐blind assessment of participant outcomes following two weeks or more of treatment, although the emphasis of the review was on studies with a duration of eight weeks or longer. We required full journal publication, with the exception of online clinical trial results, summaries of otherwise unpublished clinical trials, and abstracts with sufficient data for analysis. We excluded short abstracts (usually meeting reports). We excluded studies that were non‐randomised, studies of experimental pain, case reports, and clinical observations.

Types of participants

Studies included adults aged 18 years and above with one or more chronic neuropathic pain condition including (but not limited to):

1. cancer‐related neuropathy;

2. central neuropathic pain;

3. complex regional pain syndrome Type II;

4. HIV neuropathy;

5. painful diabetic neuropathy (PDN);

6. phantom limb pain;

7. postherpetic neuralgia (PHN);

8. postoperative or traumatic neuropathic pain;

9. spinal cord injury;

10. trigeminal neuralgia.

Where we included studies of participants with more than one type of neuropathic pain, we intended to analyze results according to the primary condition. We excluded studies of migraine and headache. We excluded studies with fewer than 10 participants in each arm.

Types of interventions

Methadone at any dose, by any route, administered for the relief of neuropathic pain and compared with placebo or any active comparator.

Types of outcome measures

We anticipated that studies would use a variety of outcome measures, with most studies using standard subjective scales (numerical rating scale (NRS) or visual analogue scale (VAS)) for pain intensity or pain relief, or both. We were particularly interested in Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) definitions for moderate and substantial benefit in chronic pain studies (Dworkin 2008). These are defined as:

1. at least 30% pain relief over baseline (moderate);

2. at least 50% pain relief over baseline (substantial);

3. much or very much improved on Patient Global Impression of Change scale (PGIC; moderate);

4. very much improved on PGIC (substantial).

These outcomes are different from those used in most earlier reviews, concentrating as they do on dichotomous outcomes where pain responses do not follow a normal (Gaussian) distribution. People with chronic pain desire high levels of pain relief, ideally more than 50% pain intensity reduction, and ideally having no worse than mild pain (Moore 2013b; O'Brien 2010).

Primary outcomes

1. Participant‐reported pain relief of 30% or greater.

2. Participant‐reported pain relief of 50% or greater.

3. PGIC much or very much improved.

4. PGIC very much improved.

Secondary outcomes

1. Any pain‐related outcome indicating some improvement.

2. Withdrawals due to lack of efficacy, adverse events, and for any cause.

3. Participants experiencing any adverse event.

4. Participants experiencing any serious adverse event. Serious adverse events typically include any untoward medical occurrence or effect that at any dose results in death, is life‐threatening, requires hospitalisation or prolongation of existing hospitalisation, results in persistent or significant disability or incapacity, is a congenital anomaly or birth defect, is an 'important medical event' that may jeopardise the person, or may require an intervention to prevent one of the above characteristics or consequences. While we extracted all data related to serious adverse events, we planned to pay particular attention to reports of respiratory depression and cardiovascular events.

5. Specific adverse events, particularly somnolence and dizziness.

Search methods for identification of studies

Electronic searches

We searched the following databases, without language restrictions:

1. Cochrane Central Register of Controlled Trials (CENTRAL, via the Cochrane Register of Studies Online database (CRSO)) to 30 November 2016;

2. MEDLINE (via Ovid) from 1946 to 30 November 2016;

3. Embase (via Ovid) from 1974 to 30 November 2016.

The search strategies for CENTRAL, MEDLINE, and Embase are in Appendix 2, Appendix 3, and Appendix 4, respectively.

Searching other resources

We reviewed the bibliographies of any RCTs identified and review articles, and searched ClinicalTrials.gov (ClinicalTrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (apps.who.int/trialsearch/) to identify additional published or unpublished data. We did not contact investigators or study sponsors.

Data collection and analysis

We intended to perform separate analyses according to particular neuropathic pain conditions, but there were insufficient data to allow us to do this (the majority of participants had PHN). We intended to combine different neuropathic pain conditions in analyses for exploratory purposes only.

Selection of studies

We determined eligibility by reading the abstract of each study identified by the search. We eliminated studies that clearly did not satisfy the inclusion criteria, and obtained full copies of the remaining studies. Two review authors (EM and MF or RS) read these studies independently and reached agreement by discussion. We did not anonymise the studies before assessment. We created a PRISMA flow chart (Figure 1).

1.

Study flow diagram.

Data extraction and management

Two review authors (EM and MF or RS) extracted data independently using a standard form and checked for agreement before entry into Review Manager 5 (RevMan 2014), or any other analysis tool. We included information about the pain condition and number of participants treated, drug and dosing regimen, study design (placebo or active control), study duration and follow‐up, analgesic outcome measures and results, withdrawals, and adverse events (participants experiencing any adverse event or serious adverse event).

Assessment of risk of bias in included studies

We used the Oxford Quality Score as the basis for inclusion (Jadad 1996), limiting inclusion to studies that were randomised and double‐blind as a minimum.

Two review authors (EM and MF or RS) independently assessed risk of bias for each study, using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), and adapted from those used by the Cochrane Pregnancy and Childbirth Group, with any disagreements resolved by discussion. We assessed the following for each study:

1. Random sequence generation (checking for possible selection bias). We assessed the method used to generate the allocation sequence as: low risk of bias (i.e. any truly random process, e.g. random number table; computer random number generator); unclear risk of bias (when the method used to generate the sequence was not clearly stated). We excluded studies at a high risk of bias that used a non‐random process (e.g. odd or even date of birth; hospital or clinic record number).

2. Allocation concealment (checking for possible selection bias). The method used to conceal allocation to interventions prior to assignment determines whether intervention allocation could have been foreseen in advance of, or during, recruitment, or changed after assignment. We assessed the methods as: low risk of bias (e.g. telephone or central randomization; consecutively numbered, sealed, opaque envelopes); unclear risk of bias (when method not clearly stated). We excluded studies that did not conceal allocation and were therefore at a high risk of bias (e.g. open list).

3. Blinding of participants and personnel (checking for possible performance bias), and blinding of outcome assessment (checking for possible detection bias). We assessed the methods used to blind study personnel and participants (all outcomes were self‐assessed) from knowledge of which intervention a participant received. We assessed the methods as: low risk of bias (e.g. study stated that it was blinded and described the method used to achieve blinding, e.g. identical tablets, matched in appearance and smell); unclear risk of bias (study stated that it was blinded but did not provide an adequate description of how it was achieved). We excluded studies at a high risk of bias that were not double‐blind.

4. Incomplete outcome data (checking for possible attrition bias due to the amount, nature, and handling of incomplete outcome data). We assessed the methods used to deal with incomplete data as: low risk of bias (i.e. less than 10% of participants did not complete the study or used 'baseline observation carried forward' analysis, or both); unclear risk of bias (used LOCF analysis); or high risk of bias (used 'completer' analysis).

5. Selective reporting (checking for reporting bias). We assessed whether primary and secondary outcome measures were prespecified and whether these were consistent with those reported.

6. Size of study (checking for possible biases confounded by small size). We assessed studies as being at low risk of bias (200 participants or more per treatment arm); unclear risk of bias (50 to 199 participants per treatment arm); or high risk of bias (fewer than 50 participants per treatment arm).

Measures of treatment effect

We intended to calculate NNTBs as the reciprocal of the absolute risk reduction (McQuay 1998). For unwanted effects, the number needed to treat becomes the number needed to treat for an additional harmful outcome (NNTH) and is calculated in the same manner. We planned to use dichotomous data to calculate risk ratio (RR) with 95% CIs using a fixed‐effect model unless we found significant statistical heterogeneity (see Assessment of heterogeneity).

Unit of analysis issues

We planned to split the control treatment arm between active treatment arms in a single study if the active treatment arms were not combined for analysis.

For cross‐over studies, we planned to use only the first period, if this was available. Where only combined data for both periods were reported, as was the case with the three included studies, we treated each study as if it was a parallel study, drawing attention to the potential bias that this confers, and interpreted the results accordingly (Elbourne 2002).

Dealing with missing data

We planned to use intention‐to‐treat (ITT) analysis where the ITT population consisted of participants who were randomised, took at least one dose of the assigned study medication, and provided at least one postbaseline assessment. Missing participants would be assigned zero improvement wherever possible.

Assessment of heterogeneity

We intended to deal with clinical heterogeneity by combining studies that examined similar conditions. We planned to assess statistical heterogeneity visually (L'Abbé 1987), and with the use of the I² statistic. When the I² value was greater than 50%, we planned to consider possible reasons for this. However, due to lack of data, no meta‐analysis was performed.

Assessment of reporting biases

The aim of this review was to use dichotomous outcomes of known utility and of value to people (Hoffman 2010; Moore 2010b; Moore 2010c; Moore 2010d; Moore 2013b). The review did not depend on what the authors of the original studies chose to report or not, though clearly difficulties arose in studies failing to report any dichotomous results. We extracted and used continuous data, which probably reflect efficacy and utility poorly, and may be useful for illustrative purposes only.

We planned to assess publication bias using a method designed to detect the amount of unpublished data with a null effect required to make any result clinically irrelevant (usually taken to mean an NNTB of 10 or higher; Moore 2008).

Data synthesis

We planned to use a fixed‐effect model for meta‐analysis. We planned to use a random‐effects model for meta‐analysis if there was significant clinical heterogeneity and it was considered appropriate to combine studies. We intended to perform pooled analyses where there were data from at least 200 participants in the comparison to avoid problems with random chance effects (Moore 1998), or bias in small studies and data sets (Dechartres 2013; Dechartres 2014; Nüesch 2010). There were an insufficient number of participants in the included studies to permit pooled analysis.

Quality of the evidence

We used the GRADE approach to assess the quality of evidence related to each of the key outcomes, and reported our judgement on the quality of the evidence in Table 1 (Chapter 12, Higgins 2011; Appendix 5).

In addition, there may be circumstances where the overall rating for a particular outcome needs to be adjusted as recommended by GRADE guidelines (Guyatt 2013a). For example, if there are so few data that the results are highly susceptible to the random play of chance, or if studies use LOCF imputation in circumstances where there are substantial differences in adverse event withdrawals, one would have no confidence in the result, and would need to downgrade the quality of the evidence by three levels, to very low quality. In circumstances where there were no data reported for an outcome, we reported the level of evidence as very low quality (Guyatt 2013b).

'Summary of findings' table

We included a 'Summary of findings' table as set out in the Cochrane Pain, Palliative and Supportive Care Group (PaPaS) Author and Referee Guidance (PaPaS 2012), and recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Chapter 11, Higgins 2011). The table included outcomes equivalent to moderate or substantial benefit of at least 30% and at least 50% pain intensity reduction, PGIC (possibly at least substantial improvement and at least moderate improvement) (Dworkin 2008), withdrawals due to lack of efficacy, withdrawals due to adverse events, serious adverse events, and death (a particular serious adverse event).

Subgroup analysis and investigation of heterogeneity

We did not plan to perform subgroup analyses since experience of previous reviews indicated that there would be too few data for any meaningful subgroup analysis.

Sensitivity analysis

We did not plan to perform sensitivity analysis because the evidence base was known to be too small to allow reliable analysis; results from neuropathic pain of different origins was not pooled in the primary analyses. We planned to examine details of dose escalation schedules in the unlikely situation that this could provide some basis for a sensitivity analysis.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; and Characteristics of ongoing studies tables.

Results of the search

Our literature search yielded 103 references from CENTRAL, 901 references from MEDLINE, and 688 studies from Embase. Review of the abstracts associated with these references identified only four potentially relevant studies. The remaining references clearly did not meet inclusion criteria, without the need to obtain a full‐text for confirmation. We excluded one of these four studies as it was not blinded. Our search of clinical trial websites yielded six ongoing or completed trials from clinicaltrials.gov and 289 studies from the WHO ICTRP. From these, we found one relevant ongoing study (NCT01205516), and one study that is completed, but has not yet posted results (NCT02233452) (Figure 1).

Included studies

Three studies with heterogeneous designs fulfilled the inclusion criteria. All were cross‐over trials and all administered interventions orally. One trial was conducted in Brazil (Teixeira 2013), one in England (Morley 2003), and one in the USA (Raja 2002). Enrolment ranged from 10 to 76 participants, with the number of participants receiving methadone in each study ranging from 10 to 26.

Inclusion criteria varied among studies. Minimum duration of neuropathic pain for inclusion was three months in two studies (Morley 2003; Raja 2002), and six months in one study (Teixeira 2013). One study had no criterion for minimum baseline pain (Morley 2003; participants' baseline pain intensity ranged from 37/100 to 98/100), one study included participants with NRS pain intensity of more than 4/10 (Raja 2002), and one included participants with VAS pain intensity of more than 40/100 (Teixeira 2013). Permission to continue to use previous medications, or to use rescue analgesics, also varied among studies. Morley 2003 allowed participants to continue previous medications, some of which were opioids. In Raja 2002, all prescribed pain medications were discontinued at least one week prior to enrolment; however, NSAIDs and paracetamol (acetaminophen) were permitted during the study for analgesic rescue. Teixeira 2013 enrolled only opioid‐naive participants, but participants could continue to receive their current non‐opioid analgesic regimen.

Study design also varied. Morley 2003 studied two dosing regimens of methadone in a placebo‐controlled trial in participants with diverse neuropathic pain syndromes. Both regimens were tested over separate 20‐day phases. Participants received either methadone or placebo on odd days and a rest day on even days, that is, they received five days of methadone for each phase. The "low‐dose" phase administered methadone 5 mg (or placebo) twice daily, whereas the "high‐dose" phase administered methadone 10 mg (or placebo) twice daily. The washout period between phases was not specified. Raja 2002 compared an opioid (morphine or methadone) with a tricyclic antidepressant (nortriptyline or desipramine) and placebo in a three‐phase trial, with each phase lasting approximately eight weeks, in participants with PHN. Each phase contained a titration, maintenance, and taper subphase. Participants received methadone only if they were unable to tolerate morphine during the titration subphase. Teixeira 2013 was a pilot proof‐of‐concept study, also in participants with PHN. It consisted of two phases, each of three weeks' duration, with a three‐week washout period in between phases. Participants received either methadone 5 mg twice daily or placebo during the first phase, then the opposite treatment during the second phase.

Excluded studies

One study, Haumann 2016, compared methadone with fentanyl in participants with head and neck cancers that had a neuropathic pain component. The study was not double‐blinded.

Studies awaiting classification

We identified one 12‐week RCT on ClinicalTrials.gov that compared unspecified doses of oral methadone (n = 13), ketamine (n =13), or methadone plus ketamine (n =13) in participants with neuropathic pain of at least six months' duration. The study has been completed, but has not yet posted study results (NCT02233452).

Ongoing studies

One RCT, found on ClinicalTrials.gov, is expected to be completed in December of 2017 (NCT01205516). It aims to enrol 180 participants with neuropathic pain of central or peripheral origin, and administer oral methadone or placebo, based on a five‐week titration schedule, a six‐week maintenance dose phase, and a four‐week taper.

Risk of bias in included studies

Our findings are summarised in Figure 2 and Figure 3.

2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Both Morley 2003 and Raja 2002 employed adequate randomization methods, by using tables of random numbers (Morley 2003), or by computer‐generated randomization (Raja 2002). Teixeira 2013 did not describe how randomization was performed. Similarly, both Morley 2003 and Raja 2002 employed acceptable methods to conceal allocation, primarily through central allocation. Teixeira 2013 did not mention allocation concealment.

Blinding

All three studies reported that all interventions appeared identical. In addition, both Morley 2003 and Raja 2002 stated that interventions were prepared by third parties not directly involved in study assessments. None of the studies assessed adequacy of blinding by asking participants to guess which intervention they had received.

Incomplete outcome data

Morley 2003 assessed only those participants completing the study; however, dropouts were similar in number and reason between groups. Raja 2002 employed an ITT analysis, but did not impute any data. It was unclear how many participants completed the methadone phase. In Teixeira 2013, all participants completed the study and it appeared that data were complete for all participants at all time points.

Selective reporting

We assessed both Morley 2003 and Raja 2002 as having unclear risk of bias. For Morley 2003, the secondary outcomes of severity of adverse events and use of additional analgesics were assessed but not presented, and neuropathic pain scale scores were presented but not analyzed. For Raja 2002, participants were permitted to take non‐prescription NSAIDs and paracetamol during the study for analgesic rescue, but the authors did not state that rescue use was assessed as an outcome and did not report data in their results. Teixeira 2013 assessed several outcomes that were not reported in their results section. Conversely, a correlation analysis was presented in the results section that was not mentioned in the methods section. Consequently, we assessed this study as having a high risk of reporting bias.

Other potential sources of bias

We assessed all three studies as having a high risk of bias due to low sample size. In each study, fewer than 50 participants received methadone. In addition, only one study employed study phase durations of eight weeks (Raja 2002). The other two studies had study phase durations of 20 days (Morley 2003) or three weeks (Teixeira 2013). In Morley 2003, participants received only five days of methadone in the low‐ and high‐dose phases.

Effects of interventions

See: Table 1

Heterogeneity of participants, methodologies and outcomes assessed, and the low number of total events in the included studies, precluded us performing meta‐analyses. Data from the first phase of each study were not provided separately; therefore, we analyzed the data from both phases of each study, as if they had been derived from parallel studies.

Efficacy

Details of efficacy outcomes are provided in Appendix 6.

Participant‐reported pain relief of 30% or greater or 50% or greater

Only one study presented data in a manner that allowed us to calculate the number of participants with at least 50% pain relief (Morley 2003). The study authors presented mean individual participant data from the low‐dose phase (methadone 5 mg or placebo twice daily) for both baseline pain intensity and pain intensity during treatment. Based on these data, we ascertained that 0/19 participants achieved a 50% reduction in pain intensity, either when receiving methadone or when receiving placebo. However, from the same data, we calculated that four (21%) participants achieved a 30% or greater reduction in pain intensity when receiving methadone and two (11%) participants while receiving placebo during the low‐dose phase. Teixeira 2013 reported the number of participants with at least a 30% reduction in VAS pain intensity post interventions. Seven of 10 participants (70%) achieved this outcome while receiving methadone and 5/10 (50%) while receiving placebo (P = 0.363). Two participants achieved at least a 30% reduction in pain intensity during treatment with both methadone and placebo.

PGIC much or very much improved; PGIC much improved

None of the studies reported PGIC much or very much improved or PGIC much improved.

Any pain‐related outcome indicating some improvement

Morley 2003 supplied individual participant data; therefore, we were able to perform our own statistical analysis of each outcome and compare our results with those results reported by the authors. We performed a paired t‐test to compare the difference in mean pain intensity between methadone and placebo for both the low‐ (10 mg daily) and high‐dose (20 mg daily) phases. Both phases demonstrated that methadone was more effective than placebo in reducing mean pain intensity (low‐dose: P = 0.042; high‐dose P = 0.020). Pain relief reported with both doses of methadone also demonstrated statistical superiority versus placebo (low‐dose: P = 0.003; high‐dose: P < 0.001, Wilcoxon Signed Rank Test). The results of our analyses of pain intensity and pain relief agreed with those analyses presented in the paper for the high‐dose phase, but differed from those presented for the low‐dose phase, where the authors reported that the differences between methadone and placebo were not statistically significant for either outcome. The authors also reported statistically significant improvements in maximum pain intensity during the high‐dose phase (P = 0.013), but not during the low‐dose phase (P = 0.065). Of note, 3/18 participants completing the low‐dose phase had a mean pain intensity below 40/100 (i.e. mild), while receiving methadone, compared with only one participant receiving placebo. In the high‐dose phase, two participants had a mean pain intensity below 40/100 while receiving methadone, compared with only one while receiving placebo. Appendix 6 lists results as presented by the authors.

Raja 2002 presented the majority of results as combined data during the opioid phase (i.e. data from both participants receiving morphine and methadone were presented as a single result). Only pain intensity difference (0 to 10 NRS) data were presented separately for methadone and morphine. The authors stated that the reduction in pain was greater with morphine (reduction 2.2, 95% CI ‐2.7 to ‐1.6, n = 38) than methadone (reduction 1.2, 95% CI ‐1.8 to ‐0.5, n = 26) (P = 0.02). In comparison, participants had a 0.2‐point reduction with placebo (95% CI ‐0.7 to ‐0.2, n = 56) and a 1.4‐point reduction while receiving a tricyclic antidepressant (95% CI ‐1.8 to ‐0.9, n = 59).

In addition to presenting data for at least 30% pain relief, Teixeira 2013 also presented the number of participants with at least a 30% reduction in evoked pain (pain that occurred due to a painful or non‐painful stimulus) and an improvement in the category verbal scale (mild, moderate, or severe pain), although what constituted improvement in the category verbal scale was not defined. All 10 participants had at least 30% reduction in evoked pain while receiving methadone, 5/10 (50%) participants had at least 30% reduction in evoked pain while receiving placebo, and 5/10 participants achieved such a reduction during both phases (P = 0.031). Numbers were identical for improvement in category verbal scale (i.e. all 10 participants) while receiving methadone, five while receiving placebo, and five achieving improvement during both phases (P = 0.031). The authors further stated that the activities of daily living (no data presented) and McGill Pain Questionnaire subscores (sensitive, affective, evaluative, and miscellaneous) did not significantly change after treatment with methadone and that methadone did not have any negative impact on daily activities, such as concentration, mood, or sleep.

Quality of the evidence

We downgraded the quality of the evidence for efficacy to very low because there were only three studies with a small number of participants with various diagnoses, because of the incomplete reporting of data, and because of the low number of events (see Table 1).

Withdrawals

Details of withdrawals are provided in Appendix 7.

Withdrawals due to lack of efficacy

No participants withdrew due to lack of efficacy in either phase in Morley 2003. Raja 2002 combined data for morphine and methadone; therefore, we were unable to identify the number of participants withdrawing due to lack of efficacy while receiving methadone. In Teixeira 2013, all 10 participants completed both three‐week phases (methadone or placebo) of the study.

Withdrawals due to adverse events

Seven of 19 (37%) participants withdrew due to adverse events in Morley 2003, one during phase I (low‐dose) and six during Phase II (high‐dose). In Phase I, one participant withdrew because of severe nausea, dizziness, and sweating on the first occasion he took methadone. Of the six participants who withdrew during Phase II, three withdrew on days when they were receiving methadone, and three on days when they were receiving placebo. The reasons for withdrawal included severe nausea (three participants), dizziness, vomiting, sweating, and disorientation with severe headaches. As with withdrawals due to lack of efficacy, we were unable to identify the number of participants who withdrew due to adverse events in Raja 2002, although it was noted that 7/20 participants withdrawing while taking opioids did so because of adverse effects. As noted above, no participants withdrew from Teixeira 2013.

Withdrawals for any cause

One participant withdrew between phases due to an intercurrent illness in Morley 2003. As with withdrawals due to lack of efficacy and adverse events, we were unable to identify the number of participants who withdrew for any cause in Raja 2002; however, it was noted that primary reasons for dropping out during both periods included other medical problems (n = 6 during the opioid phase, n = 1 during the antidepressant phase) and concerns of family members (n = 5 during the opioid phase, n = 2 during the antidepressant phase). There were no withdrawals in Teixeira 2013.

Quality of the evidence

We downgraded the quality of the evidence for withdrawals to very low because there were only three studies with a small number of participants with mixed diagnoses, because data were incompletely presented, and because of the low numbers of events.

Participants experiencing adverse events

Details of adverse events are provided in Appendix 7.

Any adverse event

In Morley 2003 15/19 (79%) participants experienced an adverse event while receiving methadone, and 7/19 (37%) experienced an adverse event while receiving placebo during the low‐dose phase (Phase I). In Phase II, 11/17 (65%) participants experienced an adverse event while receiving methadone, and 8/17 (47%) experienced an adverse event while receiving placebo. Adverse events were reported as mild to moderate in participants who completed the trial. Adverse events in participants receiving methadone were not reported separately in Raja 2002. Teixeira 2013 stated that the frequency of reported adverse events did not differ significantly in either treatment period (methadone or placebo), but did not present data.

Any serious adverse events

None of the studies reported that any participant experienced a serious adverse event or that any participant died; however, none explicitly stated that no participant had such an event.

Specific adverse events

No study reported that participants had respiratory depression or cardiac abnormalities.

During Phase I of Morley 2003, the most common adverse events reported by participants receiving methadone were nausea (7/19, 37%) and dizziness (6/19, 32%), with vomiting, pruritus, constipation, and somnolence also experienced. The incidence was generally numerically higher than adverse events experienced while participants were taking placebo, although statistical significance was not reported. To compare adverse event rate differences between methadone and placebo we used the McNemar test, as recommended for comparison of dichotomous data in studies with cross‐over design (Elbourne 2002). The test is based on comparing discordant pairs of responses (e.g. adverse event developed under methadone, but not placebo) for each intervention. Our analysis of both phases did not demonstrate a statistically significant difference between methadone and placebo for the incidence of any specific adverse event with the exception of dizziness in the low‐dose phase, which occurred more frequently in participants receiving methadone (either on the methadone day or the rest day after methadone) than in participants receiving placebo (placebo day or rest day after placebo) (P = 0.041). Six of 19 (32%) participants reported dizziness while taking methadone, compared with 0/19 participants receiving placebo.

Quality of the evidence

We downgraded the quality of the evidence for adverse events to very low because there were only three studies with a small number of participants with mixed diagnoses, because the data were incompletely presented, and because of the low numbers of events.

Discussion

Summary of main results

We found only three studies to include in this review. All employed a cross‐over design and administered oral methadone. Two studies enrolled participants with PHN (Raja 2002; Teixeira 2013), while one enrolled a population with mixed neuropathies (Morley 2003). Morley 2003 and Teixeira 2013 studied relatively low fixed doses (10 mg to 20 mg daily), whereas Raja 2002 allowed dose titration to a maximum of 80 mg over a period of approximately six weeks, perhaps more closely reflecting clinical practice. Morley 2003 and Teixeira 2013 compared methadone with placebo. Raja 2002 compared methadone with placebo, a tricyclic antidepressant, and another opioid (morphine), but only presented sufficient data to allow us to compare methadone with morphine.

Efficacy

None of the three included studies presented sufficient data for us to be able to make any firm conclusions regarding the efficacy of methadone as an analgesic. Morley 2003 presented continuous VAS data with no reporting of numbers of participants with clinically significant reductions in pain. However, we were able to make such calculations, as individual participant data were listed. The study assessed neither quality of life nor functional outcomes. Raja 2002 assessed many more valid outcomes, but did not present the majority of results in a format that we could analyze, due to the fact that they did not report methadone and morphine data separately. Teixeira 2013 presented outcome data in terms of the number of predefined responders for a number of pain assessments during each phase.

The extremely limited data available to us did not demonstrate numerical or statistically significant differences in the number of participants with at least a 50% reduction in pain (in a single study no participants achieved this with methadone or placebo), and showed a numerically, but not statistically, significant difference in the number of participants with at least 30% reduction in pain (Table 1). Analysis of secondary efficacy outcomes did demonstrate a statistically significant reduction in maximum and mean pain intensity and an increase in pain relief when methadone 20 mg daily (but not 10 mg daily) was compared with placebo (Morley 2003). As noted, the authors' analysis differed from ours (we found improvements to be statistically significant in both phases), perhaps due to differences in statistical methodology. Regardless, the number of participants in these comparisons was too low to place any confidence in their findings. There were even fewer data for assessments of quality of life or functioning; therefore, we can make no conclusions related to these outcomes. None of the included studies presented data that could aid assessment of potential advantages of methadone regarding its action as an NMDA antagonist and possible related reduction in tolerance, opioid‐induced hyperalgesia, or neuropathic pain.

There were no reports of dropouts due to lack of efficacy in any of the studies, which is unusual in studies of opioids (McNicol 2013). This may be attributed to the short duration of two of the studies (Morley 2003; Teixeira 2013), and the lack of reporting in the longer study (Raja 2002), and the low overall number of participants.

Safety

Adverse events were similar to those commonly reported in opioid studies (McNicol 2008). Morley 2003 compared incidence of various adverse events with methadone versus placebo. Only dizziness was shown to occur statistically more frequently, and only during the low‐dose (10 mg daily) phase. Neither Raja 2002 nor Teixeira 2013 provided usable safety data. Where reported, around 14% of participants withdrew from the methadone phases because of adverse events. This is similar to percentages reported for opioids as a class when treating neuropathic pain (McNicol 2013).

There were no reports of QTc prolongation or respiratory depression in any study, probably as a result of several factors: none of the manuscripts explicitly mentioned assessing the occurrence of these events; the exclusion of participants with known risk factors for cardiac or respiratory events; the relative infrequency with which these adverse effects occurred, and the very low total numbers of participants; because participants were more closely monitored than in regular clinical settings; and the low doses of methadone received across the three studies. It is suggested that both adverse effects occur more frequently in people receiving methadone than in people receiving other opioids, but there were no data from the current review to support these assumptions (McNicol 2008). Therefore, we were unable to demonstrate any difference between methadone's and other opioids' safety profiles.

Overall completeness and applicability of evidence

We have evidence from only three studies enrolling 105 participants, of whom only 55 received methadone. The majority of the participants had PHN (n = 89). Only four participants had PDN, with the remaining participants having other, less common, neuropathies. Morley 2003 and Raja 2002 included participants who had previously or were currently receiving opioids, whereas Teixeira 2013 enrolled only opioid‐naive participants.

There was very little evidence describing the efficacy of methadone according to our primary outcomes (i.e. those that have been shown to be clinically important to patients). The evidence from secondary outcomes of mean pain intensity and pain relief is probably of little value, given the low number of participants and the limited utility of such outcomes.

The doses of methadone employed in two of the studies were low, ranging from 10 mg to 20 mg daily (Morley 2003; Teixeira 2013). In the study in which participants were allowed to titrate dose to a maximum of 80 mg daily, the mean maintenance dose was only 15 mg (Raja 2002). It is unclear what mean doses are prescribed in clinical settings, but it is possible that the low doses employed in these studies reflect real‐world prescribing. One large retrospective cohort study of prescribing practices in a single US state reported prescribing of a median daily dose of methadone 40 mg in people with non‐cancer pain, but showed that people receiving doses of methadone 20 mg or less daily were also at an increased risk of mortality versus people receiving morphine (Ray 2015). There are insufficient data from the included studies to demonstrate a dose‐efficacy response, or an increase in adverse events with higher doses; therefore, the limited findings from this review may not be applicable to people receiving doses above 20 mg daily.

Based on such limited data, we cannot draw any conclusions regarding variations in methadone's analgesic efficacy for different types of neuropathic pain or when administered via different routes. We have very‐limited‐to‐no data to make conclusions regarding short‐ versus long‐term administration, low‐ versus high‐dose regimens, or efficacy compared to other analgesics.

Quality of the evidence

When assessing the quality of findings using GRADE, we ranked quality as very low across all efficacy and safety outcomes, as shown in Table 1.

The included studies had several limitations related to risk of bias (Risk of bias in included studies), particularly due to incomplete reporting, selective outcome reporting, and small‐study sizes. At least two of the studies had risk of bias due to incomplete reporting (Moore 2012). Morley 2003 analyzed only those participants completing the study (only 11/19 participants enrolled completed both phases). Raja 2002 performed an ITT analysis, but did not impute data, instead choosing to use the last three available pain ratings for participants that did not complete the study. It is possible, therefore, that both studies may be at risk for overestimating the efficacy of methadone. Although Teixeira 2013 did not specify how missing data were dealt with, it appears that all participants completed the study and provided data for all outcomes at all times. While there were multiple inconsistencies across the studies in the manner in which outcomes were reported, perhaps the most serious risk of bias from selective outcome reporting was the lack of inclusion of results for the outcomes of number of participants with at least 30% or 50% reduction in pain. None of the included studies presented dichotomous data allowing assessment of number of participants with 50% pain relief or better. We were able to derive these numbers from one study (Morley 2003), but the very small size of the study makes any findings meaningless. Similarly, there were very few participants and events for the outcomes of 30% pain relief or better. The very low number of participants receiving methadone in all studies also produces a high potential for bias (Moore 2010a).

In addition to the noted risks of bias, there were several other issues that affected the quality of the body of evidence. There were inconsistencies across the studies related to populations, interventions, and outcomes assessed. As noted in the Summary of main results, the populations differed with respect to their diagnosis and time since diagnosis, and the analgesic regimens currently or previously employed. Although doses of methadone were similar across the studies, the manner in which they were administered differed. Neither Morley 2003 nor Teixeira 2013 allowed the opportunity for participants to titrate doses to effectiveness. The study design of Morley 2003 also did not reflect clinical practice, in that participants alternated days of receiving either methadone or placebo with rest days. There are problems of indirectness of evidence, related to the populations studied; there were almost no data for any diagnosis other than PHN. Imprecision of findings was most problematic, given the very low sample sizes and number of events. The possibility of publication bias due to unpublished or unidentified studies cannot be excluded, given that few studies would be required to change the results seen in the included studies.

Two of the studies had follow‐up durations of three weeks or less for each phase (Morley 2003; Teixeira 2013). Short‐duration studies of chronic diseases may overestimate treatment effects (Moore 2010a).

Last, there were problems related to the nature of cross‐over studies. Whereas Raja 2002 had a one‐week washout between interventions and Teixeira 2013 had a three‐week washout, Morley 2003 was designed so that participants had only a one‐day washout ("rest day") between receiving methadone and placebo. Given the long half‐life of methadone, the possibility of a carryover effect for both positive and adverse effects cannot be ruled out. Additionally, only one study reported that baseline pain data were similar at the start of each phase of treatment (Raja 2002).

The very low quality of the evidence means that we have very little confidence in the effect estimate, and that the true effect is likely to be substantially different from the estimate of effect.

Potential biases in the review process

Despite our assessment that few unpublished or unidentified studies would be required to change the findings from our review, we think it unlikely that there are many of these. We attempted to minimise the potential for publication bias related to unpublished or unidentified studies by assessing clinical trial registries and multiple databases, respectively.

Our inclusion criteria may have introduced bias. We accepted cross‐over studies despite the methodological issues that exist with this type of study (Elbourne 2002). Specifically, it is recommended that only data from the first phase (if presented) be analyzed (i.e. as if the study was a parallel trial). This was not possible for any of the included studies, and if it had been, we would have been forced to exclude two of the studies as there would have been fewer than 10 participants in each arm (Moore 1998; Moore 2010a). Instead, we treated the overall results as if they had been derived from a parallel study. We chose to do this as there were so few studies available, but recommend that the findings of our analysis be treated with extreme caution.

Agreements and disagreements with other studies or reviews

Recent Cochrane Reviews and guidelines from professional organisations came to broadly similar conclusions to our review.

This review replaces an earlier review, "Methadone for chronic non‐cancer pain in adults" (Haroutiunian 2012). While this review updates the original and includes only studies of neuropathic pain, the lack of data in both reviews and the overall conclusions are similar. Two of the studies in this review were also included in the earlier review (Morley 2003; Raja 2002). An overview of opioids as a class for neuropathic pain was published in the Cochrane Library in 2013 (McNicol 2013). It assessed similar outcomes and included two of the three studies included in this review (Morley 2003; Raja 2002). It concluded that there was limited evidence of efficacy of opioids, based on there being a greater number of studies than included in our review, but highlighted many of the same methodological shortcomings of existing trials. It made no recommendations specific to methadone.

One 2015 review of all pharmacotherapy for neuropathic pain in adults proposed that opioids be considered third‐line agents, based on safety concerns surrounding long‐term use (Finnerup 2015). The majority of data for opioids came from studies of oxycodone and morphine. Specific recommendations related to methadone use were not made.

Finally, Nicholson 2017 performed a review for the Cochrane Library using similar criteria to ours, but instead looking at methadone for cancer pain. This review also lacked data due to the small number of studies meeting the inclusion criteria and their design heterogeneity preventing meta‐analysis of outcomes. Therefore, the possibility of extrapolating data from people with cancer pain to people with neuropathic pain does not currently exist.

Authors' conclusions

Implications for practice.

There is very limited evidence to judge the effectiveness of methadone in neuropathic pain. Equally, there is no evidence from randomised controlled trials that methadone has a different safety profile to other opioids. However, epidemiological data suggesting an increased risk of adverse events in people receiving methadone versus other opioids (CDC 2012; Ray 2015), imply that methadone should not be considered a first‐line opioid.

For people with neuropathic pain

There is insufficient evidence to support or reject the suggestion that methadone has any efficacy in any neuropathic pain condition.

For clinicians

There is insufficient evidence to support or reject the suggestion that methadone has any efficacy in any neuropathic pain condition.

For policy makers

There is insufficient evidence to support or reject the suggestion that methadone has any efficacy in any neuropathic pain condition. In the absence of any supporting evidence, it should probably not be recommended as a first‐line treatment, or even as a first‐line opioid.

For funders

There is insufficient evidence to support or refute the suggestion that methadone has any efficacy in any neuropathic pain condition. In the absence of any supporting evidence, it should probably not be recommended as a first‐line treatment, or even as a first‐line opioid. Its only definite advantage over other pharmaceutically manipulated long‐acting opioids is its lower cost.

Implications for research.

General implications

This review highlights the lack of high‐quality evidence investigating the use of methadone for neuropathic pain. In particular, safety issues such as respiratory depression, cardiac arrhythmias, and addiction have not been adequately addressed.

Design

While well‐designed, large (at least 200 participants in each arm), long‐duration (at least 12 weeks), randomised, controlled studies would be highly desirable, this review highlights the fact that only two RCTs have been completed since 2003, from which we have only the results of one, perhaps reflecting methadone's generic status. Future studies should enrol participants with various neuropathic diagnoses and report data for each diagnosis separately. If possible, dose‐ranging studies would also help to clarify differences in efficacy and safety between high‐ and low‐dose regimens.

Measurement (endpoints)

Future studies should also focus on patient‐centred outcomes, such as those of at least a 30% and at least a 50% reduction in pain intensity over baseline at the end of a trial.

What's new

Date	Event	Description
27 January 2021	Review declared as stable	See Published notes.

History

Protocol first published: Issue 1, 2017
Review first published: Issue 5, 2017

Date	Event	Description
11 January 2019	Amended	Contact details updated.
18 December 2018	Review declared as stable	See Published notes.

Notes

Assessed for updating in 2018

A restricted search in December 2018 did not identify any potentially relevant studies. Therefore, this review has now been stabilised following discussion with the authors and editors. If appropriate, we will update the review if new evidence likely to change the conclusions is published, or if standards change substantially which necessitate major revisions.

Assessed for updating in 2021

At January 2021, we are not aware of any potentially relevant studies likely to change the conclusions. Therefore, this review has now been stabilised following discussion with the authors and editors. The review will be reassessed for updating in two years. If appropriate, we will update the review before this date if new evidence likely to change the conclusions is published, or if standards change substantially which necessitate major revisions.

Appendices

Appendix 1. Methodological considerations for chronic pain

There have been several changes in how the efficacy of conventional and unconventional treatments is assessed in chronic painful conditions. The outcomes are now better defined, particularly with new criteria for what constitutes moderate or substantial benefit (Dworkin 2008); older trials may only report participants with 'any improvement'. Newer trials tend to be larger, avoiding problems from the random play of chance. Newer trials also tend to be of longer duration, up to 12 weeks, and longer trials provide a more rigorous and valid assessment of efficacy in chronic conditions. New standards have evolved for assessing efficacy in neuropathic pain, and we are now applying stricter criteria for the inclusion of trials and assessment of outcomes, and are more aware of problems that may affect our overall assessment. We summarised some of the recent insights that must be considered in this new review below.

1. Pain results tend to have a U‐shaped distribution rather than a bell‐shaped distribution. This is true in acute pain (Moore 2011a; Moore 2011b), back pain (Moore 2010d), arthritis (Moore 2010c), and fibromyalgia (Straube 2010); in all cases, mean results usually describe the experience of almost no‐one in the trial. Data expressed as means are potentially misleading, unless they can be proven to be suitable.

2. As a consequence, we have to depend on dichotomous results (the person either has or does not have the outcome) usually from pain changes or patient global assessments. The Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) group has helped with their definitions of minimal, moderate, and substantial improvement (Dworkin 2008). In arthritis, trials of less than 12 weeks' duration, and especially those shorter than eight weeks, overestimate the effect of treatment (Moore 2010c); the effect is particularly strong for less‐effective analgesics, and this may also be relevant in neuropathic‐type pain.

3. The proportion of people with at least moderate benefit can be small, even with an effective medicine, falling from 60% with an effective medicine in arthritis to 30% in fibromyalgia (Moore 2009; Moore 2010c; Moore 2013a; Moore 2014b; Straube 2008; Sultan 2008). One Cochrane Review of pregabalin in neuropathic pain and fibromyalgia demonstrated different response rates for different types of chronic pain (higher in diabetic neuropathy and postherpetic neuralgia and lower in central pain and fibromyalgia) (Moore 2009). This indicates that different neuropathic pain conditions should be treated separately from one another, and that pooling should not be done unless there are good grounds for doing so.

4. Individual participant analyses indicate that people who get good pain relief (moderate or better) have major benefits in many other outcomes, affecting quality of life in a significant way (Moore 2010b; Moore 2014a).

5. Imputation methods such as last observation carried forward, used when participants withdraw from clinical trials, can overstate drug efficacy especially when adverse event withdrawals with drug are greater than those with placebo (Moore 2012).

Appendix 2. Search strategy for CENTRAL via CRSO

#1 MESH DESCRIPTOR Methadone EXPLODE ALL TREES

#2 ((methadon* or d‐methadone or l‐methadone or r‐methadone or s‐methadone or dolophine or phenadone or physeptone or phymet or symoron or metadol or metasedin or methaddict or methadose or methex or pinadone or amidone or biodone)):TI,AB,KY

#3 #1 OR #2

#4 MESH DESCRIPTOR Pain EXPLODE ALL TREES

#5 MESH DESCRIPTOR Peripheral Nervous System Diseases EXPLODE ALL TREES

#6 MESH DESCRIPTOR Somatosensory Disorders EXPLODE ALL TREES

#7 (((pain* or discomfort*) adj10 (central or complex or nerv* or neuralg* or neuropath*))):TI,AB,KY

#8 (((neur* or nerv*) adj6 (compress* or damag*))):TI,AB,KY

#9 #4 OR #5 OR #6 OR #7 OR #8

#10 #3 AND #9

Appendix 3. Search strategy for MEDLINE (via Ovid)

1 exp Methadone/
2 (methadon* or d‐methadone or l‐methadone or r‐methadone or s‐methadone or dolophine or phenadone or physeptone or phymet or symoron or metadol or metasedin or methaddict or methadose or methex or pinadone or amidone or biodone).tw. 
3 1 or 2
4 exp Pain/
5 exp Peripheral Nervous System Diseases/
6 exp Somatosensory Disorders/
7 ((pain* or discomfort*) adj10 (central or complex or nerv* or neuralg* or neuropath*)).mp. [mp=title, abstract, original title, name of substance word, subject heading word, keyword heading word, protocol supplementary concept word, rare disease supplementary concept word, unique identifier]
8 ((neur* or nerv*) adj6 (compress* or damag*)).mp. 
9 or/4‐8
10 3 and 9
11 randomized controlled trial.pt. 
12 controlled clinical trial.pt.
13 randomized.ab.
14 placebo.ab.
15 drug therapy.fs.
16 randomly.ab.
17 trial.ab.
18 groups.ab.
19 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18
20 exp animals/ not humans.sh. 
21 19 not 20 
22 10 and 21

Appendix 4. Search strategy for Embase via Ovid

1 exp Methadone/

2 (methadon* or d‐methadone or l‐methadone or r‐methadone or s‐methadone or dolophine or phenadone or physeptone or phymet or symoron or metadol or metasedin or methaddict or methadose or methex or pinadone or amidone or biodone).tw.

3 1 or 2

4 exp Pain/

5 exp Peripheral Neuropathy/

6 exp Somatosensory Disorder/

7 ((pain* or discomfort*) adj10 (central or complex or nerv* or neuralg* or neuropath*)).mp. [mp=title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading]

8 ((neur* or nerv*) adj6 (compress* or damag*)).mp.

9 or/4‐8

10 3 and 9

11 random$.tw.

12 factorial$.tw.

13 crossover$.tw.

14 cross over$.tw.

15 cross‐over$.tw.

16 placebo$.tw.

17 (doubl$ adj blind$).tw.

18 (singl$ adj blind$).tw.

19 assign$.tw.

20 allocat$.tw.

21 volunteer$.tw.

22 Crossover Procedure/

23 double‐blind procedure.tw.

24 Randomized Controlled Trial/

25 Single Blind Procedure/

26 or/11‐25

27 (animal/ or nonhuman/) not human/

28 26 not 27

29 10 and 28

Appendix 5. GRADE: criteria for assigning grade of evidence

The GRADE system uses the following criteria for assigning a quality level to a body of evidence (Chapter 12, Higgins 2011).

1. High: randomised trials; or double‐upgraded observational studies.

2. Moderate: downgraded randomised trials; or upgraded observational studies.

3. Low: double‐downgraded randomised trials; or observational studies.

4. Very low: triple‐downgraded randomised trials; or downgraded observational studies; or case series/case reports.

Factors that may decrease the quality level of a body of evidence are:

1. limitations in the design and implementation of available studies suggesting high likelihood of bias;

2. indirectness of evidence (indirect population, intervention, control, outcomes);

3. unexplained heterogeneity or inconsistency of results (including problems with subgroup analyses);

4. imprecision of results (wide confidence intervals);

5. high probability of publication bias.

Factors that may increase the quality level of a body of evidence are:

1. large magnitude of effect;

2. all plausible confounding would reduce a demonstrated effect or suggest a spurious effect when results show no effect;

3. dose‐response gradient.

Appendix 6. Summary of outcomes: efficacy

Study	Treatment	Pain outcome	Other efficacy outcome
Morley 2003	Phase I (20 days): methadone orally: 5 mg twice daily alternating with placebo on odd days and rest on even days (n = 19). Phase II (20 days): methadone orally: 10 mg twice daily alternating with placebo on odd days and rest on even days (n = 17).	Pain intensity or relief (0‐100), methadone vs placebo (mean ± SD), as reported in manuscript (review author analysis differed): Phase I Maximum: ‐4.71 ± 10.1 (P = 0.065). Mean: ‐3.20 ± 8.23 (P = 0.117). Pain relief: 5.07 ± 10.83 (P = 0.064). Phase II Maximum: ‐9.25 ± 10.14 (P = 0.013). Mean: ‐6.56 ± 7.86 (P = 0.020). Pain relief: 9.07 ± 10.28 (P = 0.015). Use of additional analgesics assessed but not presented. Neuropathic pain scale scores presented but not analyzed.	‐
Raja 2002	All medications orally; titrated to effect. Approximately 8 weeks' total treatment. Methadone 5‐80 mg/day. Morphine 15‐240 mg/day. Mean titration duration for both opioids = 4.4 weeks. TCA 10‐160 mg/day. Mean titration duration = 4.1 weeks. Placebo 1‐16 tablets/day. Mean titration duration = 3.6 weeks.	Change (95% CI) in pain intensity (0‐10 NRS) during each treatment period: morphine ‐2.2 (‐2.7 to ‐1.6, n = 38); methadone ‐1.2 (‐1.8 to ‐0.5, n = 26); placebo ‐0.2 (‐0.7 to 0.2, n = 56); TCA ‐1.4 (‐1.8 to ‐0.9, n = 59). Pain relief from predrug baseline levels. Treatment responders (33% reduction in pain intensity from baseline). No data presented separately for methadone.	Symbol substitution task from the Wechsler Adult Intelligence Scale ‐ Revised. The Hopkins Verbal Learning Test. The grooved pegboard task. Physical functioning; sleep; treatment preference; and mood. No data presented separately for methadone.
Teixeira 2013	Methadone 5 mg orally twice daily × 3 weeks. Placebo Twice daily orally × 3 weeks. 3‐week washout between interventions.	≥ 30% improvement in VAS (0‐100) pain intensity: methadone 7/10; placebo 5/10. Improvement in categorical scale pain intensity (mild to severe): methadone 10/10; placebo 5/10. ≥ 30% improvement in evoked pain (0‐100): methadone 10/10; placebo 5/10.	≥ 3 decreases in subscores of MPQ: Sensitive: methadone 7/10; placebo 3/10. Affective: methadone 6/10; placebo 7/10. Evaluative: methadone 6/10; placebo 8/10. Miscellaneous: methadone 4/10; placebo 8/10. Daily activities scale from the BPQ (normal = 0, to abolished = 2). No data presented; described as not significantly changed after active treatment.

BPQ: Brief Pain Questionnaire; CI: confidence interval; MPQ: McGill Pain Questionnaire; n: number of participants; NRS: numeric rating scale; SD: standard deviation; TCA: tricyclic antidepressant (nortriptyline or desipramine); VAS: visual analogue scale.

Appendix 7. Summary of outcomes: adverse events and withdrawals

Study	Treatment	Adverse events	Withdrawals
Morley 2003	Phase I (20 days): methadone orally: 5 mg twice daily alternating with placebo on odd days and rest on even days (n = 19). Phase II (20 days): methadone orally: 10 mg twice daily alternating with placebo on odd days and rest on even days (n = 17).	Any AE: Phase I Methadone: 15/19. Placebo: 7/19. Phase II Methadone: 11/17. Placebo: 8/17. SAE: none reported. Deaths: none reported. Specific AE occurring in ≥ 10% participants: Phase I Nausea: Methadone 7/19; placebo 4/19. Vomiting: Methadone 4/19; placebo 1/19. Somnolence: Methadone 2/19; placebo 2/19. Dizziness: Methadone 6/19; placebo 0/19. Constipation: Methadone 2/19; placebo 1/19. Pruritus: Methadone 2/19; placebo 0/19. Headache: Methadone 5/19; placebo 2/19. Sweating: Methadone 2/19; placebo 0/19. Phase II Nausea: Methadone 8/17; placebo 4/17. Somnolence: Methadone 3/17; placebo 2/17. Dizziness: Methadone 3/17; placebo 1/17. Constipation: Methadone 3/17; placebo 1/17. Pruritus: Methadone 2/17; placebo 1/17. Diarrhoea: Methadone 2/17; placebo 0/17. Sweating: Methadone 3/17; placebo 1/17. AEs reported as mild to moderate in participants who completed the trial; severe in participants who dropped out.	AE: Phase I Methadone: 1/19. Placebo: 0/19. Phase II Methadone: 3/17. Placebo: 3/17. LoE: Phase I Methadone: 0/19. Placebo: 0/19. Phase II Methadone: 0/17. Placebo: 0/17.
Raja 2002	All medications orally; titrated to effect. Approximately 8 weeks' total treatment. Methadone 5‐80 mg/day. Morphine 15‐240 mg/day. Mean titration duration for both opioids = 4.4 weeks. TCA 10‐160 mg/day. Mean titration duration = 4.1 weeks. Placebo 1‐16 tablets/day. Mean titration duration = 3.6 weeks.	AEs in participants receiving methadone not reported separately. No SAEs or deaths reported.	Data not presented for methadone separately.
Teixeira 2013	Methadone 5 mg orally twice daily × 3 weeks. Placebo Twice daily orally × 3 weeks. 3‐week washout between interventions.	"The frequency of reported adverse events, such as constipation, nausea and dizziness, did not differ significantly in either treatment period." No data presented. No SAEs or deaths reported.	No dropouts.

AE: adverse event; LoE: lack of efficacy; n: number of participants; SAE: serious adverse event.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Morley 2003.

Study characteristics
Methods	Double‐blind, randomised, cross‐over, 2‐phase study, with inter‐ and intra‐participant comparisons. Both phases 20 days' duration.
Participants	Type of neuropathic pain: mixed, non‐malignant, with duration > 3 months. Pain had not been relieved by other analgesics. Participants could continue current analgesic regimen during study. Low‐dose phase Entered/completing: 19/18. Age (mean, SD): 59.6 ± 13.1 years. Sex (male, %): 13, 68%. High‐dose phase Entered/completing: 17/11. Age (mean, SD): 58.4 ± 13.3 years. Sex (male, %): 12, 71%.
Interventions	Low‐dose phase: methadone orally: 5 mg twice daily alternating with placebo on odd days and rest on even days. 20 days' duration. High‐dose phase: methadone orally: 10 mg twice daily alternating with placebo on odd days and rest on even days. 20 days' duration. In each phase of the trial, the randomization was such that dosing of methadone could occur on any of 5 non‐consecutive days over a 20‐day period.
Outcomes	Primary: VAS maximum pain intensity; mean pain intensity; pain relief. Participants recorded each outcome every evening (at least 3 hours after taking their last dose of either methadone or placebo on odd numbered days, and at the same time on even numbered 'rest' days). Secondary: Neuropathic pain scale of Galer and Jensen (Galer 1997, assessed at baseline and the end of each phase); adverse events (incidence and severity, recorded in participant diary); requirement for additional analgesics (participant diary).
Source of funding	Stanley Thomas Johnson Foundation, Berne, Switzerland.
Were treatment groups comparable at baseline?	NA: cross‐over study.
Notes	Participants had neuropathic pain that had not been satisfactorily relieved by other interventions or by current or previous drug regimens. No criterion for pain intensity at baseline; participants' baseline pain intensity ranged from 37/100 to 98/100. For each phase, results of 5 days with active intervention were compared with results of 5 days with placebo. Participants were permitted to continue with concurrent medications, some of which were opioids.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	8 replications of a Latin square design.
Allocation concealment (selection bias)	Low risk	Central allocation.
Blinding (performance bias and detection bias) All outcomes	Low risk	Medication containers appeared identical and medications "were not distinguishable by taste or appearance."
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Only participants completing study were analyzed. Numbers of participants withdrawing due to adverse events were similar while receiving methadone (n = 4) or placebo (n = 3).
Selective reporting (reporting bias)	Unclear risk	Severity of adverse effects and use of additional analgesics assessed but not presented. Neuropathic pain scale scores presented but not analyzed.
Size	High risk	19 participants in cross‐over study.

Raja 2002.

Study characteristics
Methods	Randomised, double‐blind, placebo‐ and active‐controlled, 3‐phase cross‐over trial. Each treatment phase lasted approximately 8 weeks and had a titration, maintenance, and taper subphase. The treatment phases were separated by a 1‐week drug free, washout period.
Participants	Type of neuropathic pain: 76 participants (26 received methadone) with PHN persisting > 3 months after resolution of cutaneous lesions, with NRS > 4. Methadone group Entered/completing: 26/unclear. Age (mean, SD): not reported. Sex (male, %): not reported. Morphine group Entered/completing: 40/unclear. Age (mean, SD): not reported. Sex (male, %): not reported. Tricyclic antidepressant group Entered/completing: 59/52. Age (mean, SD): not reported. Sex (male, %): not reported. Placebo group Entered/completing: 56/55. Age (mean, SD): not reported. Sex (male, %): not reported.
Interventions	All medications orally. Methadone Variable dose, 5‐80 mg/day, based on titration to effect. Mean (± SD) daily dose = 15 ± 2 mg. Mean titration duration for both opioids (methadone and morphine) = 4.4 weeks. Morphine Variable dose, 15‐240 mg/day, based on titration to effect. Mean (± SD) daily dose = 91 ± 49.3 mg (range 15 mg to 225 mg). Tricyclic antidepressant Variable dose, 10‐160 mg/day, based on titration to effect. Mean (± SD) daily dose = nortriptyline 89 ± 27.1 mg (range 40 mg to 140 mg); desipramine 63 ± 3.6 mg. Mean titration duration = 4.1 weeks. Placebo Variable dose, 1‐16 tablets/day, based on titration to effect. Mean number of tablets daily = 9.4 (range 2 to 16). Mean titration duration = 3.6 weeks.
Outcomes	Primary: Pain intensity. Mean overall pain during previous 24 hours (0 to 10 NRS: 0 = no pain, 10 = most intense pain imaginable). Change in pain intensity during each treatment period calculated as the difference between the mean of the last 3 ratings during the treatment phase, typically made during the end of the maintenance period, and the mean of 3 baseline ratings obtained prior to the initiation of the drug treatment. Baseline ratings obtained over a 9‐ to 10‐day period and included the last pain rating at the end of the drug taper period and the 2 pain ratings obtained during the drug‐free baseline period. Pain relief from predrug baseline levels over the last 24 hours (0 to 100% NRS: 0% = no relief, 100% = complete relief). Pain relief for each treatment period calculated as mean of the 3 final ratings during the maintenance phase or, if the participant withdrew earlier, mean of the last 3 available relief ratings. Cognitive function. Concentration and psychomotor function: symbol substitution task from the Wechsler Adult Intelligence Scale ‐ Revised. Verbal learning: The Hopkins Verbal Learning Test. Manual dexterity and psychomotor speed: the grooved pegboard task. Secondary: Physical functioning; sleep; mood; adverse effects (assessed as mild, moderate, or severe); treatment preference. Pain intensity and pain relief values were collected by twice‐weekly telephone interviews during the trial. All other outcome measures were obtained during clinic visits at the end of the drug‐free baseline period and at the end of the maintenance phase for each drug.
Source of funding	Supported by NIH grant no. NS 32386 and GCRC grant no. RR0052.
Were treatment groups comparable at baseline?	N/A: cross‐over study
Notes	Study compared opioid (morphine or methadone) vs tricyclic antidepressant (nortriptyline or desipramine) vs placebo. Participants received methadone only if they did not tolerate morphine. Separate demographic data (morphine or methadone) not reported. Separate outcome data only presented for 1 outcome: reduction in pain (baseline to end of maintenance period). Previous medications for PHN: tricyclic antidepressants (n = 14), short‐acting opioids (n = 15), or both (n = 8). All prescribed pain medications were discontinued at least 1 week prior to enrolment. NSAIDs and paracetamol (acetaminophen) were permitted during the study for analgesic rescue.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"The randomization sequence was computer generated by the biostatistician."
Allocation concealment (selection bias)	Low risk	"The randomization sequence was… provided in sealed envelopes to the pharmacist and the monitoring committee. Study medications were mailed directly to the patients by the pharmacy in a box marked Study Drug A, B, or C, depending on the treatment period."
Blinding (performance bias and detection bias) All outcomes	Low risk	"The pharmacist formulated the study drugs in identical gel capsules to maintain the blinding. All investigators were blinded to the drug treatments during the study."
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Intention‐to‐treat analysis employed. For participants who did not complete a treatment period, the last 3 available pain ratings were used. No data were imputed. Study authors reported that intention‐to‐treat analysis included the relatively high pain scores of participants who dropped out prior to completion of the drug titration phase due to adverse effects and thus likely provided a conservative estimate of the effectiveness of opioids. Number of participants who did not complete methadone phase not reported separately.
Selective reporting (reporting bias)	Unclear risk	All outcomes described in methods section of the report reported adequately in results section. Participants were permitted to take non‐prescription NSAIDs and paracetamol during the study for analgesic rescue, but this did not appear to have been assessed as an outcome and was not reported in the results section of the report.
Size	High risk	Cross‐over, with only 26 participants receiving methadone.

Teixeira 2013.

Study characteristics
Methods	Randomised, placebo‐controlled cross‐over trial. 2 phases, each 3 weeks' duration, with 3‐week washout period in between phases.
Participants	Type of neuropathic pain: PHN ≥ 6 months' duration, VAS > 40/100, and opioid naive. Mean (± SD) duration of pain 41 ± 19 months. Methadone phase Entered/completing: 10/10. Age (mean, SD): 71 ± 21 years. Sex (male, %): 4, 40%. Placebo phase Entered/completing: 10/10. Age (mean, SD): 71 ± 21 years. Sex (male, %): 4, 40%.
Interventions	Methadone 5 mg orally twice daily × 3 weeks. Placebo As with methadone. Participants had 3‐week washout between interventions.
Outcomes	Primary: Spontaneous pain intensity: VAS and categorical scale (mild, moderate, and severe); evoked pain (dynamic mechanical allodynia intensity in the painful area using a standardised brush at a speed of 2 cm/second and covering a 6‐cm distance inside the PHN painful area) scored as none (= 0) to severe (= 3) after 3 strokes. Secondary: Daily activities scale from the Brief Pain Questionnaire (normal = 0 to abolished = 2); McGill Pain Questionnaire; and adverse events assessed by directly questioning participants on the presence of new symptoms presenting during treatment.
Source of funding	Not reported.
Were treatment groups comparable at baseline?	N/A: cross‐over study.
Notes	Participants had failed first‐ and second‐line treatments (tricyclic antidepressants, venlafaxine, and gabapentinoids) and had indications for adding an opioid to their current regimen. Participants could remain on current regimen.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Not described.
Allocation concealment (selection bias)	Unclear risk	Not mentioned.
Blinding (performance bias and detection bias) All outcomes	Low risk	"The methadone and placebo pills looked identical."
Incomplete outcome data (attrition bias) All outcomes	Low risk	No participants left the study. Appeared that data were reported for all participants at all time points.
Selective reporting (reporting bias)	High risk	Several assessed outcomes not reported in results section of report. No data reported for adverse events. Correlation analysis performed that was not mentioned in methods section of report.
Size	High risk	10 participants in each phase.

n: number of participants; N/A: not applicable; NSAID: nonsteroidal anti‐inflammatory drug; NRS: numeric rating scale; PHN: postherpetic neuralgia; SD: standard deviation; VAS: visual analogue scale.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Haumann 2016	Not blinded.

Characteristics of studies awaiting classification [ordered by study ID]

NCT02233452.

Methods	Randomised, double‐blind, parallel, active‐controlled trial.
Participants	People with neuropathic pain for > 6 months, poorly responsive to drugs used to treat neuropathic pain (i.e. opioid, non‐opioid, anticonvulsants, antidepressants), aged 22‐77 years.
Interventions	Oral methadone (n = 13), ketamine (n =13), or methadone + ketamine (n =13). Doses not specified. Duration of study: 3 months.
Outcomes	Primary: pain intensity (VAS) assessed at baseline, and 8, 15, 30, 60, and 90 days after beginning treatment. Secondary: symptoms of neuropathic pain such as allodynia, burning or shooting pain; adverse events.
Notes	Study completed, but results not posted on trial site. Does not appear to be a corresponding journal publication.

n: number of participants; VAS: visual analogue scale.

Characteristics of ongoing studies [ordered by study ID]

NCT01205516.

Study name	Methadone in Neuropathic Pain.
Methods	Double‐blind, randomised, active‐controlled, 3‐centre trial.
Participants	Adults with neuropathic pain of central or peripheral origin for ≥ 3 months, and a score of ≥ 4/10 on the DN4 (Douleur Neuropathique 4). Estimated enrolment: 180.
Interventions	Methadone: 2.5 mg tablets, 1‐12 tablets taken every 12 hours (5‐60 mg/day). Morphine controlled‐release: 10 mg tablets, 1‐12 tablets taken every 12 hours (20‐240 mg/day). 5‐week titration phase to adequate pain relief without unacceptable levels of adverse effects. 6‐week dose phase, then a 4‐week tapering off phase.
Outcomes	Not specified.
Starting date	January 2013.
Contact information	Mary E Lynch, MD; 902‐473‐6428; mary.lynch@dal.ca.
Notes	Estimated study completion date: December 2017.

Contributions of authors

Searched for studies: EM, MF, RS.

Obtained copies of studies: EM.

Selected which studies to include (two plus one arbiter): EM, MF, RS.

Extracted data from studies: EM, MF, RS.

Entered data into Review Manager 5: EM, MF.

Carried out the analysis: EM.

Interpreted the analysis: EM, MF, RS.

Drafted the final review: EM, MF.

Edited the final review: EM, MF, RS.

Sources of support

Internal sources

• Saltonstall Fund for Pain Research, USA

External sources

• The National Institute for Health Research (NIHR), UK

  NIHR Cochrane Programme Grant: 13/89/29 ‐ Addressing the unmet need of chronic pain: providing the evidence for treatments of pain

Declarations of interest

EM: none known; EM is a pharmacist with a Master's degree in Pain Research, Education and Policy, and is involved in the pharmacological management of patients with chronic pain issues in a tertiary setting.

MF: none known.

RS: none known; RS is an anaesthesiologist with experience and expertise in acute pain management, including managing patients with chronic pain undergoing surgical procedures.

Stable (no update expected for reasons given in 'What's new')