Lidocaine is a local anesthetic of the amide type also used systemically as an antiarrhythmic drug. Early reports described the use of IV lidocaine or procaine to relieve cancer and postoperative pain (1–4). Subsequent experimental studies indicated that lidocaine produces analgesia by blockade of peripheral and central sodium ion gate channels, including in the spinal dorsal horn (5), and if given IV can alleviate deafferentation pain or central pain (6). More recent research has suggested that the antinociceptive action of lidocaine is more complex and that inhibition of neuronal ectopic discharges is only one of several mechanisms involved (7). Mao and Chen (8) suggest that lidocaine administration for neuropathic pain has regained much attention over the past decade for two reasons: first, animal and laboratory data have provided a plausible mechanism by which lidocaine might be clinically effective for neuropathic pain; and second, oral analogues of local anesthetics have been developed to permit chronic maintenance therapy. The International Association for the Study of Pain defines neuropathic pain as “pain resulting from damage to the peripheral or central nervous system” (9). Neuropathic pain has several key clinical characteristics: variable onset after injury, lancinating and spontaneous character, association with allodynia and hyperesthesia, and evoked summation and hyperpathia (10).
Neuropathic pain may be attributable to primary afferent nociceptive or non-nociceptive input or to spontaneous ectopic discharges without activation of peripheral nociceptors (8,11). Cell membranes of injured peripheral nerves express sodium channels with unusual density and produce persistent spontaneous discharges that maintain a central hyperexcitable state (12). Ectopic discharges can be initiated along the injured nerve, in the dorsal root ganglion, and in peripheral neuromata (13–17). Lidocaine inhibits these aberrant electrical discharges at concentrations well below those necessary to produce conduction blockade in nerves. Studies in animal preparations clearly indicate that systemically administered lidocaine can silence ectopic discharges without blocking nerve conduction (11,14,16,18,19). Further studies have shown that the systemic administration of local anesthetics provides clinical analgesia in a broad range of neuropathic pain states (8,11,20–25).
Clinical trials testing lidocaine or its oral analogues for neuropathic pain have, in general, enrolled few patients, used varied dosage regimens, had different experimental designs, and assessed end-points across diverse time periods in patients with neuropathic pain of different etiologies. The result is a lack of clear definition on the role of these drugs in the treatment of neuropathic pain. To consolidate this sometimes contradictory literature, we have critically evaluated and quantitatively synthesized the existing data on the effect of local anesthetics administered systemically to relieve neuropathic pain.
Our objectives were 1) To determine if systemically administered local anesthetics relieve neuropathic pain and to quantify the degree of pain relief compared with other control interventions (placebo and active control drugs); 2) To examine whether particular clinical subpopulations benefit more from this therapy; and 3) To assess the safety of these agents.
We performed a sensitive search (26) to identify published trials with parallel or crossover design in patients with neuropathic pain of any etiology. We combined a series of search terms relevant to randomized double-blind, placebo-controlled trials with pain-specific terms and with the subject headings “lidocaine,” “lignocaine,” “mexiletine,” “flecainide,” and “tocainide.” The search strategy was adapted to each of the following databases: The Cochrane Central Register of Controlled Trials (Software Update, version 4, 2000); EMBASE (January 1980 to December 2002); MEDLINE (January 1966 to December 2002); CancerLit (1963 to October 2002); LILACS; and the System for Information on Grey Literature in Europe (SIGLE). This latter database was used to search for conference proceedings. We placed no language restriction on the search. We also reviewed the reference sections of retrieved articles to identify additional randomized controlled trials. We contacted investigators in this field for assistance in identifying other published or unpublished trials or to request additional data on published trials.
Three reviewers independently screened all titles and abstracts identified in the literature search. The titles and abstracts were not blinded as to author names, affiliated institutions, journal of publication, or study results. Identified trials were then examined for internal validity using the Oxford Quality Score criteria as described by Jadad et al. (27) and examined for other methodological bias. The reviewers resolved any disagreement by discussion.
We extracted data on patient characteristics, study design methods, interventions, outcome measurements, and adverse events from the original papers. The outcomes and instruments to measure them varied across studies. Outcome measurements were categorized as binary or continuous. For continuous data, a change in the 0-100 mm Visual Analog Scale (VAS) ratings of pain intensity was the most common pain-related outcome (28–54). When possible, we linearly transformed other scales or categorical assessments so as to map into this scale. Continuous data included medians, means with standard deviations, or sem. Because medians do not measure variance we did not analyze medians. If no sd was presented, we calculated it from sem and sample size. Adverse events were recorded as presented and defined by original authors; no distinction was made as to their severity or to whether they were major or minor, and no judgment was made regarding drug causation. We combined the aggregate, continuous data on pain response and dichotomous data on adverse events to obtain a pooled effect size for each outcome.
We analyzed the data collected with MetaView (version 4.1), the companion module for RevMan 4.1.1 (both available at www.cochrane.org). We calculated the weighted mean differences (WMD) between placebo control or active control and treatment groups. Every analysis was assessed for statistical heterogeneity using the χ2 and I2 tests. All data were combined with a random effects model which incorporates both between-studies and within-study heterogeneity. We calculated the odds ratio (OR) for adverse events.
Our literature search identified 1902 potentially relevant titles. Forty-one trials of systemically administered local anesthetic-type drugs met inclusion criteria. Fourteen trials were excluded: two were duplicate publications (55,56); one study of flecainide was terminated when the drug was removed from the market (57); five trials examined the use of IV lidocaine in experimentally induced acute pain in normal volunteers (58–62); three trials did not have or describe random allocation (63–65); two trials were unblinded (66,67); and one study was single blinded, without random allocation (1). Twenty-seven randomized, double-blind, controlled clinical trials for chronic neuropathic pain were included (28–54): 13 used lidocaine; 12 used mexiletine; one tested lidocaine and mexiletine sequentially (37); and one used tocainide (41). Eight studies were parallel trials. Nineteen had a crossover design. Two of the 19 trials did not specify washout periods (41,42). Three randomized studies appeared as abstracts (29,43,44). We retrieved complete information from one of these trials by contacting the author (29). The age (mean ± sd) of the participants in all included trials was 51.7 ± 10.3 yr. We expected that trials enrolling participants with different diseases despite having overlapping etiological features (e.g., poststroke and spinal cord injury pain, or causalgia/reflex sympathetic dystrophy and diabetic polyneuropathy) would contribute to heterogeneity (Table 1).
Eleven clinical trials (41%) were of good methodological quality, scoring four points (28,29,31–33,45,52) or five points (36,40,53,54) on the Jadad scale (27). Sixteen trials (59%) scored two points (38,43,44) or three points (30,34,35,37,39,41,42,46–51). The median score was three for all trials with either lidocaine or its oral analogues. Of the 27 trials included in this review, seven (26%) described a method for random allocation (29,31,32,36,40,53,54); six (22%) included sample size calculations (33,36,40,45,53,54), and 10 (37%) described some method to check that blinding was effective (28,29,31,35,36,40,45,46,52,54). The sample sizes of the arms receiving lidocaine or oral analogues varied but in general were small; the median size for all trials was 28 (range, 8-87).
Meta-Analysis of Relief of Spontaneous Pain with IV Lidocaine or Oral Mexiletine Versus Placebo.
All placebo-controlled trials involving lidocaine and mexiletine that provided continuous data (means and sd or sem) on pain relief were eligible for meta-analysis (Fig. 1). For trials using more than one dose of lidocaine or mexiletine, the largest dose was selected. For trials measuring pain at different times, the last measurement time was most commonly chosen. From the trial reported by Bruera et al. (31) we pooled the data from all time points because of the unusually large number of measurements; however, the negative findings of this study would not have been affected were we to have used data from any single time point. We pooled daytime and nocturnal pain scores reported in one trial (46), and for a trial on postamputation pain evaluating stump and phantom pain, we chose stump pain (54).
Pretreatment and posttreatment mean pain scores were available from 10 lidocaine and 9 mexiletine trials (n = 706), for a total of 349 patients receiving the treatment drug and 357 patients receiving placebo. The summary effect size for lidocaine and mexiletine pooled together indicates that both drugs were superior to placebo in relieving chronic neuropathic pain (WMD = −10.60 mm; 95% CI: −14.52 to −6.68 mm; P < 0.00001). Both drugs had similar efficacy in decreasing VAS pain scores: in the lidocaine trials included for meta-analysis 165 patients received lidocaine and 164 patients were treated with placebo. Lidocaine was superior to placebo (WMD = −10.02 mm; 95% CI: −16.51 to −3.54 mm, P = 0.002); in the mexiletine trials, 184 patients received mexiletine and 193 patients received placebo. Two trials had a wide dispersion of data around the mean (36,48). The combined effect size also favored mexiletine over placebo (WMD = −10.97 mm; 95% CI: −16.14 to −5.80 mm, P = 0.0001).
Meta-Analysis of Either Lidocaine or Mexiletine Versus Other Active Treatments.
Five trials (n = 206: 102 treated with lidocaine or analogues, 104 treated with an active control) compared the analgesia obtained with local anesthetic-type drugs with that obtained with other active controls: carbamazepine (41), gabapentin (36), amantadine (45), and morphine (47,54) (Fig. 2). There was no evidence of heterogeneity (χ2 = 2.16; df = 4; P = 0.71) and no evidence that these drugs were better than lidocaine or oral congeners to relieve neuropathic pain (WMD = −0.60 mm; 95% CI: −6.96 to 5.75 mm).
Trials were categorized into 2 subgroups: fewer than 25 participants (n = 16, 6 mexiletine trials and 10 lidocaine trials); and more than 25 participants (n = 3, 2 mexiletine trials and one lidocaine trial). Despite the larger number of trials and the fact that nearly 40% were studies with mexiletine, the subgroup of studies with fewer than 25 patients was not statistically heterogeneous, the less numerous subgroup of studies that enrolled more than 25 patients was statistically heterogeneous.
Time of Outcome Measurement.
Stratification analysis indicated that the time of outcome assessment may contribute to the heterogeneity of the study findings. Although there was no indication of statistical heterogeneity in the subgroup with outcome measurements recorded up to 24 h (n = 9, all lidocaine trials), the second subgroup (outcome measurements recorded later than 24 h, n = 10, including 2 lidocaine trials) was borderline heterogeneous (P = 0.07). Further, exclusion of one trial (48) lessened the heterogeneity considerably. In fact, when all trials were combined to yield a global effect size, the statistical heterogeneity was borderline (χ2 = 27.1; df = 18; P = 0.08). We performed an analysis in which all trials with widely spread data were suppressed (30,36,48). Effect sizes for both subgroups separately or combined into one summary statistic showed no evidence of statistical heterogeneity (χ2 = 11.40; df = 15; P = 0.72).
Trials were classified according to design as either crossover or parallel trials. χ2 tests for heterogeneity suggested that the crossover trials were less heterogenous than the parallel trials (P = 0.29 and P = 0.02, respectively), largely because of three of the trials within the latter group (36,40,48). However, statistical tests such as χ2 are insensitive measures of heterogeneity (68), and, in fact, visual inspection of the forest plots for both types of trials disclosed heterogeneity in both. Therefore, trial design may be a source of heterogeneity.
We used the Jadad score (27) to categorize trials into 3 subgroups: between 2 to 3 points (low and fair quality, n = 8); 4 points (good quality, n = 8); and studies with 5 points (very good quality, n = 3). The reported estimates of efficacy did not differ significantly according to study quality.
All trials were stratified into six subgroups by etiology:
1) Peripheral, metabolic: five trials, of which four used mexiletine and one used lidocaine. The only disease was diabetic polyneuropathy.
2) Peripheral, infectious: three trials including human immunodeficiency virus (HIV)-1-related polyneuropathy treated with mexiletine (n = 1) and postherpetic neuralgia treated with lidocaine (n = 2).
3) Peripheral, posttraumatic: four trials with mexiletine and lidocaine (two each).
4) Peripheral, cancer: one trial using lidocaine.
5) Peripheral, mixed: two trials using lidocaine.
6) Central/mixed, vascular or posttraumatic: three trials included participants with pain resulting from postamputation, stroke, and spinal cord injury. Lidocaine was used in two of these trials, while mexiletine was used in one.
The subgroup with peripheral neuropathic pain of metabolic etiology showed heterogeneity. No conclusion can be drawn from subgroup 4 (peripheral, cancer), as there is only one trial published with data that could be included in the meta-analysis. These findings suggest that etiology is another important source of clinical and statistical heterogeneity.
The most common adverse events noted in the retrieved trials were nausea, vomiting, abdominal pain, diarrhea, dizziness, and perioral numbness. Less frequent side effects were metallic taste, tremor, dry mouth, insomnia, allergic reactions, and tachycardia. Serious adverse events, such as cardiac arrhythmias and hemodynamic instability, were notably absent from these trials. With mexiletine the most common adverse events were nausea and vomiting (17.24%), slightly more than the proportion of patients receiving lidocaine who presented with these symptoms (16.4%). Side effects referable to the central nervous system were more frequent in those patients who received lidocaine, with dizziness ranking first (29.5%). Instead, neurologic symptoms in the patients treated with mexiletine were far less frequent: for example, dizziness was reported in two patients (39,46), and tremor in two other patients (32,46).
Lidocaine and Oral Analogues Versus Placebo.
Twenty studies provided rates of adverse events for placebo and lidocaine or oral analogues (28–36,38–40,42,46–50,52,53) (Fig. 3). Three of these (31,33,38) do not appear in Figure 3 because they found no adverse events in participants exposed to the treatment drug or placebo. Of 749 participants, 410 were treated with lidocaine or mexiletine and 339 received placebo. One-hundred-thirty-two patients (32.2%) receiving the test drug experienced adverse events, compared with 39 patients (11.5%) given placebo (OR = 4.16, 95% CI: 2.68 to 6.46). There was no evidence of statistical heterogeneity (P = 0.98). These results indicate that treatment with lidocaine or mexiletine was associated with significantly more adverse events than placebo.
Lidocaine and Oral Analogues Versus Other Analgesics Used as Active Controls.
Because of possible safety issues related to using local anesthetics systemically for indications other than cardiac arrhythmias, we compared lidocaine and oral analogues with other active control drugs, with the aim of determining their relative safety in the treatment of neuropathic pain. Of 7 studies, 4 provided information on adverse events in 185 patients. Ninety-four of these were treated with lidocaine or its oral analogues. Ninety-one were treated with an active control, including carbamazepine, morphine, amitriptyline, and gabapentin (36,40,41,47) (Fig. 4). Of the 94 patients treated with lidocaine or its oral congeners, 27 had adverse events, compared with 22 of 91 treated with active control drugs (OR: 1.23; 95% CI: 0.22 to 6.90; P = 0.81). Active control drugs produced a slightly less, but nonsignificant, likelihood of adverse events than lidocaine or oral analogues. However, the results are limited because of the small number of trials with adequate information on this end-point, and the heterogeneity of the model (χ2 = 7.56; df = 3; P = 0.06).
Considerable debate now surrounds the young field of evidence-based medicine and many cautionary notes have been struck regarding the potential for misapplication of its methods (69). For example, meta-analysis of low quality reports may produce unreliable estimates of treatment effect (68,70,71).
Data from 19 randomized, placebo-controlled trials indicate that the analgesia resulting from the systemic administration of local anesthetics is superior to that from placebo. The main source of heterogeneity in the present meta-analysis was pain etiology. Other factors with the potential to produce heterogeneity, such as trial design, study quality, and time of outcome measurement, may have been present. We emphasize that the identification of heterogeneity in a meta-analysis does not necessarily undermine its conclusions.
The diverse etiologies of neuropathic pain may have a role in explaining heterogeneity, both clinical and statistical. Given the variability of etiologies, mechanisms, and presentations of neuropathic pain (72), an analogy can be drawn with headache as a symptom. The headache from a tumor may be mistaken at times for migraine or tension-type headache, and thus the symptomatic responses to identical treatments can differ. Lidocaine and its oral analog mexiletine relieved pain from diabetic polyneuropathy but were ineffective against plexopathy from tumor infiltration or HIV-related polyneuropathy. Though not conclusive, subgroup analyses to explore and explain heterogeneity suggest that neuropathic pain from diabetes, trauma, and cerebrovascular disease are more likely than pain of postinfectious etiology to respond. Systemic morphine and lidocaine were equally effective for postherpetic neuralgia (47) and postamputation stump pain (54) and both were superior to placebo. In the latter study, however, morphine was more effective than lidocaine for phantom pain (54). Tocainide (no longer marketed in the United States) appeared to be as effective as carbamazepine for trigeminal neuralgia (41). Future trials should restrict enrollment of patients to those with neuropathic pain of a single etiology to clarify the interpretation of their findings. Even for a single entity (e.g., HIV/acquired immune deficiency syndrome neuropathic pain) it is unclear how well tolerated lidocaine or mexiletine are in malnourished patients with active systemic illness, but it is plausible that cachectic patients may be at increased risk for adverse events.
Our analyses show that the systemic administration of sodium channel blocking drugs can relieve pain in selected patients with neuropathic pain and that this effect is superior to placebo. A more difficult question to answer is whether a mean difference of 11 mm on a 0-100 scale (or 1.1 on a 0-10 numeric scale) represents a true clinical difference for patients. The analysis of continuous data from pain scales using central tendency measures (means or medians) is a mathematical attempt to make a multidimensional, subjective variable, pain, more objective. A limitation of the use of mean pain scores is that individual responses may not follow a normal probability distribution but rather a bimodal pattern in which a mean difference of 11 mm can be a clinical difference for some patients (73). Even in the absence of a frankly bimodal distribution, individual patients may experience a larger response. For other public health issues, e.g., arterial blood pressure reduction, it is accepted that small mean changes in systolic blood pressure across large populations can translate into significant improved clinical outcomes (74). One solution to this problem is the use of binary data, expressing results as response rates. The definition of the smallest decline in pain intensity considered successful or clinically significant by the patient has been explored (75). Several studies analyzing data from large randomized clinical trials showed that a clinically meaningful difference begins around a 30% reduction in pain intensity, or a 2-point reduction from moderate-to-severe baseline pain intensity (0-10 scale) (73,76,77). We collected and analyzed the responder rates published in 14 of the mexiletine and lidocaine trials and found that these drugs were better than placebo (odds ratio: 2.04; 95% CI: 1.57 to 2.65, P < 0.00001, data not shown). This result is in agreement with the WMD between oral anesthetics and placebo, and further suggests that such a difference is clinically important in the context of chronic neuropathic pain.
The analgesic effects of lidocaine and mexiletine were similar. However, lidocaine’s short serum half-life of 120 min precludes the use of this drug for chronic use. Accordingly, pain relief with lidocaine has been measured within 24 h in all trials because in most patients the effect disappears a few hours after treatment. The mechanism of prolonged relief reported in animal models (78) and in some patients described in nonrandomized (6,79,80) and randomized studies (28,38,49) is still unknown. As discussed in a previous review (25), controlled studies are still required to determine the long-term effects of lidocaine on pain. A clinical approach might involve the chronic subcutaneous delivery of lidocaine via infusion pump but we are not aware of any randomized controlled studies using this technique. Despite its therapeutic limitations, lidocaine is likely to be used in future clinical trials as it is a prototypical blocking drug for tetrodotoxin-resistant sodium channels and one with which other novel, promising and more selective drugs can be compared.
In direct comparisons, the incidences of adverse events reported for lidocaine, tocainide, and mexiletine were similar to those for other active control drugs. Statistical heterogeneity, limited data, and diversity of drug classes in the comparison control groups limit external validity of the analysis of relative safety. However, indirect comparison of the incidences of adverse effects reported with lidocaine and mexiletine and those seen with gabapentin (a drug generally considered to have a good safety profile) in separate, controlled trials using similar pain models suggests a similar rate of occurrence (81,82). Across-trial differences in the definitions and methods to ascertain side effects precluded meta-analyses of the severity and nature of these adverse effects. It remains to be determined whether patients find the adverse effects experienced with lidocaine and oral analogues to be less tolerable than those seen with other drugs used for neuropathic pain.
The precise role of lidocaine and its oral analogues in controlling neuropathic pain of differing etiologies is not yet fully established because of the multifaceted nature of neuropathic pain, the statistical and clinical heterogeneity of many of the trials, and the uncertainty of results from small studies. Well-structured, high-quality trials with active placebos or active drug controls, conducted according to uniform, and hence combinable, designs are needed to clarify the efficacy of these drugs in the treatment of neuropathic pain from specific etiologies. Future trials should also explore subcutaneous infusions and move from mexiletine to testing newer oral analogues with higher therapeutic ratios. In addition, growing interest in topically applied local anesthetics to treat acute (83,84) and chronic (85) pain suggests that there is a need to compare the benefits and harms of systemic versus topical local anesthetics for the relief of neuropathic pain from diverse etiologies.
We thank Mrs. Frances Fairman of the Cochrane Collaborative Review Group on Pain, Palliative and Supportive Care, Pain Research Unit (Churchill Hospital and the University of Oxford, England) for conducting a literature search in EMBASE. Miroslav-Misha Backonja, MD, University of Wisconsin, Madison, WI, provided individual patient data for the study cited as Reference 29. Andrew Baranowski, MD, London, UK provided means and standard errors of the means of VAS pain scores for the study cited as Reference 30. Faye Chiou, MD, Baylor College of Medicine, Houston, TX, reviewed and verified the absence of adverse events from mexiletine in the study cited as Reference 33. Mark Wallace, MD, University of California, San Diego, CA, provided complete information on means, standard deviations of pain scores, and adverse events in the study cited as Reference 51. Per-Eric Lins, MD, and Per Oskarsson MD, Huddinge University Hospital, Karolinska Institute, Stockholm, Sweden, supplied mean pain scores and standard errors for daytime and nocturnal pain for the study cited as Reference 46. Prabhav Tella, MBBS, MS, Johns Hopkins University, Baltimore, MD, USA provided us with means and standard deviations of pretreatment and posttreatment pain scores for lidocaine, placebo, and morphine for the study cited as Reference 54. Trolsen Nielsen, MD, Aarhus University, Aarhus, Denmark, tried to assist in supplying additional information regarding study cited as Reference 35.
1. Keats AS, D’Alessandro GL, Beecher HK. A controlled study of pain relief by intravenous procaine. JAMA 1951;147:1761–3.
2. Gilbert CR, Hanson IR, Brown AB, Hingson RA. Intravenous use of Xylocaine. Curr Res Anesth Analg 1951;30:301–13.
3. De Clive-Lowe SG, Desmond J, North J. Intravenous lignocaine anesthesia. Anesthesiology 1958;13:138–46.
4. Bartlett EE, Hutaserani O. Xylocaine for the relief of postoperative pain. Anesth Analg 1961;40:296–304.
5. Woolf CJ, Wiesenfeld-Hallin Z. The systemic administration of local anaesthetics produces a selective depression of C-afferent fibre evoked activity in the spinal cord. Pain 1985;23:361–74.
6. Boas RA, Covino BG, Shahnarian A. Analgesic response to i.v. lignocaine. Br J Anaesth 1982;54:501–5.
7. Nagy I, Woolf CJ. Lignocaine selectively reduces C fibre-evoked neuronal activity in rat spinal cord in vitro
by decreasing N
-methyl-d-aspartate and neurokinin receptor-mediated post-synaptic depolarizations: implications for the development of novel centrally acting analgesics. Pain 1996;64:59–70.
8. Mao J, Chen LL. Systemic lidocaine for neuropathic pain relief. Pain 2000;87:7–17.
9. Merskey H, Bogduk N. Classification of chronic pain: description of chronic pain syndromes and definitions of pain terms, 2nd ed. Seattle: IASP Press, 1994:222.
10. Wall PD. Neuropathic pain and injured nerve: central mechanisms. Br Med Bull 1991;47:631–47.
11. Devor M. Neuropathic pain and injured nerve: peripheral mechanisms. Br Med Bull 1991;47:619–30.
12. Cepeda MS, Lau J, Carr DB. Defining the therapeutic role of local anesthetic sympathetic blockade in complex regional pain syndrome: a narrative and systematic review. Clin J Pain 2002;18:216–33.
13. Wall PD, Gutnick M. Properties of afferent nerve impulses originating from a neuroma. Nature 1974;248(5451):740–3.
14. Chabal C, Russell LC, Burchiel KJ. The effect of intravenous lidocaine, tocainide, and mexiletine on spontaneously active fibers originating in rat sciatic neuromas. Pain 1989;38:333–8.
15. Kajander KC, Wakisaka S. Bennett GJ. Spontaneous discharge originates in the dorsal root ganglion at the onset of a painful peripheral neuropathy in the rat. Neurosci Lett 1992;138:225–8.
16. Abdi S, Lee DH, Chung JM. The anti-allodynic effects of amitriptyline, gabapentin, and lidocaine in a rat model of neuropathic pain. Anesth Analg 1998;87:1360–6.
17. Stebbing MJ, Bornstein JC. Electrophysiological analysis of the convergence of peripheral inputs onto neurons of the coeliac ganglion in the guinea pig. J Auton Nerv Syst 1994;46:93–105.
18. Sotgiu ML, Biella G, Castagna A, et al. Different time-courses of i.v. lidocaine effect on ganglionic and spinal units in neuropathic rats. Neuroreport 1994;5:873–6.
19. Omana-Zapata I, Khabbaz MA, Hunter JC. QX-314 inhibits ectopic nerve activity associated with neuropathic pain. Brain Res 1997;771:228–37.
20. Tanelian DL, MacIver MB. Analgesic concentrations of lidocaine suppress tonic A-delta and C fiber discharges produced by acute injury. Anesthesiology 1991;74:934–6.
21. Abram SE, Yaksh TL. Systemic lidocaine blocks nerve injury-induced hyperalgesia and nociceptor-driven spinal sensitization in the rat. Anesthesiology 1994;80:383–91.
22. Yaksh TL. Regulation of spinal nociceptive processing: where we went when we wandered onto the path marked by the gate. Pain 1999;Suppl 6:S149–52.
23. Devor M, Wall PD, Catalan N. Systemic lidocaine silences ectopic neuroma and DRG discharge without blocking nerve conduction. Pain 1992;48:261–8.
24. Ferrante FM, Paggioli J, Cherukuri S, Arthur GR. The analgesic response to intravenous lidocaine in the treatment of neuropathic pain. Anesth Analg 1996;82:91–7.
25. Kalso E, Tramer MR, McQuay HJ, Moore RA. Systemic local-anaesthetic-type drugs in chronic pain: a systematic review. Eur J Pain 1998;2:3–14.
26. Lefebvre C, Clarke MJ. Identifying randomized trials. In: Egger M, Smith GD, Altman DG, eds. Systematic reviews in health care: meta-analysis in context, 2nd ed. London: BMJ Publishing Group, 2001:69–86.
27. Jadad AR, Moore RA, Carrol D. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1–12.
28. Attal N, Gaude V., Brasseur L, et al. Intravenous lidocaine in central pain: a double-blind, placebo-controlled, psychophysical study. Neurology 2000;54:564–74.
29. Backonja MM, Tremont-Lukats IW. Pain relief from lidocaine at three different doses: A randomized placebo-controlled trial. Neurology 2000;54:A81.
30. Baranowski AP, De Courcey J, Bonello E. A trial of intravenous lidocaine on the pain and allodynia of postherpetic neuralgia. J Pain Symptom Manage 1999;17:429–33.
31. Bruera E, Ripamonti C, Brennis C, et al. A randomized double-blind controlled trial of intravenous lidocaine in the treatment of neuropathic cancer pain. J Pain Symptom Manage 1992;7:138–41.
32. Chabal C, Jacobson L, Mariano A, et al. The use of oral mexiletine for the treatment of pain after peripheral nerve injury. Anesthesiology 1992;76:513–7.
33. Chiou-Tan FY, Tuel SM, Johnson JC, et al. Effect of mexiletine on spinal cord injury dysesthetic pain. Am J Phys Med Rehabil 1996;75:84–7.
34. Dejgard A, Petersen P, Kastrup J. Mexiletine for treatment of chronic painful diabetic neuropathy. Lancet 1988;1:9–11.
35. Ellemann K, Sjogren P, Banning AM, et al. Trial of intravenous lidocaine on painful neuropathy in cancer patients. Clin J Pain 1989;5:291–4.
36. Fassoulaki A, Patris K, Sarantopoulos C. Hogan Q. The analgesic effect of gabapentin and mexiletine after breast surgery for cancer. Anesth Analg 2002;95:985–91.
37. Galer BS, Harle J, Rowbotham MC. Response to intravenous lidocaine infusion predicts subsequent response to oral mexiletine: a prospective study. J Pain Symptom Manage 1996;12:161–7.
38. Kastrup J, Petersen P, Dejgard A, et al. Intravenous lidocaine infusion—a new treatment of chronic painful diabetic neuropathy? Pain 1987;28:69–75.
39. Kemper CA, Kent G, Burton S, Deresinski SC. Mexiletine for HIV-infected patients with painful peripheral neuropathy: a double-blind, placebo-controlled, crossover treatment trial. J Acquir Immune Defic Syndr Hum Retrovirol 1998;19:367–72.
40. Kieburtz K, Simpson D, Yiannoutsos C, et al. A randomized trial of amitriptyline and mexiletine for painful neuropathy in HIV infection. AIDS Clinical Trial Group 242 Protocol Team. Neurology 1998;51:1682–8.
41. Lindstrom P, Lindblom U. The analgesic effect of tocainide in trigeminal neuralgia. Pain 1987;28:45–50.
42. Marchettini P, Lacerenza M, Marangoni C, et al. Lidocaine test in neuralgia. Pain 1992;48:377–82.
43. Matsuoka K, Hirata Y, Kanazawa Y. Optimal dose evaluation of mexiletine hydrochloride (MX-PDN) in patients with diabetic polyneuropathy. Igaku Yakugaku 1996;36:665–93.
44. Matsuoka K, Kanazawa Y, Ohtake M, Kaihara S. Double blind trial of mexiletine on painful diabetic polyneuropathy. Diabetologia 1997;40:A559.
45. Medrik-Goldberg T, Lifschitz D, Pud D, et al. Intravenous lidocaine, amantadine, and placebo in the treatment of sciatica: a double-blind, randomized, controlled study. Reg Anesth Pain Med 1999;24:534–40.
46. Oskarsson P, Ljunggren JG, Lins PE. Efficacy and safety of mexiletine in the treatment of painful diabetic neuropathy. The Mexiletine Study Group. Diabetes Care 1997;20:1594–7.
47. Rowbotham MC, Reisner-Keller LA, Fields HL. Both intravenous lidocaine and morphine reduce the pain of postherpetic neuralgia. Neurology 1991;41:1024–8.
48. Stracke H, Meyer U, Schumacher H, et al. Mexiletine in the treatment of painful diabetic neuropathy. Med Klin 1994;89:124–31.
49. Sorensen J, Bengtsson A, Backman E, et al. Pain analysis in patients with fibromyalgia: effects of intravenous morphine, lidocaine, and ketamine. Scand J Rheumatol 1995;24:360–5.
50. Wallace MS, Dyck JB, Rossi SS, Yaksh TL. Computer-controlled lidocaine infusion for the evaluation of neuropathic pain after peripheral nerve injury. Pain 1996;66:69–77.
51. Wallace MS, Ridgeway BM, Leung AY, et al. Concentration-effect relationship of intravenous lidocaine on the allodynia of complex regional pain syndrome types I and II. Anesthesiology 2000;92:75–83.
52. Wallace MS, Magnuson S, Ridgeway B. Efficacy of oral mexiletine for neuropathic pain with allodynia: a double-blind, placebo-controlled, crossover study. Reg Anesth Pain Med 2000;25:459–67.
53. Wright JM, Oki JC, Graves L 3rd. Mexiletine in the symptomatic treatment of diabetic peripheral neuropathy. Ann Pharmacother 1997;31:29–34.
54. Wu CL, Tella P, Staats PS, et al. Analgesic effects of intravenous lidocaine and morphine on postamputation pain: a randomized double-blind, active placebo-controlled, crossover trial. Anesthesiology 2002;96:841–8.
55. Kastrup J, Petersen P, Dejgard A, Hilsted J. Treatment of chronic painful diabetic neuropathy with intravenous lidocaine infusion. BMJ (Clin Res Ed) 1986;292:173.
56. Stracke H, Meyer UE, Schumacher HE, Federline K. Mexiletine in the treatment of diabetic neuropathy. Diabetes Care 1992;15:1550–5.
57. Dunlop RJ, Hockley JM, Tate T, Turner P. Flecainide in cancer nerve pain. Lancet 1991;337:1347.
58. Rowlingson JC, Difazio CA, Foster J, Carron H. Lidocaine as an analgesic for experimental pain. Anesthesiology 1980;52:20–2.
59. Wallace MS, Laitin S, Licht D, Yaksh TL. Concentration-effect relations for intravenous lidocaine infusions in human volunteers: effects on acute sensory thresholds and capsaicin-evoked hyperpathia. Anesthesiology 1997;86(6):1262–72.
60. Ando K, Wallace MS, Braun J, et al. Effect of oral mexiletine on capsaicin-induced allodynia and hyperalgesia: a double-blind, placebo-controlled, crossover study. Reg Anesth Pain Med 2000;25:468–74.
61. Dirks K, Fabricius P, Petersen KL, et al. The effect of systemic lidocaine on pain and secondary hyperalgesia associated with the heat/capsaicin sensitization model in healthy volunteers. Anesth Analg 2000;91:967–72.
62. Gottrup H, Hansen PO, Arendt-Nielsen L, Jensen TS. Differential effects of systemically administered ketamine and lidocaine on dynamic and static hyperalgesia induced by intradermal capsaicin in humans. Br J Anaesthesia 2000;84:155–62.
63. Bach FW, Jensen TS, Kastrup J, et al. The effect of intravenous lidocaine on nociceptive processing in diabetic neuropathy. Pain 1990;40:29–34.
64. Reljanovic M, Pibernik M, Beer Z, et al. Mexiletine treatment in painful diabetic polyneuropathy. Diabetologia 1996;39:A260.
65. Sakurai M, Kanazawa I. Positive symptoms in multiple sclerosis: their treatment with sodium channel blockers, lidocaine and mexiletine. J Neurol Sci 1998;162:162–8.
66. Posner IA. Treatment of fibromyalgia syndrome with intravenous lidocaine: a prospective, randomized pilot study. J Musculoskeletal Pain 1994;2/4:55–65.
67. Catala E, Ferrandiz M, Aliaga L, et al. Intravenous lidocaine compared with sympathetic blocks as treatment for post-herpetic neuralgia: a 1-year survey. The Pain Clinic 1994;7:205–10.
68. Lau J, Ioannidis JP, Schmid CH. Quantitative synthesis in systematic reviews. Ann Intern Med 1997;127:820–6.
69. Feinstein AR, Horwitz RI. Problems in the “evidence” of “evidence-based medicine.” Am J Med 1997;103:529–35.
70. Kjaergard LL, Villumsen J, Gluud C. Reported methodologic quality and discrepancies between large and small randomized trials in meta-analyses. Ann Intern Med 2001;135:982–9.
71. Hedges LV, Pigott TD. The power of statistical tests in meta-analysis. Psychol Methods 2001;6:203–17.
72. Woolf CJ, Bennett GJ, Doherty M, et al. Towards a mechanism-based classification of pain? Pain 1998;77:227–9.
73. Farrar JT, Portenoy RK, Berlin JA, et al. Defining the clinically important difference in pain outcome measures. Pain 2000;88:287–94.
74. Appel LJ. The verdict from ALLHAT: thiazide diuretics are the preferred initial therapy for hypertension. JAMA 2002;288:3039–42.
75. Sriwatanakul K, Kelvie W, Lasagna L. The quantification of pain: an analysis of words used to describe pain and analgesia in clinical trials. Clin Pharmacol Ther 1982;32:143–8.
76. Cepeda MS, Africano JM, Polo R, et al. What decline in pain intensity is meaningful to patients with acute pain? Pain 2003;105:151–7.
77. Farrar JT, Young JP Jr, LaMoreaux L, et al. Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain 2001;94:149–58.
78. Chaplan SR, Bach FW, Shafer SL, Yaksh TL. Prolonged alleviation of tactile allodynia by intravenous lidocaine in neuropathic rats. Anesthesiology 1995;83:775–85.
79. Edwards WT, Habib F, Burney RG, Begin G. Intravenous lidocaine in the management of various chronic pain states: a review of 211 cases. Reg Anesth 1985;10:1–6.
80. Atkins RL. Intravenous lidocaine for the treatment of intractable pain of adiposis dolorosa. Int J Obesity 1982;6:351–7.
81. Rowbotham M, Harden N, Stacey B, et al. Gabapentin for the treatment of postherpetic neuralgia. JAMA 1998;280:1837–42.
82. Gorson KC, Schott C, Herman R, et al. Gabapentin in the treatment of painful diabetic neuropathy: a placebo controlled, double blind, crossover trial. J Neurol Neurosurg Psychiatry 1999;66:251–2.
83. Eidelman A, Weiss J, Enu I, et al. A comparative efficacy and costs of various topical anesthetics for repair of dermal lacerations: a systematic review of randomized, controlled trials. J Clin Anesth 2005;17:106–16.
84. Eidelman A, Weiss J, Lau J, Carr D. Topical anesthetics for dermal instrumentation: a systematic review of randomized, controlled trials. Ann Emerg Med. In press.
85. Gammaitoni AR, Alvarez NA, Galer BS. Safety and tolerability of the lidocaine patch 5%, a targeted peripheral analgesic: a review of the literature. J Clin Pharmacol 2003;43:111–7.
86. Jarvis B, Coukell AJ. Mexiletine: a review of its therapeutic use in painful diabetic neuropathy. Drugs 1998;56:691–707.