7. Smoking;8. High alcohol intake;9. Low socioeconomic status; and10.Renal failure.

Approximately 20 to 40% ofpatients with diabetes develop someform of neuropathy. Autonomic andmotor involvement is less commonthan sensory neuropathy. Auto-nomic nerve damage can cause car-diovascular abnormalities andsystemic symptoms, such as indiges-tion, diarrhoea or constipation,

PAIN MANAGEMENT

Medical Progress August 2003 13

Recommendations for the Management of

Painful Diabetic Peripheral NeuropathyThe Multidisciplinary Panel on Neuropathic Pain*

I. Pathophysiology, Prevalenceand Symptoms

Diabetic neuropathy is afamily of progressive,degenerative disordersaffecting the sensory,

motor or autonomic peripheralnerves. Poor glycaemic control andchronic hyperglycaemia are believedto be responsible for peripheralnerve damage, although the precisemechanism is not known. Abnor-

malities in nerve growth factors,autoimmune disorders, ischaemiaand hypoxia may also contribute toloss of nerve fibres.

Risk factors for the develop-ment and progression of diabeticneuropathy include:1. Poor glycaemic control;2. Increasing age;3. Undiagnosed type 2 diabetes;4. Long duration of diabetes;5. Cardiovascular disease;6. Peripheral vascular disease;

* Panel members: Chen Phoon Ping, MBBS, FANZCA, FFPMANZCA, FHKCA, FHKAM, DipPainMgtConsultant, Department of Anaesthesiology, Intensive Care and Operating Services, Alice Ho Miu Ling Nethersole Hospital, and Adjunct AssociateProfessor, The Chinese University of Hong Kong, Hong Kong SAR.Josephine WY Ip, MBBS, MS, FRCS(Ed), FHKAM(Ortho), DipHandSurg(FESSH)Chief, Division of Hand and Foot Surgery, Department of Orthopaedic Surgery, The University of Hong Kong, Queen Mary Hospital, Hong Kong SAR.Joseph MK Lam, MBChB, FRCS(Edin), FCSHK, FHKAM(Surg)Consultant Neurosurgeon and Adjunct Associate Professor, Division of Neurosurgery, Department of Surgery, The Chinese University of Hong Kong, Princeof Wales Hospital, Hong Kong SAR.Lee Tsun Woon, MBBS, FANZCA, FHKCA, FHKAM(Anaesthesiology), DipPainMgtChief of Service, Department of Anaesthesia and Intensive Care, Tuen Mun Hospital, Hong Kong SAR.Tsoi Tak Hong, MBBS, MRCP, FRCP(Edin), FRCP(Glas), FHKCP, FHKAM(Med)Specialist in Neurology and Consultant Physician, Department of Medicine, Pamela Youde Nethersole Eastern Hospital, Hong Kong SAR.Wong Chun Por, MBBS, FHKAM(Med), FRCP(Lond), FRCP(Glas)Chief of Service, Integrated Medicine Service, Consultant Geriatrician, and Head, Department of Geriatrics, Ruttonjee Hospital, Hong Kong SAR.Lawrence KS Wong, MBBS, MD, MRCP, FHKAM(Med), FRCPAssociate Professor, Division of Neurology, Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Prince of Wales Hospital,Hong Kong SAR.The Multidisciplinary Panel on Neuropathic Pain is supported by an educational grant from Pfizer Corporation Hong Kong Limited.

© 2005 CMPMedica Pacific Ltd. Reprinted with permission from Medical Progress 2003 Vol. 30 No. 8.

14 Medical Progress August 2003

PAIN MANAGEMENT

dizziness, bladder infections anderectile dysfunction. Diabetic amyo-trophy, predominantly a disorder ofmotor neurones, is usually seen inelderly patients with poorly con-trolled type 2 diabetes. Musclesaround the pelvis and thigh becomeweak and painful, and mobility isaffected.

The most common form of dia-betic neuropathy, distal symmetricpolyneuropathy, predominantlyaffects sensory functions. Thesepolyneuropathies usually involve theperipheral nerves of the feet and legsand, in some cases, the hands andarms. Onset, type and severity ofsymptoms vary widely amongpatients, making diagnosis andprognosis imprecise. Early symp-toms include numbness, tingling,burning or pain, which may laterdevelop into loss of reflexes, footdeformities (Charcot’s joint), muscleweakness or paralysis. Up to 10%of affected patients experience per-sistent neuropathic pain that mayinclude dysaesthesia. Depressionmay occur in patients with severe

pain. Diabetic neuropathy may leadto foot ulceration and even the needfor amputation. Early diagnosis andmanagement of at-risk patientsmight prevent at least half of all diabetes-related amputations.

II. Diagnosis

Diagnosis of diabetic neuropathy isbased on clinical symptomatologyand a comprehensive neurologicalexamination. In addition, otherunderlying pathologies for neuropa-thy should be excluded (e.g., vascu-lar disease, HIV, vitamin B12

deficiency, hypothyroidism). Clini-cal features vary widely, and peoplewith diabetic neuropathy may evenbe pain-free. However, the classicpresentation of advanced polyneu-ropathy is distal wasting and weak-ness, absent tendon reflexes, andglove-and-stocking sensory loss orpain. Patients may also experienceallodynia. The following historyand examinations should be consid-ered in the diagnosis of diabeticneuropathy:

1. Full patient history to determine:• Type, duration and level of

control of diabetes;• Nature of symptoms, if any

(intensity, duration, progres-sion, nocturnal exacerbation,recurrent foot problems);

• Pain characteristics using stan-dard pain questionnaires(chronic or acute pain, bilat-eral, type of dysaesthesia,hyperaesthesia); and

• Lifestyle factors that maycontribute to progression ofneuropathy.

2. Neurological examination:• Characterize distal sensory

function and reflexes, e.g., pin-prick test, light touch, vibrationtest, ankle reflex, pressure per-ception, temperature assess-ment, monofilament test oftwo-point discrimination; and

• Electrophysiological assessmentto document neuropathy, ifrequired, e.g., nerve conductionstudy and electromyography, orDoppler sonography to deter-mine the presence of vasculardisease.

III. Management

The goal of treatment for painfuldiabetic peripheral neuropathy is torelieve painful symptoms, preventfurther tissue damage and improvepatient education.

Patients without clinical neu-ropathy should be educated onlifestyle, foot care and the impor-tance of controlling glycaemia toslow disease progression. Referpatients to a diabetes specialistnurse or chiropodist for a yearlyfoot examination, if necessary.

Patients with suspected diabeticamyotrophy or a decreased qualityof life due to clinical symptomatic

Approximately 20 to 40% of diabetic patientsdevelop some form of neuropathy.

Medical Progress August 2003 15

PAIN MANAGEMENT

neuropathy should be referred to adiabetologist or neurologist for fur-ther evaluation. In the interim, treat-ment for acute or chronic painshould be started.

Patients with peripheral neuropa-thy and complete or partial loss ofsensation should be educated ongood glycaemic control and footcare. Refer patients to a diabetesfoot specialist.

Trauma, cellulitis or acuteischaemia of the foot require urgentreferral to a specialist diabetes foot-care team to prevent new or recur-rent lesions and amputation.

In all diabetic patients, the impor-tance of good glycaemic controlshould be stressed, as this may slowor prevent the development ofperipheral neuropathy and othercomplications, including retinopathy,nephropathy and angiopathy.

IV. Pain Treatments

The mainstay therapeutic agents formanaging diabetic neuropathic painare tricyclic antidepressants (TCAs)and anticonvulsants. Combinationsof pharmacological, physical andpsychological interventions are likelyto attain the optimum level of painrelief for most patients.

For chronic pain, TCAs (e.g.,amitriptyline, imipramine, nortripty-line, desipramine) should be consid-ered first-line therapies. Pain reliefmay not be apparent for up to 3weeks. TCAs are contraindicated inpatients with cardiac and hepaticdisease, which include many olderpatients. Some patients cannot toler-ate the side effects of TCAs –drowsiness, anticholinergic effectsand postural hypotension – butthese can be minimized by startingwith a low dose at night andincreasing gradually (e.g., for

amitriptyline, start with 10-25 mgdaily and increase to 50-100 mgdaily). Nortriptyline, imipramineand desipramine are less sedatingthan amitriptyline.

For acute pain, start with simpleanalgesics and progress to TCAs orother adjuvant analgesics, if neces-sary.

If TCAs are contraindicated,ineffective and/or not well tolerated,anticonvulsants (e.g., gabapentin orcarbamazepine) should be consid-ered as an alternative first-linechoice. Side effects can be common,but are minimized by adopting aslow titration schedule. Gabapentinis generally associated with fewerside effects than TCAs, carba-mazepine or phenytoin. Gabapentinshould be commenced at 300 mg atbedtime and increased by 300 mgevery 3 days up to a dose of 1,800 mgdaily after 1 week (given in 3divided doses). If higher daily dosesare required for maintenance, themaximum recommended dose is3,600 mg daily (a lower dose is rec-ommended in patients with renalimpairment). For elderly patients orpatients susceptible to side effects, it is recommended to increasegabapentin dosage by 300 mg everyweek, or to commence with a lowerdose (e.g., 100 mg).

Tramadol may be an effectivealternative for some patients.

Patients remaining refractory toa reasonable trial of pharmacother-apy (e.g., 2-3 months with 2-3 dif-ferent agents) should be referred toa multidisciplinary pain clinic forfurther therapeutic initiatives.

Physical stimulation, such astranscutaneous electrical nerve stim-ulation (TENS) and acupuncture,may counteract painful sensations.However, acupuncture and topicaltreatments should be used with cau-

tion in the lower leg in patients withdiabetes, as these may aggravate theskin and lead to infection. Painmanagement programmes andbehavioural therapy can also be used with pharmacologicalapproaches to teach patients how tolive with pain. Regular walking,warm baths or elastic stockings mayalso help to relieve leg pain.

V. Appendix on Evidence-BasedManagement of PainfulDiabetic Peripheral Neuropathy

The pharmacological treatmentsincluded in these recommendationsare based on published clinical evidence in diabetic neuropathypatients and current clinical practice.However, some agents may not beapproved for use in neuropathic painsyndromes. Full prescribing informa-tion should be consulted before initiating drug therapy.

Pharmacological ManagementTricyclic antidepressantsSeveral clinical trials have shownthat TCAs are effective in treatingpainful diabetic neuropathy, al-though they are not licensed for thisindication. It has been postulatedthat TCAs relieve pain independ-ently from their antidepressantaction, and dampen sensory nervefunction by inhibiting muscarinicand α-adrenergic receptors.

Data from systematic reviews: Arecent systematic review of random-ized, placebo-controlled trials of anti-depressants in diabetic neuropathypooled data from eight studies usingTCAs (amitriptyline, clomipramine,desipramine, imipramine and mapro-tiline) with a total of 283 patientepisodes.1 The relative benefit oftreatment was 1.9 (95% CI: 1.5-2.3)and the number needed to treat

16 Medical Progress August 2003

PAIN MANAGEMENT

(NNT) for one patient to achieve a50% reduction in pain was 3.5(95% CI: 2.5-5.6). This demon-strates the efficacy of TCAs in dia-betic neuropathy. Doses were withinthe low to mid range of those recom-mended for depression. A similarconclusion was reached by a reviewcompleted 4 years earlier.2 Further-more, one review of placebo-con-trolled trials from 1989 to 1999reported an NNT of 1.4 forimipramine compared with 2.4 forother TCAs.3

The incidence of adverse events issignificantly greater with TCAs thanplacebo. For minor adverse events,the number needed to harm (NNH)was 3.2 (95% CI: 2.3-5.2) and formajor adverse effects (i.e., thosenecessitating drug withdrawal), theNNH was 14 (95% CI: 8.5-38).1

Other antidepressants: Pooleddata from three trials (162 patientepisodes) was used to investigate theeffectiveness of the selective sero-tonin-reuptake inhibitors (SSRIs)citalopram, fluoxetine and paroxe-tine in treating painful diabetic neu-ropathy. These agents had a relativebenefit of 1.3 (95% CI: 1.0-1.8),demonstrating no significant differ-ence between the SSRIs andplacebo.1 Doses used were two-thirds of the maximum recom-mended daily dose for depression.Another review reported an NNTfor SSRIs of 6.7.3 The efficacy ofvenlafaxine in painful diabetic neu-ropathy has not yet been assessed incontrolled clinical trials.

AnticonvulsantsAnticonvulsants act as membrane-stabilizing agents and reduce thepotential for the transmission ofabnormal pain signals. Anticonvul-sants are often used to managechronic neuropathic pain when TCAs

are inappropriate or ineffective. Data from systematic reviews: A

systematic review of anticonvulsantsin diabetic neuropathy from all ran-domized, placebo-controlled trialspublished up to 1999 has been per-formed.1 Four studies investigatedanticonvulsants in 247 patients (2studies on phenytoin, 1 on carba-mazepine and 1 on gabapentin). Thepooled data showed that anticonvul-sants were significantly superior toplacebo in the treatment of painfuldiabetic neuropathy. The relativebenefit of treatment was 2.4 (95%CI: 1.8-3.2) and the NNT for one

patient to achieve a 50% reductionin pain was 2.7 (95% CI: 2.2-3.8). Asimilar conclusion was reached by areview of trials from 1966 to 1994.4,5

From one review of placebo-controlled trials, the specific NNTfor carbamazepine was 3.3, and forgabapentin, 3.7.3

Anticonvulsants had a signifi-cantly greater incidence of adverseeffects than placebo. For minoradverse events, the NNH for allanticonvulsants combined was 2.7(95% CI: 2.2-3.4) and for pheny-toin, the NNH was 3.2 (95% CI:2.1-6.3).1 For every eight patientsachieving a 50% reduction in painon an anticonvulsant, one will ex-perience a major adverse event tocause drug withdrawal. An earlier

review calculated that the NNH formajor side effects was 20.4,5

Anticonvulsants and antidepres-sants had comparable efficacy out-comes for painful diabetic neu-ropathy (NNT for all antidepres-sants was 3.4 [95% CI: 2.6-4.7] andfor all anticonvulsants was 2.7[95% CI: 2.2-3.8]). There was nosignificant difference in the inci-dence of minor adverse events forboth drugs; however, more patientsrequired withdrawal from antide-pressants than from anticonvulsantsbecause of major side effects.1

Gabapentin: Gabapentin is thefirst oral therapy to be licensed forthe management of painful diabeticneuropathy as a result of a large,multicentre, double-blind, placebo-controlled trial.6 Patients with a 1-to 5-year history of pain attributedto diabetic neuropathy were ran-domly assigned to gabapentin(n=84; titrated from 900 mg/day to3,600 mg/day or maximum toler-ated dosage) or placebo (n=81).Patients treated with gabapentin hadsignificantly lower mean daily painscores, at 8 weeks than placebo-treated patients (p<0.001). Thisimprovement was apparent fromweek 2. Significant improvementswith gabapentin were also seen insleep interference scores, the Short-Form McGill Pain Questionnairescores, and the Patient and ClinicianGlobal Impression of Change scores.Quality of life, particularly bodilypain, mental health and vitalityimproved significantly withgabapentin. Eight percent ofpatients in the gabapentin groupwithdrew because of adverse events,compared with 6% in the placebogroup. Dizziness and somnolencewere the only two adverse eventsthat occurred more frequently ingabapentin-treated patients.

“The mainstay therapeutic

agents for managing

diabetic neuropathic pain are

tricyclic antidepressants

and anticonvulsants”

Medical Progress August 2003 17

PAIN MANAGEMENT

With a sound evidence base forefficacy and safety, gabapentinshould be considered as an alterna-tive first-line therapy to TCAs forpainful diabetic neuropathy.

Lamotrigine: A 6-week, random-ized, controlled trial compared lam-otrigine (n=29) with placebo (n=30)for the treatment of diabetic neu-ropathy. Lamotrigine (200-400 mgdaily) reduced daily numerical painscores significantly more thanplacebo (p<0.001). Global assess-ment of efficacy was also better withlamotrigine. With an adverse eventprofile similar to placebo, lamo-trigine is an effective therapy fordiabetic neuropathy.7

Phenytoin: There is little evi-dence from clinical trials on the effi-cacy of phenytoin in diabeticneuropathy. Moreover, comparedwith newer anticonvulsants, bothphenytoin and carbamazepine haveunfavourable safety profiles, andcan cause haematological changesand cardiac arrhythmias.

TramadolTramadol is a synthetic, centrallyacting, non-opioid analgesic thatmay be a useful alternative to TCAsand anticonvulsants. A double-blind,randomized, controlled trial of 6weeks’ duration evaluated 131patients with painful diabetic neu-ropathy. Tramadol (average dailydose of 210 mg) provided signifi-cantly more effective pain relief(p<0.001) and greater improvementof both physical (p=0.02) and socialfunctioning (p=0.04) than placebo.No benefits were seen in sleep dis-turbance.8 In a 6-month extension ofthis study, mean pain relief scoreswere well maintained.9

Topical CapsaicinSeveral large-scale, placebo-con-

trolled studies have shown that0.075% capsaicin cream was morebeneficial than vehicle-only cream inreducing pain intensity.10-12 Capsaicincream allowed greater participationin work and recreational activities,while sleep quality improved signifi-cantly.11 Smaller, vehicle-controlledstudies report similar benefits forcapsaicin.13,14 Furthermore, a meta-analysis of efficacy trials reported anodds ratio in favour of capsaicinover placebo of 2.74 (95% CI: 1.73-4.32) for neuropathic pain associ-ated with diabetes.15 However, localapplication of topical agents to thelower limbs should only be per-formed under clinician supervision,as capsaicin and herbal remediesmay irritate the skin, leading toinfection.

Topical capsaicin has also beencompared with oral amitriptyline ina double-blind, multicentre, parallel-group study involving 235 patients.Both agents produced equal and sta-tistically significant improvements inpain over 8 weeks; however,amitriptyline treatment was associ-ated with systemic side effects.16

Topical capsaicin is not associatedwith any severe systemic adverseeffects. However, stinging and burn-ing, particularly during the first weekof therapy, is reported by manypatients. Topical capsaicin meritsconsideration as adjuvant therapy fordiabetic neuropathy that is chroni-cally painful and difficult to treat.

MexiletineMexiletine has been evaluated in sev-eral randomized, placebo-controlledtrials in patients with painful dia-betic neuropathy. Visual analoguescale (VAS) pain ratings improved inall studies that used this measure;however, the improvement was significantly greater than placebo in

only two studies.17 Patients (n=16)receiving mexiletine 10 mg/kg/dayfor 10 weeks had greater improve-ments in pain ratings (as measuredby VAS) than placebo-treatedpatients.18 Nocturnal pain was alsoreduced in 31 patients receivingmexiletine 675 mg/day for 3 weeks.19

Patients with stabbing or burningpain, heat sensations or formicationwere more likely to benefit frommexiletine than patients with otherpain sensations.20 In general, mexile-tine did not have a significant influ-ence on sleep quality in patients withdiabetic neuropathy.17

Other Agents and Novel Therapiesfor the FutureIntravenous lignocaine infusion maybe useful for providing short-termrelief in patients with chronic painfuldiabetic neuropathy.21 Long-term opi-oid therapy may also be considered ifpatients remain refractory to otherforms of treatment. Supplementationwith the vitamins thiamine or pyri-doxine was associated with improveddiabetic peripheral neuropathy symp-toms in an African study.22 Therefore,a balanced diet with vitamin supple-mentation, if necessary, is importantfor diabetic patients.

New agents may eventually beavailable for the treatment of dia-betic neuropathy. A recent random-ized, placebo-controlled study of 279patients with diabetic neuropathycompared the aldose reductaseinhibitor fidarestat (1 mg daily for52 weeks) with placebo.23 Fidarestattreatment improved nerve conduc-tion and subjective symptoms of dia-betic neuropathy. Another agent thatmay improve some aspects of nerveconduction is the antioxidant thioticacid (alpha-lipoic acid).24 In contrast,a phase III, randomized, placebo-controlled study investigating

18 Medical Progress August 2003

PAIN MANAGEMENT

recombinant human nerve growthfactor in diabetic neuropathy didnot demonstrate significant treat-ment benefit.25 Isosorbide dinitratespray was assessed in a double-blind, placebo-controlled, cross-overstudy in 22 patients with diabeticneuropathy.26 The spray was foundto reduce overall neuropathic painand burning sensation, which maybe due to increased generation ofnitric oxide. Further studies arerequired to assess the benefit andpotential clinical role of these novelagents.

Non-pharmacological ManagementPhysical StimulationTENS may be effective in somepatients with painful diabetic neu-ropathy. Transcutaneous electrother-apy may also be combined with apharmacological agent, such asamitriptyline, to increase symptomrelief.27 Further, percutaneous electri-cal nerve stimulation (PENS) wasevaluated in a randomized, cross-over, sham-controlled study in 50patients with diabetic neuropathy.28

Active PENS treatment reducedpain, improved physical activity andquality of sleep, and was associatedwith a reduced requirement for non-opioid analgesic medication. Moretraditional therapies, such asacupuncture, may also provide reliefin some patients. However, use ofacupuncture, particularly on thelower limbs, may lead to skin aggra-vation and infection. Therefore,acupuncture in the lower limbsshould be avoided.

Psychological TreatmentMultidisciplinary pain managementprogrammes may also provide psy-chological approaches, such asrelaxation and diversion techniques,and behavioural therapy, to help

patients manage pain. Cognitivebehavioural therapy can be intro-duced to help prevent pain behav-iours in patients with chronic pain.Referral to a psychiatrist can be considered if the patient is depressed.

References1. Collins SL, Moore RA, McQuay HJ, Wiffen P.Antidepressants and anticonvulsants for diabeticneuropathy and postherpetic neuralgia: a quanti-tative systematic review. J Pain Symptom Manage2000;20:449-458.2. McQuay HJ, Tramer M, Nye BA, et al. A system-atic review of antidepressants in neuropathic pain.Pain 1996;68:217-227.3. Sindrup SH, Jensen TS. Efficacy of pharmaco-logical treatments of neuropathic pain: an updateand effect related to mechanism of drug action.Pain 1999;83:389-400.4. McQuay H, Carroll D, Jadad AR, et al. Anticon-vulsant drugs for management of pain: a system-atic review. BMJ 1995;311:1047-1052.5. Wiffen P, Collins S, McQuay H, et al. Anticon-vulsant drugs for acute and chronic pain. CochraneDatabase Syst Rev 2000;(3):CD001133.6. Backonja M, Beydoun A, Edwards KR, et al.Gabapentin for the symptomatic treatment ofpainful neuropathy in patients with diabetes melli-tus: a randomized controlled trial. JAMA 1998;280:1831-1836.7. Eisenberg E, Lurie Y, Braker C, et al. Lamotriginereduces painful diabetic neuropathy: a randomized,controlled study. Neurology 2001;57:505-509.8. Harati Y, Gooch C, Swenson M, et al. Double-blind randomized trial of tramadol for the treat-ment of the pain of diabetic neuropathy.Neurology 1998;50:1842-1846.9. Harati Y, Gooch C, Swenson M, et al. Main-tenance of the long-term effectiveness of tramadolin treatment of the pain of diabetic neuropathy. JDiabetes Complications 2000;14:65-70.10. Treatment of painful diabetic neuropathy withtopical capsaicin. A multicenter, double-blind,vehicle-controlled study. The Capsaicin StudyGroup. Arch Intern Med 1991;151:2225-2229.11. Effect of treatment with capsaicin on dailyactivities of patients with painful diabetic neu-ropathy. The Capsaicin Study Group. Diabetes Care1992;15:159-165.12. Scheffler NM, Sheitel PL, Lipton MN. Treatmentof painful diabetic neuropathy with capsaicin0.075%. J Am Podiatr Med Assoc 1991;81:288-293.13. Tandan R, Lewis GA, Krusinski PB, et al. Topicalcapsaicin in painful diabetic neuropathy. Con-trolled study with long-term follow-up. DiabetesCare 1992;15:8-14.

14. Basha KM, Whitehouse FW. Capsaicin: a thera-peutic option for painful diabetic neuropathy.Henry Ford Hosp Med J 1991;39:138-140.15. Zhang WY, Li Wan Po A. The effectiveness oftopically applied capsaicin. A meta-analysis. Eur JClin Pharmacol 1994;46:517-522.16. Biesbroeck R, Bril V, Hollander P, et al. A dou-ble-blind comparison of topical capsaicin and oralamitriptyline in painful diabetic neuropathy. AdvTher 1995;12:111-120.17. Jarvis B, Coukell AJ. Mexiletine. A review of itstherapeutic use in painful diabetic neuropathy.Drugs 1998;56:691-707.18. Dejgard A, Petersen P, Kastrup J. Mexiletine fortreatment of chronic painful diabetic neuropathy.Lancet 1988;1:9-11. 19. Oskarsson P, Ljunggren JG, Lins PE. Efficacy andsafety of mexiletine in the treatment of painfuldiabetic neuropathy. The Mexiletine Study Group.Diabetes Care 1997;20:1594-1597.20. Stracke H, Meyer U, Schumacker H, et al. Mex-iletine in treatment of painful diabetic neuropathy.Med Klin 1994;89:124-131.21. Kastrup J, Petersen P, Dejgard A, et al. Intra-venous lidocaine infusion – a new treatment ofchronic painful diabetic neuropathy? Pain1987;28:69-75.22. Abbas ZG, Swai AB. Evaluation of the efficacyof thiamine and pyridoxine in the treatment ofsymptomatic diabetic peripheral neuropathy. EastAfr Med J 1997;74:803-808.23. Hotta N, Toyota T, Matsuoka K, et al. Clinicalefficacy of fidarestat, a novel aldose reductaseinhibitor, for diabetic peripheral neuropathy: a 52-week multicenter placebo-controlled double-blindparallel group study. Diabetes Care 2002;24:1776-1782.24. Reljanovic M, Reichel G, Rett K, et al. Treatmentof diabetic polyneuropathy with the antioxidantthiotic acid (alpha-lipoic acid): a two year multi-center randomized double-blind placebo-controlledtrial (ALADIN II). Alpha Lipoic Acid in Diabetic Neu-ropathy. Free Radic Res 1999;31:171-179.25. Apfel SC, Schwartz S, Adornato BT, et al. Effi-cacy and safety of recombinant human nervegrowth factor in patients with diabetic polyneu-ropathy: A randomized controlled trial. JAMA2000;284:2215-2221.26. Yuen KCJ, Baker NR, Rayman G. Treatment ofchronic painful diabetic neuropathy with isosorbidedinitrate spray. Diabetes Care 2002;25:1699-1703.27. Kumar D, Alvaro MS, Julka IS, Marshall HJ. Dia-betic peripheral neuropathy. Effectiveness of elec-trotherapy and amitriptyline for symptomaticrelief. Diabetes Care 1998;21:1322-1325.28. Hamza MA, White PF, Craig WF, et al. Percuta-neous electrical nerve stimulation: a novel anal-gesic therapy for diabetic neuropathic pain.Diabetes Care 2000;23:365-370.