Vol. 43 No. 1 January 2012 Journal of Pain and Symptom Management 87

Original Article

Scrambler Therapy May Relieve ChronicNeuropathic Pain More Effectively ThanGuideline-Based Drug Management: Resultsof a Pilot, Randomized, Controlled TrialGiuseppe Marineo, PhD, Vittorio Iorno, MD, Cristiano Gandini, MD,Vincenzo Moschini, MD, and Thomas J. Smith, MDDelta Research & Development (G.M.), Centro Ricerche Bioingegneria Medica, University of Rome

‘‘Tor Vergata,’’ Rome, Italy; Centro di Medicina del Dolore ‘‘Mario Tiengo’’ (V.I., C.G., V.M.), IRCCS

Fondazione Ospedale Maggiore Policlinico, Mangiagalli and Regina Elena, Milan, Italy; and

Division of Hematology/Oncology and Palliative Care (T.J.S.), Massey Cancer Center, Virginia

Commonwealth University, Richmond, Virginia, USA

Abstract

Context. Neuropathic pain is common, disabling, and often difficult to treat.Objectives. To compare guideline-based drug management with Scrambler

therapy, a patient-specific electrocutaneous nerve stimulation device.Methods. A clinical trial with patients randomized to either guideline-based

pharmacological treatment or Scrambler therapy for a cycle of 10 daily sessionswas performed. Patients were matched by type of pain including postsurgicalneuropathic pain, postherpetic neuralgia, or spinal canal stenosis. Primaryoutcome was change in visual analogue scale (VAS) pain scores at one month;secondary outcomes included VAS pain scores at two and three months, painmedication use, and allodynia.

Results. Fifty-two patients were randomized. The mean VAS pain score beforetreatment was 8.1 points (control) and 8.0 points (Scrambler). At one month, themeanVAS score was reduced from 8.1 to 5.8 (�28%) in the control group, and from8 to 0.7 points (�91%) in the Scrambler group (P< 0.0001). At two and threemonths, the mean pain scores in the control group were 5.7 and 5.9 points,respectively, and 1.4 and 2 points in the Scrambler group, respectively (P< 0.0001).More relapses were seen in polyradicular pain than monoradicular pain, butretreatment andmaintenance therapy gave relief. No adverse effects were observed.

Conclusion. In this pilot randomized trial, Scrambler therapy appeared torelieve chronic neuropathic pain better than guideline-based drugmanagement. J Pain Symptom Manage 2012;43:87e95. � 2012 U.S. Cancer PainRelief Committee. Published by Elsevier Inc. All rights reserved.

Address correspondence to: Giuseppe Marineo, PhD,Delta Research & Development, Via di Mezzocam-mino, 85 00187 Rome, Italy. E-mail: g.marineo@mclink.it, or Thomas J. Smith, MD, Duffy PalliativeCare Service, Sidney Kimmel Cancer Center, Johns

Hopkins, 600 North Wolfe Street, Blaylock 369,Baltimore, MD 21287-0005. E-mail: tsmit136@jhmi.edu

Accepted for publication: March 15, 2011.

� 2012 U.S. Cancer Pain Relief CommitteePublished by Elsevier Inc. All rights reserved.

0885-3924/$ - see front matterdoi:10.1016/j.jpainsymman.2011.03.015

88 Vol. 43 No. 1 January 2012Marineo et al.

Key Words

Chronic neuropathic pain, analgesics, refractory pain, Scrambler therapy, electroanalgesia

IntroductionNeuropathic pain is common, chronic, dis-

abling, and often difficult to effectively treat.1

Common types of neuropathic pain includepostsurgical pain, postherpetic neuralgia(PHN), spinal cord stenosis (SCS) (also knownas narrow canal syndrome), and chemotherapy-induced peripheral neuropathy.2 Althoughconventional treatments such as opioids, neu-roleptics, and other drugs help, all have sideeffects and limited effectiveness.3

Scrambler therapy is a novel approach topain control that attempts to relieve painby providing ‘‘nonpain’’ information via cuta-neous nerves to block the effect of pain infor-mation. Scrambler therapy synthesizes 16different types of nerve action potentials simi-lar to endogenous ones, assembles them intosequences, and uses algorithms to determinea patient-specific cutaneous electrostimulationto reduce pain. Scrambler therapy has relievedrefractory chronic pain in uncontrolled clini-cal trials. In the pilot trial, 11 cancer patientswith abdominal pain received 10 daily one-hour treatment sessions.4 Pain was reducedfrom 8.6 to 2.3, on a numeric rating scalefrom 0 to 10, after the first treatment and to<0.5 (P< 0.0001) at the end of 10 sessions.In the second trial, 226 patients with neuro-pathic pain, including failed back surgerypain, brachial plexus neuropathy, and others,were treated.5 Eighty percent of patientshad greater than 50% pain relief. Smithet al.6 treated 16 patients with refractorychemotherapy-induced neuropathic pain with10 daily hour-long sessions to the painful areas.Pain scores were reduced by 58% from thestart of treatment to the end. No toxicity hasbeen observed in any trial. However, all thesetrials were single-arm trials with no controlgroup.

The purpose of this study was to directly com-pare management of chronic refractory neuro-pathic pain by Scrambler therapy withmanagement using current guideline-baseddrug treatment7 in a randomized controlledtrial.

MethodsStudy PopulationPatients were eligible if they had chronic

neuropathic pain rated as 6 or more on a visualanalogue scale (VAS) on at least four days eachweek, for the prior three months despite treat-ments including antidepressants, anticonvul-sants, and opioids. Patients were treated atOspedale Maggiore, Policlinico Mangiagalliand Regina Elena of Milan at the Centre forPain Medicine ‘‘Mario Tiengo.’’ Most of the re-cruited patients were already undergoing treat-ment in this facility and had been given thestandard treatment for neuropathic pain bytheir physicians. The inclusion and exclusioncriteria are listed in Table 1.The monitoring of patients during and after

treatments always took place at the same facil-ity, the Centre for Pain Medicine in Milan. Tworesearch assistants performed the Scramblertherapy under the direction of the treatingphysicians. These two research assistants col-lected the pain assessments in both arms ofthe study.Informed consent was obtained from all pa-

tients. The Scrambler therapy device had beengranted Ministry of Health approval for hospi-tal and ambulatory use. The hospital’s sci-entific and health unit, acting as theinstitutional review board, approved and mon-itored the study.

Randomization and StratificationFor this trial at one center, the patients who

were eligible (pain $6 for at least threemonths, despite treatment according to localpractice) were classified according to theirpain diagnosis (postsurgical, PHN, SCS).They were then assigned to a treatment group:alternative drug therapy according to theEuropean Federation of Neurological Socie-ties (EFNS) guidelines and in standard prac-tice at this center, or Scrambler therapy withno change in the ineffective drug regimen.They were assigned consecutively in the orderthey were enrolled in the trial, for example,the first PHN patient to drug treatment, the

Table 1Inclusion and Exclusion Criteria

Inclusion Exclusion

Presence of mainly neuropathic pain Cancer-related painVAS pain intensity $6 in the preceding

three monthsPresence of serious psychiatric disorder (schizophrenia, manic-

depressive psychosis, primary major depression)Failure to respond to currently used pharmacologic

treatment of neuropathic pain (antidepressants,anticonvulsants, and opioids)

Presence of dermatologic conditions that preclude application ofskin electrodes

Absence of significant responses to TENS or othersimilar electroanalgesic methods

Uncontrolled seizures

In addition to pain, presence of related sensitivitysymptoms: allodynia, hyperpathia, hyperesthesia

Use of antiseizure medications

Presence of pain for at least six months Any form of medical ‘‘metal’’ device (e.g., pacemakers, defibrillators,vascular clips or stents, cardiac valve or joint replacements)

Frequency of pain more than four days per weekAge >18 yearsConsent to Scrambler therapy treatment

Vol. 43 No. 1 January 2012 89Scrambler Therapy for Chronic Neuropathic Pain

second PHN patient to Scrambler therapy, thethird PHN patient to drug treatment, and soforth. The assignment was done in order bythe research assistants, with no exceptions.Allocation was not concealed (Fig. 1).

Standardized Control TreatmentThe control group patients were managed

by the same team of pain specialists using themost up-to-date EFNS guidelines. The mostcommon baseline therapy (typically amitripty-line, gabapentin, and tramadol) before ran-domization, which had resulted in a baselineVAS pain score of 8.1, was switched to

All patients, n=56 Assessed for eligibility

Randomized; stratified by type

of pain

Control Group, n=26 Start new drugs, e.g.,

amitriptyline, clonazepam, oxycodone

Control Group, Evaluable n=26

Scrambler Group, Evaluable n=26

Scrambler Group, n=26 Start Scrambler therapy;

stay on prior drugs

Fig. 1. Assignment of patients to treatment groups.

a potentially more effective one (typicallyamitriptyline, clonazepam, and oxycodone).Low-dose clonazepam is classified as an anti-convulsant, has documented efficacy in caseseries,8e10 and is the standard practice at thiscenter when gabapentin has not beeneffective.

Standardized Scrambler TreatmentThe Scrambler attempts to deliver ‘‘non-

pain’’ information to the area in pain by simu-lating five external artificial neurons. Actionpotentials that resemble normal nerve im-pulses are digitally synthesized, assembledinto packets of information strings, and deliv-ered using standard silver gel electrodes simi-lar to electrocardiogram electrodes. Eachnew packet is created with an algorithm thattakes into account the previous outputs,dynamically modifying four main variables.These variables include the following: 1) typeof action potential to use (16 different possiblecombinations); 2) packet-associated frequency(from 43 to 52 Hz); 3) packet time duration(0.7e10 seconds); and 4) the amplitude ofmodulation. The system quickly tries differentcombinations until pain relief is achieved.(The technology details are described in pat-ent number PCT/IT2007/000647.) The im-pulses are transmitted by surface electrodesplaced on the skin in the dermatome areasof pain above and below the dermatome.The electrical charge used in Scrambler ther-apy is low, and the U.S. Food and Drug Admin-istration has approved it as safe. At the highestsetting, ‘‘70’’ on the dial from 10 to 70, the

90 Vol. 43 No. 1 January 2012Marineo et al.

amperage (A) is 3.50e5.50 mA, and themaximum current density is only 0.0002009W/cm2.Each Scrambler therapy patient was given

a 45-minute daily treatment for 10 consecutivedays, Monday through Friday. The stimuluswas increased to the maximum intensity indi-vidually bearable by the patient that did notcause any additional pain or discomfort. Theprincipal investigator (V. I.) chose the besttreatment areas during the first visit, whichwere then replicated daily. The Scrambler ther-apy group maintained their starting drugtreatment with no changes. Normally the elec-trodes are never applied directly on the pain-ful area but in the dermatomes above andbelow the pain affected area. For example, ifthe pain involves L3, the first electrode is posi-tioned close to but outside of the painful areain dermatome L2 or L3 and the other on theopposite side of the pain area in zone L3 orL4. Once the electrodes are positioned, theoperator slowly raises the stimulation level un-til pain relief is obtained; if not obtained, theoperator can add or move channels to increasethe coverage. There are a total of five chan-nels, or paired sets of electrodes. If pain isnot relieved, the electrode placement or stim-ulus is changed.

Data and Statistical ConsiderationsThe primary endpoint was the change in

pain VAS scores11 from entry to the scores atone, two, and three months. Pain intensitywas measured by an unmarked 10-cm longVAS.12 The patients were classified as having‘‘monoradicular pain’’ if they had one domi-nant area of pain, for example, one area of

Table 2Demographic Compariso

Patient Description Control Group (n¼ 26)

SexMale 10 (38.46)Female 16 (61.53)

Age, mean (SD), years 53 (16.14)

DiagnosesPostsurgical neuropathic pain 14 (53.84)PHN 8 (30.76)SCS 4 (15.38)

Pain score at entry, after six monthsof standard treatment

8.11/10� 1.03

SD¼ standard deviation.

PHN, and ‘‘polyradicular pain’’ if they hadmultiple areas of pain.13 Allodynia was testedwith von Frey elements by the research assis-tants and recorded as ‘‘present’’ or ‘‘not pres-ent.’’ The sample size of 26 in each arm wasdetermined using an anticipated effect size of�1.59, with a starting VAS pain score of 6and a standard deviation of 2 to give a 5%alpha error margin and 80% power. All statisti-cal calculations were done with StatMate 2(GraphPad Software, Inc., La Jolla, CA;http://www.graphpad.com/StatMate/statmate.htm).For the primary endpoint of pain, a repeated-

measures analysis of variance (ANOVA) wasdone with the VAS score as the dependent vari-able and time (baseline, onemonth, etc.), treat-ment (treated or control), and treatment bytime interaction terms as the independent vari-ables. The repeated-measures model accountsfor the correlations that might arise from thesame individuals being observed over time. Sec-ondary endpoints included change in painscores by the type of pain and mono- or polyra-dicular nature of the pain, change in allodynia,and change in medication use and doses; allwere measured at entry, one, two, and threemonths. Changes in pain intensity over timewere analyzed with one-way ANOVA and theTukey-Kramer Multiple Comparisons Test.The difference in medication type and in dos-age, and in allodynia, was evaluated usingrepeated-measures ANOVA.

ResultsThe randomized patients were similar in

both groups, as shown in Table 2. The study

n of the Groups

, n (%) Scrambler Therapy Group (n¼ 26), n (%)

9 (34.61)17 (65.38)

56 (16.26)

14 (53.84)8 (30.76)4 (15.38)

8.01� 1.12

Table 3Mean Variation in Pain Intensity (by VAS) Over

Time

Time of Assessment(months)

Group

Control Scrambler Therapy

T0, entry 8.11/10� 1.03 8.01� 1.12T1, one month 5.84/10� 1.34 0.78� 1.78T2, two months 5.76/10� 1.40 1.49� 2.39T3, three months 5.91/10� 1.44 2.03� 3.14

All results statistically significant, P< 0.001 by ANOVA.

Vol. 43 No. 1 January 2012 91Scrambler Therapy for Chronic Neuropathic Pain

sample statistical validity was confirmed withthe normality test of Kolmogorov and Smirnov(P> 0.1).

The primary endpoint results are shown inFig. 2. The VAS pain score in the control groupfell by 28% at one month. Average VAS painintensity in the Scrambler therapy groupdecreased from entry to T1 (one month),T2 (two months), and T3 (three months)(ANOVA P< 0.0001). The treatment by timeinteraction term was significant (P< 0.0001)suggesting that the decline in the VAS scoreover time in the treated group was significantlysteeper than the control group (Fig. 2.) Thecomparison between the two arms of the studyat monthly intervals confirmed this signifi-cance (P< 0.001, Tukey-Kramer MultipleComparisons Test). Results shown in Table 3document that pain scores were reduced sig-nificantly with Scrambler therapy comparedwith the control group (all results P< 0.001by ANOVA). Pain was reduced in all groups:postsurgical neuropathic pain (P< 0.0001 byANOVA), PHN (P< 0.0001), and SCS(P¼ 0.0108). Fig. 3 shows the effect in the dif-ferent diagnoses.

Scrambler therapy appeared to be effectivein both mono- and polyradicular pain, butmore patients relapsed in the polyradicular

Fig. 2. Effect of treatment on pain scores over time.A repeated-measures ANOVA with the VAS score asthe dependent variable and time (baseline, onemonth, etc.), treatment (treated or control), andtreatment by time interaction terms as the indepen-dent variables. The repeated-measures modelaccounts for the correlations that might arise fromthe same individuals being observed over time. Thetreatment by time interaction term was significant(P< 0.0001), suggesting that the decline in the VASscore over time in the treated group was significantlysteeper than that in the control group.

group, as shown in Fig. 4. These pain relapsesat three months were statistically significant inthe Scrambler therapy group (P< 0.001,Tukey-Kramer Multiple Comparisons Test)but not in the control group (P> 0.05). Thisrelapse was observed only in patients with poly-radicular neuropathy, compared with thosewith mononeuropathy (P< 0.001, ANOVAtest by Tukey-Kramer Multiple ComparisonsTest). The control group changes had no dif-ference in the relapse rate when comparingthe type of pain (P> 0.05).

Scrambler therapy appeared to have a posi-tive effect on allodynia. Allodynia was reducedat one and three months in the Scramblertherapy group (Fig. 5). The changes were sta-tistically significant comparing the Scramblertherapy group with the control group at onemonth (P¼ 0.0017, Fisher’s Exact Test), twomonths (P¼ 0.0094), and three months(P¼ 0.0644). However, given the simplemethod of assessment (‘‘yes’’ or ‘‘no’’) donein this pilot trial, further work is needed toconfirm this effect.

Scrambler therapy was associated with signif-icant pain medication dose reductions, asshown in Fig. 6. The percentage reductionwas calculated compared with the initial doseat entry, then reassessed at times T2 and T3.

Fig. 3. Effect of treatment by type of pain.PSP¼ postsurgical pain.

Fig. 4. Effect of therapy by mono- or polyneuropathy.

92 Vol. 43 No. 1 January 2012Marineo et al.

At T3, opioids were totally eliminated in 11 of17 cases, halved in one case, and unvaried infive cases. Anticonvulsants were eliminated in17 of 24 cases, reduced in one case, and unvar-ied in six cases. Lastly, antidepressants wereeliminated in nine of 19 cases, reduced infour cases, and unvaried in six cases. Dosagevariation was statistically significant (repeated-measures ANOVA: P< 0.0001).

DiscussionIn this small, randomized, controlled trial,

Scrambler therapy appeared to reduce pain, al-lodynia, and pain drug use significantly betterthan guideline-based drug therapy. In 21 of 26patients, pain could be relieved entirely.Scrambler therapy was associated with a 91%pain reduction compared with a 28% reduc-tion using new medications.

Chronic pain syndromes can be helped bypatient-specific (adjusted to the tolerance ofthe individual patient) direct nerve stimu-lation.14e16 The mechanisms by which directpatient-specific nerve stimulation reduce pain

Fig. 5. Effect of treatment on allodynia. Allodyniawas assessed as present or absent. The graph showsthe percentage of subjects in each arm that hadallodynia at each time point.

include raising the ‘‘gate’’ threshold for painat the spinal cord, reducing ‘‘wind up’’ (centralsensitization of the spinal cord and brain thatamplifies the abnormal feelings), reducing im-pulses from the damaged nerve, and reducingpsychological maladaptation to pain.17 Spinalcord stimulation gave a 50% pain reductionin small nonrandomized series for chronicpain from complex regional pain syndrome I(median change in VAS 10 to 0e2, P< 0.01,sustained)18 and PHN (median change inVAS 9 to 1, sustained).19 Spinal cord stimula-tion gave over 50% pain reduction in a ran-domized trial compared with conventionalmedical management in patients with failedback syndrome (mean scores fell from 7.6 to3.8 or less, sustained for 24 months,P< 0.001).20 This same >50% effect size hasbeen observed in randomized trials of intraspi-nal drug delivery systems compared with con-ventional pain management, when opioidsand local anesthetics can be infused21 to actdirectly on nerves. However, spinal cord stimu-lation and intrathecal drug delivery involveinvasive expensive technology with the possi-bility of serious complications.Scrambler therapy differs from transcutane-

ous electrical nerve stimulation (TENS) in

Fig. 6. Effect of treatment on pain medication use.

Vol. 43 No. 1 January 2012 93Scrambler Therapy for Chronic Neuropathic Pain

many aspects, although both provide stimula-tion via peripheral nerves. Clinically, TENS ther-apy has been shown effective in postoperativepain and musculoskeletal pain, but the numberand quality of randomized controlled trials areoften inadequate for particular conditions.22

We were not able to find any randomized trialsof TENS for SCS or chronic postsurgical pain.As reviewed by Niv et al.,23 the TENS effect inPHN has been limited in randomized trialsand disappears a few hours after treatment.TENS provides an on-off biphasic current with-out variation, whereas Scrambler therapy pro-vides continuously changing variable nonlinearwaveforms. Recent studies with TENS unitshave used a continuous pulse pattern, pulsewidth of 200 microseconds, and a pulse fre-quency of 80 Hz, increaseduntil the patient feelsa strong sensation. The Scrambler therapy aver-age charge per phase is 38.8 microcoulombs,similar to conventionalTENSdevices. Thephaseduration is 6.8e10.9 microseconds, and thepulse rate is 43e52 Hz. Because the frequencydelivered by the device never exceeds 52 Hz,the mean energy delivered per second is gener-ally less thanmost standard TENSdevices, whichdeliver a square wave with frequencies greaterthan 52 Hz.

How Scrambler therapy causes pain reliefrequires further study, but our observationsmay inform mechanisms. First, Scrambler ther-apy gives new ‘‘nonpain’’ information suchthat patients report new sensations in thepain area (pressure, itching, ‘‘bee sting’’ sensa-tions, and a flow of impulses). Second, it is notsimple C-fiber electrical stimulation, whichwould produce pain. Third, Scrambler therapyis not producing paresthesias because thepatient does not feel numbness and can stillfeel other noxious stimuli. Fourth, Scramblertherapy analgesia occurs quickly, suggestingthat the receptors are transmitting the‘‘nonpain’’ information. Fifth, the sustainedpain relief for days or months suggests eitherresetting of calcium channels (as with zicono-tide) or remodulation of the pain system’sresponse. Finally, the patient feels the sensa-tion throughout the dermatome, not justunder the electrode patch, suggesting thespread of ‘‘nonpain’’ information along thelines of nerve transmission. Clearly, more studyis needed to define the effect and themechanisms.

There are limitations to this study. First, this isa well-balanced, randomized, controlled studysimilar to that done comparing implantabledrug delivery systems with guideline-basedcare,24 but it is not a ‘‘sham’’ trial. Some re-searchers will insist that the only proof of effi-cacy is a randomized, double-blind, believableplacebo, or sham-controlled trial, but thesemay be difficult to perform.25 It has beenhard to devise appropriate blinded studies be-cause Scrambler therapy requires adjustmentsof the electrode placement and dose, titratedto pain relief, before the actual daily treatmentis begun. Alternative methods to control forplacebo effect include two strategies donehere: first, to set a pain relief goal that wouldlikely be unobtainable with placebo, and sec-ond, to allow an effective comparison controlgroup. Here, the reduction in pain was 91%,much higher than typically seen with placebo.For example, in a randomized trial of electrosti-mulation for back pain, the sham group hada 9% reduction in back pain,26 and a study in-volving various nonpharmacological therapiesin low back pain showed that the placebo effectwas less than two points on a normalized 0e10scale.27 The second alternative to a placebo-controlled trial allows an effective control com-parison group. Here, the control group hada 28% reduction in pain by the end of the firstmonth, consistent with the 14%e20% reduc-tion seen in worst and usual pain in theguideline-managed group of a large random-ized trial,28 and the 39% reduction in painseen in the guideline-managed control groupof a cancer pain trial.20,25 A second limitationto the study is the type of treatment providedto the control group. Although some clinicianswould suggest alternative drug treatments, thiswas the current practice at this Italian pain cen-ter, and the control group had a very reasonable28% reduction in pain. A third limitation is thesmall sample size, but the study was powered todetect a relatively large difference in pain con-trol and accomplished this.

There are strengths to this study, in additionto the limitations. The patients were well bal-anced in the two arms. The patient-reportedoutcomes are all standard, reproducible,and valid. The magnitude of the pain reliefeffect is large, persistent, consistent with thereduction in pain medication use, and consis-tent with the size of the pain relief in the

94 Vol. 43 No. 1 January 2012Marineo et al.

single-arm uncontrolled Scrambler therapystudies. The comparison group had good re-lief of pain from standard guideline-basedmeasures applied by the expert group, asnoted above. The pain relief was well beyondthe 50%29 and 33%30 reductions proposed asbeing clinically important for chronic pain.

In conclusion, the pain relief obtained inthis small, pilot, randomized trial encouragesfurther development of both treatment andof knowledge regarding Scrambler therapy.This knowledge will provide a better under-standing of the mechanisms of action andnew opportunities for the treatment of allforms of pain. It also provides more knowledgeof effect size for further randomized placebo-or sham-controlled trials, which are underway.

Disclosures and AcknowledgmentsDr. Giuseppe Marineo holds property rights

to the Scrambler Therapy basic researchand international patent to its technologydevelopment.

References1. Botez SA, Herrmann DN. Sensory neuropathies,from symptoms to treatment. Curr Opin Neurol2010;23:502e508.

2. Wolf S, Barton D, Kottschade L, Grothey A,Loprinzi C. Chemotherapy-induced peripheral neu-ropathy: prevention and treatment strategies. Eur JCancer 2008;44:1507e1515.

3. Finnerup NB, Sindrup SH, Jensen TS. The evi-dence for pharmacological treatment of neuro-pathic pain. Pain 2010;150:573e581.

4. Marineo G. Untreatable pain resulting from ab-dominal cancer: new hope from biophysics? JOP2003;4:1e10.

5. Sabato AF, Marineo G, Gatti A. Calmare therapy.Minerva Anestesiol 2005;71:479e482.

6. Smith TJ, Coyne PJ, Parker G, Dodson P,Ramakrishnan V. Pilot trial of a patient-specific cuta-neous electro-stimulation device (MC5-A Calmare�)for chemotherapy induced peripheral neuropathy.J Pain Symptom Manage 2010;40:883e891.

7. Attal N, Cruccu G, Haanp€a€a M, et al.EFNS TaskForce. EFNS guidelines on pharmacological treat-ment of neuropathic pain. Eur J Neurol 2006;13:1153e1169.

8. Hugel H, Ellershaw JE, Dickman A. Clonazepamas an adjuvant analgesic in patients with cancer-related neuropathic pain. J Pain Symptom Manage2003;26:1073e1074.

9. Fishbain DA, Cutler RB, Rosomoff HL,Rosomoff RS. Clonazepam open clinical treatmenttrial for myofascial syndrome associated chronicpain. Pain Med 2000;1:332e339.

10. Zakrzewska JM. Medical management of trigem-inal neuropathic pains. Expert Opin Pharmacother2010;11:1239e1254.

11. Williamson A, Hoggart B. Pain: a review of threecommonly used pain rating scales. J Clin Nurs 2005;14:798e804.

12. Dixon JS, Bird HA. Reproducibility alonga 10 cm vertical visual analogue scale. Ann RheumDis 1981;40:87e89.

13. Tan JC. Practical manual of physical medicineand rehabilitation, 2nd ed. Philadelphia, PA: ElsevierMosby, 2006.

14. Folletti A, Durrer A, Buchser E. Neurostimula-tion technology for the treatment of chronic pain:a focus on spinal cord stimulation. Expert RevMed Devices 2007;4:201e214.

15. deLeon-CasasolaOA. Spinal cord andperipheralnerve stimulation techniques for neuropathic pain.J Pain SymptomManage 2009;38(Suppl 2):S28eS38.

16. Mailis-Gagnon A, Furlan AD, Sandoval JA,Taylor R. Spinal cord stimulation for chronic pain.Cochrane Database Syst Rev 2004;(3). CD003783.

17. Jensen MP. A neuropsychological model ofpain: research and clinical implications. J Pain2010;11:2e12.

18. Harke H, Gretenkort P, Ladleif HU, Rahman S.Spinal cord stimulation in sympathetically main-tained complex regional pain syndrome type Iwith severe disability. A prospective clinical study.Eur J Pain 2005;9:363e373.

19. Harke H, Gretenkort P, Ladleif HU, Koester P,Rahman S. Spinal cord stimulation in postherpeticneuralgia and in acute herpes zoster pain. AnesthAnalg 2002;94:694e700.

20. Kumar K, Taylor RS, Jacques L, et al. Spinalcord stimulation versus conventional medical man-agement for neuropathic pain: a multicentre rando-mised controlled trial in patients with failed backsurgery syndrome. Pain 2007;132:179e188.

21. Smith TJ, Staats PJ, Pool G, et al. Intrathecalimplantable drug delivery systems give sustainedpain control, less side effects, and possibly bettersurvival for six months: results of a randomized clin-ical trial vs. comprehensive medical management.Ann Oncol 2005;16:825e833.

22. DeSantana JM, Walsh DM, Vance C, Rakel BA,Sluka KA. Effectiveness of transcutaneous electri-cal nerve stimulation for treatment of hyperalge-sia and pain. Curr Rheumatol Rep 2008;10:492e499.

23. Niv D, Maltsman-Tseikhin A, Lang E. Posther-petic neuralgia: what do we know and where arewe heading? Pain Physician 2004;7:239e247.

Vol. 43 No. 1 January 2012 95Scrambler Therapy for Chronic Neuropathic Pain

24. Smith TJ, Staats PJ, Pool G, et al. Randomizedclinical trial of comprehensive medical manage-ment of refractory cancer pain vs. comprehensivemedical management plus intrathecal implantabledrug delivery systems. J Clin Oncol 2002;20:4040e4049.

25. Dworkin RH, Turk DC, Peirce-Sandner S, et al.Research design considerations for confirmatorychronic pain clinical trials: IMMPACT recommenda-tions. Pain 2010;149:177e193.

26. Ghoname EA, Craig WF, White PF, et al. Percu-taneous electrical nerve stimulation for low backpain: a randomized crossover study. JAMA 1999;281:818e823.

27. Chou R, Huffman LH, American Pain Society,American College of Physicians. Nonpharmacologic

therapies for acute and chronic low back pain: a re-view of the evidence for an American Pain Society/American College of Physicians clinical practiceguideline. Ann Intern Med 2007;147:492e504.

28. DuPen S, DuPen A, Polossar N, et al. Imple-menting guidelines for cancer pain management:results of a randomized controlled clinical trial.J Clin Oncol 1999;17:361e370.

29. McQuay HJ, Moore RA, Eccleston C, Morley S,Williams AC. Systematic review of outpatient ser-vices for chronic pain control. Health TechnolAssess 1997;1:1e135.

30. Farrar JT, Young JP Jr, LaMoreaux L, Werth JL,Poole RM. Clinical importance of changes inchronic pain intensity measured on an 11-pointnumerical pain rating scale. Pain 2001;94:149e158.