Diabetic Somatic Neuropathies - Diabetes Carecare.diabetesjournals.org/content/diacare/27/6/1458.full.pdfAcute sensory neuropathy is a distinct variety of the symmetrical polyneuropa-thies

Diabetic Somatic NeuropathiesANDREW J.M. BOULTON, MD, FRCP

1,2

RAYAZ A. MALIK, MB, PHD2

JOSEPH C. AREZZO, PHD3

JAY M. SOSENKO, MD, MS1

SECTION 1:INTRODUCTION — The neuropa-thies are among the most common of thelong-term complications of diabetes, af-fecting up to 50% of patients (1–4). Theirclinical features vary immensely, and pa-tients may present to a wide spectrum ofspecialties, from dermatology to podiatry,for example, or from urology to cardiol-ogy. Neuropathies are characterized by aprogressive loss of nerve fibers, whichmay affect both principle divisions of theperipheral nervous system. This reviewwill focus on the somatic neuropathies;those affecting the autonomic divisionwere recently reviewed by Vinik et al. (5).There is increasing evidence that mea-sures of neuropathy, such as electrophys-iology and quantitative tests, arepredictors of not only end points, includ-ing foot ulceration, but also of mortality(6).

The epidemiology and natural historyof diabetic neuropathy (DN) remainpoorly defined, partly because of poor pa-tient selection and the variable criteria forwhat constitutes a diagnosis of DN. Theseaspects, as well as the pathogenesis of DN,will be covered in detail in this review.Studies have confirmed the major contri-bution of prolonged hyperglycemia in theetiopathogenesis of neuropathy and neu-

ropathic pain (7–10), and this and otherputative mechanisms will be discussed.

The clinical features, diagnosis, andmanagement of the focal and multifocalneuropathies will be described. A majorportion of this review will discuss theclinical features, assessment, and man-agement of the patient with the most com-mon form of DN, diabetic distal sensorypolyneuropathy (DPN). The late sequelaeof DPN and their prevention will also bedescribed.

Finally, practical guidelines for thescreening of DPN in clinical practice willbe provided. For further details on thistopic, please refer to recent reviews (11–18).

SECTION 2: DEFINITIONSAND CLASSIFICATION OFTHE DNs

A. DefinitionsMembers of an international consensusmeeting on the outpatient diagnosis andmanagement of DN agreed on a simpledefinition of DN as “the presence of symp-toms and/or signs of peripheral nerve dys-function in people with diabetes after theexclusion of other causes ” (19). It wasalso agreed that neuropathy cannot be di-agnosed without a careful clinical exami-

nation—absence of symptoms cannot beequated with absence of neuropathy, asasymptomatic neuropathy is common.The importance of excluding nondiabeticcauses was emphasized in the RochesterDiabetic Neuropathy Study, in which upto 10% of peripheral neuropathy in dia-betic patients was deemed to be of nondi-abetic causation (1).

A more detailed definition of neurop-athy was previously agreed upon at theSan Antonio Consensus Conference: “di-abetic neuropathy is a descriptive termmeaning a demonstrable disorder, eitherclinically evident or subclinical, that oc-curs in the setting of diabetes mellituswithout other causes for peripheral neu-ropathy. The neuropathic disorder in-cludes manifestations in the somaticand/or autonomic parts of the peripheralnervous system” (20). It is generallyagreed that DN should not be diagnosedon the basis of one symptom, sign, or testalone: a minimum of two abnormalities(from symptoms, signs, nerve conductionabnormalities, quantitative sensory tests,or quantitative autonomic tests) is recom-mended by Dyck (21). Certainly, for clin-ical trials or epidemiological studies, oneof these two abnormalities should includequantitative tests or electrophysiology.

B. Classification of DNsNumerous classifications of the variety ofsyndromes affecting the peripheral ner-vous system in diabetes have been pro-posed in recent years. Some have beenbased on presumed etiology, topographi-cal features, or pathological features.However, until we have a clear under-standing of the etiopathogenesis of neu-ropathy, classifications based on theclinical manifestations are most com-monly used (22–25). Three slightly dif-ferent clinical classifications are presentedin Table 1. Table 1A describes a purelyclinical classification (11,22), whereasTable 1B bases its classification on a mix-ture of clinical and anatomical findings(25). The classification proposed byThomas (23,24) will be used throughoutthis review (Table 1C). This classificationis based on the premise that DN is not aunitary condition but is the result of anumber of disturbances in the peripheral

● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

From the 1Division of Endocrinology, Metabolism and Diabetes, University of Miami School of Medicine,Miami, Florida; the 2University Department of Medicine, Manchester Royal Infirmary, Manchester, U.K; andthe 3Department of Neuroscience, Albert Einstein College of Medicine of Yeshiva University, Bronx, NewYork.

Address correspondence and reprint requests to Andrew J.M. Boulton, MD, FRCP, Division of Endocri-nology, University of Miami School of Medicine, P.O. Box 016960 (D-110). E-mail: [email protected].

Abbreviations: AGE, advanced glycation end product; AR, aldose reductase; ARI, AR inhibitor; CIDP,chronic inflammatory demyelinating polyneuropathy; CMAP, compound muscle action potential; CTS,carpal tunnel syndrome; DAG, 1,2-diacylglycerol; DCCT, Diabetes Control and Complications Trial; DN,diabetic neuropathy; DPN, diabetic distal sensory polyneuropathy; GLA, �-lipoic acid; ICNT, intermediatecutaneous nerve of the thigh; IGT, impaired glucose tolerance; IL, interleukin; LA, lipoic acid; MMP, matrixmetalloproteinase; MNSI, Michigan Neuropathy Screening Instrument; MRI, magnetic resonance imaging;NAD, neuroaxonal dystrophy; NCV, nerve conduction velocity; NDS, Neuropathic Disability Score; NGF,nerve growth factor; NIS, Neuropathy Impairment Score; PKC, protein kinase C; PNS, Peripheral NerveSociety; QOL, quality of life; QST, quantitative sensory testing; RAGE, AGE receptor; SNAP, sensory nerveaction potential; SSRI, selective serotonin-reuptake inhibitor; STZ, streptozotocin; VEGF, vascular endothe-lial growth factor; VPT, vibration perception threshold.

A table elsewhere in this issue shows conventional and Systeme International (SI) units and conversionfactors for many substances.

© 2004 by the American Diabetes Association.

R e v i e w s / C o m m e n t a r i e s / P o s i t i o n S t a t e m e n t sT E C H N I C A L R E V I E W

1458 DIABETES CARE, VOLUME 27, NUMBER 6, JUNE 2004

nervous system as a consequence ofhyperglycemia.Rapidly reversible hyperglycemic neu-ropathy. It has been recognized formany years that rapidly reversible abnor-malities of nerve conduction may occur inpatients with recently diagnosed or tran-siently poorly controlled diabetes; theseabnormalities may be accompanied bydistal uncomfortable sensory symptoms(11,23,24). Such changes are unlikely tobe caused by structural abnormalities, as

recovery soon follows restoration of eu-glycemia. It remains unknown whetherthese temporary abnormalities result in agreater risk of developing other chronicneuropathies in later life.Generalized symmetrical polyneuropa-thies.Chronic sensorimotor neuropathy is themost common form of DN that is dis-cussed in detail below. It is usually of in-sidious onset and may be present at thediagnosis of type 2 diabetes in �10% of

subjects (8,26). Whereas up to 50% ofpatients may be asymptomatic, 10–20%may experience troublesome sensorysymptoms that require specific treatment.Sensorimotor neuropathy is often accom-panied by autonomic dysfunction. Latesequelae of neuropathy, which includeinsensate foot ulceration, Charcot (neuro-pathic) arthropathy, and occasionallyeven amputation (27), are also discussedbelow.

Acute sensory neuropathy is a distinctvariety of the symmetrical polyneuropa-thies with an acute or subacute onsetcharacterized by severe sensory symp-toms, usually with few if any clinicalsigns. The natural history is one of grad-ual improvement of these symptoms withestablishment of stable glycemic control.

Autonomic neuropathy is also com-mon, though rarely severely symptom-atic. Autonomic neuropathy was a topicof focus in a recent technical review byVinik et al. (5) and will not be furtherdescribed here.Focal and multifocal neuropathies. Allthe neuropathies under this heading arerecognized as being more common inolder type 2 diabetic patients. Focal limbneuropathies are often, but not always,due to entrapment (e.g., carpal tunnelsyndrome), indicating the greater suscep-tibility of diabetic nerve to compression.Recent data suggest that there is a three-fold risk of having diabetes in 514patients with carpal tunnel syndromecompared with a normal control group(28). Among the cranial nerves, thosesupplying the external ocular musclesare most commonly involved. Thoraco-lumbar radiculoneuropathies maypresent with girdle-like pain, occasionallywith motor weakness of abdominal wallmuscles. Proximal motor neuropathy(amyotrophy) may be unilateral or asym-metrically bilateral with pain, wasting,and weakness that may be relatively acutein onset. All of these focal/multifocal neu-ropathies are discussed in greater detailbelow.

It seems probable that chronic inflam-matory demyelinating polyneuropathy(CIDP) occurs more commonly in peoplewith diabetes (24,29), although a case-control study has not been performed. Itsfeatures, differential diagnosis, and man-agement will be discussed in more detailbelow.

Table 1—Three classification systems for DNs

A: Clinical Classification of DNsPolyneuropathy Mononeuropathy

Sensory Isolated peripheral● Acute sensory● Chronic sensorimotor Mononeuritis multiplex

Autonomic● Cardiovascular Isolated peripheral● Gastrointestinal● Genitourinary Truncal● Other

Proximal motor (amyotrophy)TruncalAdapted from Boulton and Ward (22) and Boulton and Malik (11).

B: Patterns of Neuropathy in DiabetesLength-dependent diabetic polyneuropathy● Distal symmetrical sensory polyneuropathy● Large fiber neuropathy● Painful symmetrical polyneuropathy● Autonomic neuropathiesFocal and multifocal neuropathies● Cranial neuropathies● Limb neuropathies● Proximal DN of the lower limbs● Truncal neuropathiesNondiabetic neuropathies more common in diabetes● Pressure palsies● Acquired inflammatory demyelinating polyneuropathyAdapted from Said (25).

C: Classification of DNRapidly reversible● Hyperglycemic neuropathyGeneralized symmetrical polyneuropathies● Sensorimotor (chronic)● Acute sensory● AutonomicFocal and multifocal neuropathies● Cranial● Thoracolumbar radiculoneuropathy● Focal limb● Proximal motor (amyotrophy)Superimposed chronic inflammatory demyelinating neuropathyAdapted from Thomas (23,24).

Boulton and Associates

DIABETES CARE, VOLUME 27, NUMBER 6, JUNE 2004 1459

C. Consensus statements and stagingof neuropathyThere have been a number of consensusand committee reports relating to DN inthe last two decades, the best known ofwhich is probably the San Antonio Con-ference (20), which discussed definitions,measurements, and classification primar-ily for clinical research. The use of stan-dardized measurement techniques wasrecommended, and a further develop-ment conference was convened in 1992 toreview the standardization of proceduresand approaches used for epidemiologicaland clinical studies (30). Both of thesemeetings were jointly sponsored by theAmerican Diabetes Association and theAmerican Academy of Neurology.

An international group of experts inDN held a consensus meeting to developguidelines for the management of diabeticperipheral neuropathy by the practicingclinician (19). The agreed clinical stagesof DPN are shown in Table 2. This clinicalstaging is in general agreement with thatproposed by Dyck (21,31), for use in bothclinical practice and epidemiologicalstudies or controlled clinical trials. Thus,the clinical “no neuropathy” is equivalentto Dyck’s N0 or N1a; “clinical neuropa-thy” is equivalent to N1b, N2a, or N2b;and “late complications” is equivalent toDyck’s N3 (Table 3).

There have been a number of other

relevant reports, including two on mea-sures for use in clinical trials to assesssymptoms (32) and quantitative sensorytesting (QST) (33). Most recently, a com-mittee of the American Academy of Neu-rology reported on the use of QST forclinical and research purposes (34).

SECTION 3:PATHOGENESIS OF DN — Thi ssection will focus predominantly on thepathogenesis of DPN. Studies in animalmodels and cultured cells provide aconceptual framework for the cause andtreatment of DN (35). However, limited

translational work in diabetic patientscontinues to generate much debate andcontroversy over the cause(s) of humanDN, and to date we have no effectivetreatment.

A. HyperglycemiaLongitudinal data from the Rochester co-hort support the contention that the du-ration and severity of exposure tohyperglycemia are related to the severityof neuropathy only (36). Similarly, in astudy of newly diagnosed patients withtype 2 diabetes followed up from baselineand at 5 and 10 years, the overall severity,and not the development of neuropathy,was related to the degree of hyperglyce-mia (8). Recent studies in patients withimpaired glucose tolerance (IGT) provideimportant insights into the role of the de-gree of glucose dysmetabolism and thedevelopment of neuropathy. In a study ofpatients with IGT, the sural nerve ampli-tude and myelinated fiber density do notdiffer significantly from those with nor-mal glucose tolerance, suggestive of a gly-cemic threshold for the development ofneuropathy (37). However, of 121 pa-tients with a painful neuropathy and elec-trodiagnostic evidence of axonal injurytogether with epidermal nerve fiber ab-normalities, 25% had IGT (38). The neu-ropathy associated with IGT is milderthan the neuropathy associated withnewly diagnosed diabetes, and smallnerve fiber involvement may be the earli-est detectable sign of neuropathy (39).Improving hyperglycemia by more inten-sive insulin therapy (7) or pancreatictransplantation (40) improves electro-physiology in patients with type 1 diabe-tes. However, the evidence is not clear intype 2 diabetes. In the VA CooperativeStudy on Type 2 Diabetes Mellitus(VACSDM), 153 patients randomized tointensive versus conventional therapyachieved a 2.07% difference in HbA1cover 2 years, but failed to demonstrate asignificant difference in the progression ofeither somatic or autonomic neuropathy(41). Similarly, the more recent Steno-2Study failed to demonstrate a benefit ofmultifactorial intervention, including gly-cemic control, on measures of somaticneuropathy (42). The U.K. ProspectiveDiabetes Study (UKPDS) represents thelargest interventional study in type 2 dia-betes that has assessed the effects of im-proved glycemic control, but the

Table 2—Stages of DPN*

Stage of neuropathy† Characteristics

No neuropathy No symptoms or signsClinical neuropathy

Chronic painful Burning, shooting, stabbing pains with orwithout “pins and needles”; increasedat night; absent sensation to severalmodalities; reduced/absent reflexes

Acute painful Severe symptoms as above(hyperesthesiae common), may followinitiation of insulin in poorlycontrolled diabetes, signs minor orabsent

Painless with complete/partialsensory loss

Numbness/deadness of feet or nosymptoms, painless injury, reduced/absent sensation, reduced thermalsensitivity, absent reflexes

Late complications Foot lesions, neuropathic deformity,nontraumatic amputation

*Types of DN: frequent, sensorimotor symmetrical neuropathy (mostly chronic, sensory loss, or pain),autonomic neuropathy (history of impotence and possibly other autonomic abnormalities); rare, mono-neuropathy (motor involvement, acute onset, may be painful), diabetic amyotrophy (weakness/wastingusually of proximal lower-limb muscles). †Staging does not imply automatic progression to the next stage.The aim is to prevent, or at least delay, progression to the next stage.

Table 3—Staging severity of diabetic poly-neuropathy

N0: No objective evidence of DNN1: Asymptomatic polyneuropathy

N1a: No symptoms or signs butneuropathic test abnormalitiesN1b: Test abnormalities* plusneuropathy impairment on neurologicalexam

N2: Symptomatic neuropathyN2a: Symptoms, signs, and testabnormalityN2b: N2a plus significant ankledorsiflexor weakness

N3: Disabling polyneuropathy

Adapted from Dyck (21,31). *Nerve conduction,QST, or autonomic test abnormalities.

Diabetic somatic neuropathies


neuropathy data have not yet been re-ported (26).

B. Other pathogenetic mechanismsPolyol pathway. Animal models of dia-betes consistently demonstrate an associ-ation between increased flux through thepolyol pathway and a reduction in nerveconduction velocity (NCV), both ofwhich can be ameliorated with aldose re-ductase inhibitors (ARIs) (43). However,in humans the situation is not clear. Arecent study has demonstrated enhancedaldose reductase (AR) but minimal sorbi-tol dehydrogenase expression in the pe-ripheral nerve of diabetic patients (44). Inone of the earliest clinical studies, sorbitoland fructose levels were increased in onlyone-third of the sural nerve biopsies stud-ied and could not be related to clinical,neurophysiological, or pathological se-verity of neuropathy (45). Postmortemsciatic nerves from diabetic patients dem-onstrated a significant increase in glucose,fructose, and sorbitol compared with nor-mal subjects (46). In nerve obtained atamputation, glucose, fructose, and sorbi-tol were significantly higher in diabeticpatients than in nondiabetic patients (47),but concentrations differed markedlyfrom those in postmortem samples (46).In a recent study of patients with normalglucose tolerance, IGT, and type 2 diabe-tes, only the diabetic patients demon-strated an elevation in nerve sorbitol,indicating a glycemic threshold for activa-tion of this pathway (37). Linear regres-sion analysis has demonstrated asignificant inverse correlation betweennerve sorbitol and myelinated fiber den-sity (48). Moreover, it would appear thatthose at greatest risk of developing thecomplications are those with a higher setpoint for AR activity (49). Polymorphismsin the promoter region leading to a highlysignificant decrease in the frequency ofthe Z�2 allele have been demonstrated inpatients with overt neuropathy comparedwith those without neuropathy (50).

With regard to intervention, a meta-analysis of all randomized controlled tri-als of ARIs identified 19 trials, testing fourdifferent ARIs for 4–208 weeks (median24 weeks). It demonstrated a small butstatistically significant reduction in de-cline of median (0.66 m/s, 95% CI 0.18–1.14 m/s) and peroneal (0.53, 0.02–1.04)motor NCV without benefit in sensorynerves (51). The clinical relevance of aneffect on motor but not sensory nerve

function is questionable, especially as thelatter is primarily responsible for the mostcommon manifestations of human DN:severe pain and the insensate extremity,leading to ulceration and eventual ampu-tation (52).

Possible reasons for this marginalbenefit may be related to the lack of atargeted approach identifying those mostgenetically susceptible to alterations in ARactivity and, therefore, those most likelyto benefit from AR inhibition (50). Further-more, the degree of AR inhibition may de-termine the improvement observed. Thus,in a randomized, placebo-controlled, dou-ble-blinded, multiple-dose, clinical trialwith zenarestat, dose-dependent incre-ments in sural nerve sorbitol suppressionwere accompanied by significant improve-ment in NCV, and in doses producing�80% sorbitol suppression, there was a sig-nificant increase in the density of small-diameter myelinated fibers of the suralnerve (53). More recently, fidarestat, a po-tent ARI, significantly improved mediannerve F-wave conduction velocity and min-imal latency, as well as symptoms of numb-ness, spontaneous pain, paresthesiae, andhyperesthesia, in 279 diabetic patients (54).Myoinositol. While myoinositol defi-ciency has been proposed to play a role inthe pathogenesis of DN, there is little ev-idence to support this contention. In asural nerve biopsy study, myoinositol lev-els did not vary among patients with nor-mal glucose tolerance, IGT, and type 2diabetes (37).Glycation. Hyperglycemia results in theformation of advanced glycation endproducts (AGEs), which in turn act onspecific receptors (RAGEs), inducingmonocytes and endothelial cells to in-crease the production of cytokines andadhesion molecules (55). Glycation hasalso recently been shown to have an effecton matrix metalloproteinases (MMPs), inparticular MMP-2, which degrades typeIV collagen, but also on membrane type 1MMP, t issue inhibi tors of MMPs(TIMP)-1 and -2, and transforminggrowth factor-� (TGF-�) (56). Other ef-fects include prevention of epidermalgrowth factor–induced autophosphoryla-tion and activation of extracellular signal–regulated kinases (ERKs) (55). Inexperimental diabetes, these changes canbe prevented by AGE inhibitors, such asthe nucleophilic compounds pyridoxam-ine, tenilsetam, 2,3-diaminophenazone,or aminoguanidine (56,57). Alterna-

tively, the administration of recombinantRAGE hinders the AGE-RAGE interaction(56,57). Human sural nerves obtainedfrom diabetic and nondiabetic amputa-tion specimens demonstrate normal fu-rosine, an early reversible glycationproduct, but significantly elevated pento-sidine levels in both cytoskeletal and my-elin protein (58). Enhanced staining forcarboxymethyllysine in the perineurium,endothelial cells, and pericytes of endo-neurial microvessels, as well as myelin-ated and unmyelinated fibers, has beenshown to correlate with a reduction inmyelinated fiber density in peripheralnerve from five patients with type 2 dia-betes compared with five nondiabeticcontrol subjects (59). Pyrraline, an AGE,is also increased in postmortem samplesof optic nerve from diabetic patients (60).Intervention trials have focused on ne-phropathy, and no trial data are currentlyavailable for human DN.Oxidative stress. An increasing body ofdata supports the role of oxidative stress inthe pathogenesis of DN in animal models(35). Again, there is emerging evidence thatsingle-nucleotide polymorphisms of thegenes for mitochondrial (SOD2) and extra-cellular (SOD3) superoxide dismutasesmay confer an increased risk for the devel-opment of neuropathy (61). This may par-tially explain the lack of benefit observedwith a number of antioxidants. However,benefits have been observed with �-lipoicacid (LA), a powerful antioxidant that scav-enges hydroxyl, superoxide, and peroxylradicals and regenerates glutathione. In theALADIN II study, diabetic patients withsymptomatic polyneuropathy were ran-domly assigned to 5 days of intravenous LAfollowed by oral treatment for 2 years anddemonstrated a significant improvement insural sensory NCV, sensory nerve actionpotential (SNAP), and tibial motor NCV butnot neuropathic disability score (NDS)(62). ALADIN III randomized 509 diabeticpatients to LA intravenously for 3 weeks,followed by oral treatment compared withplacebo. It showed no change in the totalsymptom score, but did show an improve-ment in the neuropathy impairment scoreafter 3 weeks of intravenous therapy, whichwas maintained until the end of the study(63). Most recently, the SYDNEY study hasdemonstrated a significant improvement inthe neuropathy symptom score, neuropa-thy impairment score, and one attribute ofnerve conduction after daily intravenous



treatment with racemic LA for 5 days/weekfor 14 treatments (64).Vascular factors. The most direct evi-dence that improving tissue blood flowmay improve DN is derived from large-vessel revascularization studies, whichhave shown an improvement in NCV inone study (65) but not in another (66).However, a longer-term follow-up of thelatter study did show a prevention ofworsening of peroneal NCV (67). Thereare of course a number of pharmacologi-cal treatments that can achieve a similareffect. Angiotensin-converting enzyme(ACE) inhibitors mediate increased flow-dependent release of endothelium-derived relaxing factor (EDRF) andendothelium-dependent vessel relax-ation. In a double-blind, placebo-controlled clinical trial of trandaloprilover 12 months, peroneal motor NCV, M-wave amplitude, F-wave latency, andsural nerve amplitude improved signifi-cantly (68). Recently, the AppropriateBlood Pressure Control in Diabetes(ABCD) trial assessed the effects of inten-sive versus moderate blood pressure con-trol with either nisoldipine or enalapriland surprisingly failed to prevent pro-gression of DN, retinopathy, and neurop-athy (69).

1,2-Diacylglycerol (DAG) inducedactivation of protein kinase C (PKC); inparticular, PKC-� has been proposed toplay a major role in DN (35). However,even in nerve from diabetic animals, a fallin DAG levels and a consistent pattern ofchange in PKC activity has not been ob-served (70). Despite this, inhibition ofPKC-� in diabetic rats appears to correctreduced nerve blood flow and NCV (71).Based on the findings of a phase II clinicaltrial demonstrating some benefit in dia-betic patients with neuropathy (72), mul-ticenter, randomized, double-blind,placebo-controlled trials are under wayand are due to complete in 2005.

An increasing body of evidence sug-gests that conventional risk factors formacrovascular disease (such as derangedlipids) are also important in the pathogen-esis and progression of human DN (73).Recent studies show that hydroxymethyl-glutaryl CoA reductase inhibitors may en-hance endothelial cell nitric oxidebioavailabilty (76), prevent AGE-inducednuclear factor (NF)-��–induced pro-tein-1 activation and upregulation of vas-cular endothelial growth factor (VEGF)mRNA (75), and thereby ameliorate ex-

perimental DN (74). Simvastatin hasshown a trend toward slower progressionof neuropathy measured by vibration per-ception threshold (VPT) but no change inthe status of clinical neuropathy (76). Par-adoxically, as a cautionary note, recentobservational data suggest a link betweenchronic statin use and an increased risk ofperipheral neuropathy (77).Growth factors. Neurotrophins pro-mote the survival of specific neuronalpopulations by inducing morphologicaldifferentiation, enhancing nerve regener-ation, stimulating neurotransmitter ex-pression, and altering the physiologicalcharacteristics of neurons. Initially, theskin of diabetic patients with sensory fiberdysfunction demonstrated a depletion ofnerve growth factor (NGF) (78). How-ever, subsequent studies have shown asignificant increase in skin NGF mRNA(79) and neurotrophin-3 concentrations(80) and normal sciatic nerve ciliary neu-rotrophic factor (CNTF) levels (81). Insitu hybridization studies have also dem-onstrated an increased expression of bothtrkA, the high-affinity receptor for NGF,and trkC, the receptor for NT-3, in theskin of diabetic patients, which has beenproposed to reflect a compensatory re-sponse (82). Despite these apparentlycontradictory findings, a phase II clinicaltrial of recombinant human NGF in 250diabetic patients with symptomatic dia-betic polyneuropathy demonstrated a sig-nificant improvement in the sensorycomponent of the neurological examina-tion, in two quantitative sensory tests, andin a rather vague end point, “the clinicalimpression of most subjects that theirneuropathy had improved” (83). How-ever, a phase III trial in 1,019 diabeticpatients with sensory polyneuropathyfailed to demonstrate a significant benefit(84). More recently a randomized, double-blind, placebo-controlled study of brain-derived neurotrophic factor (rhBDNF) in30 diabetic patients demonstrated no signif-icant improvement in nerve conductionand quantitative sensory and autonomicfunction tests, including the cutaneousaxon reflex (85).Insulin-like growth factors. In culturedSchwann cells and the streptozotocin(STZ)-induced diabetic rat, insulin-likegrowth factor (IGF)-1 demonstrates aprotective effect via phosphatidylinositol(PI) 3-kinase in preventing glucose-mediated apoptosis (86). Both the STZ-diabetic and the BB/W rat develop severe

hyperglycemia and a deficiency in circu-lating IGF-1 levels and reproducibly de-velop neuroaxonal dystrophy (NAD) innerve terminals of the prevertebral sym-pathetic ganglia and the distal portions ofnoradrenergic ileal mesenteric nerves. Incontrast, the Zucker diabetic fatty (ZDF)rat, an animal model of type 2 diabetes,also develops severe hyperglycemia com-parable to that in the STZ- and BB/W-diabetic rats but maintains normal levelsof plasma IGF-1 and fails to demonstrateNAD in sympathetic ganglia and ilealmesenteric nerves as assessed by quanti-tative ultrastructural techniques (87).However, IGF-1 and IGF-1 receptormRNA levels have not been shown to dif-fer in the sural nerve of diabetic patientscompared with control subjects (88).C-peptide. Impaired insulin/C-peptideaction has emerged as a prominent patho-genetic factor. Preclinical studies havedemonstrated a range of actions that in-clude effects on Na(�)/K(�)-ATPase ac-tivity, endothelial nitric oxide synthase,expression of neurotrophic factors, regu-lation of molecular species underlying thedegeneration of the nodal apparatus intype 1 diabetic nerves, as well as DNAbinding of transcription factors and mod-ulation of apoptotic phenomena (89,90).These findings have recently been effec-tively translated into benefits in patientswith type 1 diabetes with the demonstra-tion of a significant improvement in suralsensory NCV and vibration perceptionbut without a benefit in either cold or heatperception after 12 weeks of daily subcu-taneous C-peptide treatment (91).VEGF. VEGF was originally discoveredas an endothelial-specific growth factorwith a predominant role in angiogenesis.However, recent observations indicatethat VEGF also has direct effects on neu-rons and glial cells, stimulating theirgrowth, survival, and axonal outgrowth(92). Thus, with its potential for a dualimpact on both the vasculature and neu-rons, it could represent an importanttherapeutic intervention in DN. Both theSTZ-induced diabetic rat and the alloxan-induced diabetic rabbit have demon-strated restoration of nerve vascularity,blood flow, and both large- and small-fiber dysfunction 4 weeks after intramus-cular gene transfer of plasmid DNAencoding VEGF-1 or VEGF-2, with con-firmed constitutive overexpression ofboth transgenes in tissue (93). In contrast,immunohistochemistry of sciatic nerves



and dorsal root ganglia from STZ-induceddiabetic rats demonstrates intense VEGFstaining in cell bodies and nerve fibers,whereas controls express no or very littleVEGF, and animals treated with insulin orNGF show significantly lower immuno-staining for VEGF (94). Thus, there is anintrinsic capacity to upregulate VEGF,but this appears insufficient and may re-quire exogenous delivery possibly viagene therapy. A phase I/II, single-site,dose-escalating, double-blind, placebo-controlled study to evaluate the safety andimpact of phVEGF165 gene transfer onsensory neuropathy in patients with dia-betes with or without macrovascular dis-ease involving the lower extremities iscurrently under way and will involve 192patients over a period of 4 years (95).Immune mechanisms. Studies suggestthat sera from type 2 diabetic patientswith neuropathy contains an autoim-mune immunoglobulin that inducescomplement-independent, calcium-dependent apoptosis in neuronal cells(96). The expression of these cytotoxicfactors has been related to the severity ofneuropathy and the type of neuronal cellkilled (97). Thus, it has been suggestedthat such toxic factors may contribute toDN by acting in concert with hyperglyce-mia to damage sensory/autonomic neu-rons (97).

SECTION 4: FOCAL ANDMULTIFOCALNEUROPATHIESA number of rare neuropathies referred toas the focal and multifocal neuropathieswill now be discussed in detail, particu-larly with regard to clinical presentationand management.

Factors leading to the development ofthe compression neuropathies relate to ei-ther the peripheral nerve itself or thestructures surrounding it at the point ofcompression. Many of the animal modelsof entrapment are based on acute com-pression, the mechanics of which differconsiderably in terms of nerve stretch,tethering, and associated ischemia in-volved in entrapment. Therefore, transla-tion of these experimental findings toentrapment neuropathies in diabetic pa-tients should be interpreted with caution.

A. MononeuropathiesCarpal tunnel syndrome. This is themost common entrapment neuropathyencountered in diabetic patients and oc-

curs as a result of median nerve compres-sion under the transverse carpal ligament.Idiopathic carpal tunnel syndrome (CTS)occurs in patients with rheumatoid arthri-tis, hypothyroidism, and obesity. In 20–30% of diabetic patients, it can bedemonstrated electrophysiologically butpresents as a clinically relevant problemin �5.8% (98). Painful paresthesiae of thefingers may progress to a deep-seatedache, which radiates up the forearm or,very rarely, the arm. This occurs primarilyat night but may be initiated during theday by repetitive flexion and extension ofthe wrist. Motor weakness is uncommonbut thenar muscle wasting occurs partic-ularly in the elderly. Two common clini-cal tests include the Phalen (forearms heldvertically and hands held in completeflexion for 1 min; positive if paresthesiaedevelop in the median territory within30 s) or Tinel (percussion at the wrist andpalm induces paresthesiae in the mediannerve territory) tests, but they have a highfalse-positive rate. Electrophysiologicalstudies measure the speed of conductionacross the carpal tunnel, and median sen-sory nerve conduction studies are com-pared with radial and/or ulnar sensorylatencies. Their interpretations are madedifficult if there is a coexisting peripheralneuropathy affecting the upper limbs or ifthere are no symptoms of CTS. Demyeli-nation is thought to be the primary patho-logical abnormality. Treatment optionsinclude wrist splints, which have limitedapplication because they cannot be wornduring the day, but can be effective fornocturnal symptoms. Injections of corti-sone into the carpal tunnel may provideshort-lived relief, and in the majority ofcases, repeat injections are required. Sur-gical sectioning of the transverse carpalligament provides variable degrees of painrelief but does not particularly benefitmuscle wasting or sensory loss.Ulnar neuropathy. The second mostcommon entrapment neuropathy (2.1%)occurs as a result of ulnar nerve compres-sion immediately distal to the ulnargroove beneath the edge of the flexorcarpi ulnaris aponeurosis in the cubitaltunnel. It may develop as a result of de-formity at the elbow joint secondary tofracture or as a consequence of prolongedpressure during surgery, and it has beenmost commonly associated with alcohol-ism. Typical symptoms include painfulparesthesiae in the fourth and fifth digitsassociated with hypothenar and in-

terosseous muscle wasting. The pathol-ogy is a combination of demyelinationand axonal degeneration. The key electro-physiological findings include low ampli-tude ulnar sensory nerve act ionpotentials, reduced sensory NCV, and fi-brillation potentials in the interossei (99).Management of patients is primarily con-servative, with advice to avoid pressure tothis area, as the results of surgery are verypoor. However, if symptoms and signsprogress, then a number of approachesmay be used: medial epicondylectomy,transaction of the flexor carpi ulnaris apo-neurosis, and ulnar nerve transposition(100).Radial neuropathy. Radial neuropathyis rare (0.6%), occurring as a consequenceof radial nerve compression in the spiralgroove. It presents with the characteristicmotor deficits of wrist drop with very oc-casional sensory symptoms of paresthe-siae in the dermatomes supplied by thesuperficial radial nerve. Causes of idio-pathic radial neuropathy include humeralfracture, blunt trauma over the postero-lateral aspect of the arm, and externalcompression. Electrophysiological assess-ment demonstrates a predominant effecton amplitude rather than the conductionvelocity, suggestive of predominantlyaxonal degeneration associated with sec-ondary demyelination (101). Manage-ment is conservative with pressure relief.Common peroneal neuropathy. This isthe most common of all limb mononeu-ropathies. Involvement of the motor fi-bers in the common peroneal nerveresults in weakness of the dorsiflexors and“foot drop,” but loss of the motor supplyto the tibialis anterior muscle also leads toweakness in eversion. This is accompa-nied by a sensory deficit but characteris-tically no pain or paresthesiae. Diabetes isa relatively uncommon cause (5–12% ofcases) of peroneal nerve palsy (102).Common causes include external com-pression at the fibular head during anes-thesia (in bed-ridden patients) andinappropriately placed plasters followinglower-limb fractures. An important differ-ential is a radiculopathy involving the L5root. Features that define L5 involvementinclude pain in the lower back and addi-tional loss of inversion. Electrophysiolog-ical studies suggest demyelination withconduction block in mild lesions with amarked loss of amplitude presumablysecondary to axonal degeneration in moresevere lesions (103). Because the majority



of these lesions are caused by externalpressure that, if relieved, will result in res-olution of the motor deficit within 3–6months, a conservative approach is advo-cated with removal of pressure and a footbrace in the interim.Lateral femoral cutaneous neuropathy.Compression of the lateral femoral cuta-neous nerve (meralgia paraesthetica) isuncommon and results in pain, paresthe-siae, and sensory loss in the lateral aspectof the thigh (104). Obesity is the mostcommon cause, followed by trauma dueto external injury of the nerve, as it runsdown the lateral aspect of the thigh. Mostwill resolve spontaneously and are there-fore managed conservatively.

Other nerves that may be involved in-clude the sciatic and obturator nerves.They can be a cause of significant motordeficit; however, they are extremely rareand their management is conservative.

B. Cranial neuropathiesCranial neuropathies in diabetic patientsare extremely rare (0.05%) and occur inolder individuals with a long duration ofdiabetes (105).Ocular neuropathies. Cranial nervesIII, IV, and VI are affected, and amongdiabetic patients, the relative frequency isoculomotor (3.3%) and abducent (3.3%)nerve occurring with equal and greaterfrequency than the trochlear nerve (2.1%)(106). The classical presentation of ocu-lomotor nerve palsy is that of an acute-onset diplopia with ptosis and pupillarysparing associated with ipsilateral head-ache. While pupillary sparing is oftenquoted as a means of differentiating dia-betic from other structural (aneurysm, tu-mor, or mass) ophthalmoplegias, 14 –18% of diabetic patients do developpupillary dysfunction (107). Resolutionof neurological deficits occurs over �2.5months and recurrence can occur in 25%of patients (107). A clear understandingof the underlying pathology and patho-genesis of this condition is limited due tothe difficulties in obtaining tissue in suchpatients. Based on four single postmortemcase reports demonstrating centrofascicu-lar pallor of myelin staining in paraffin-embedded sections, it is assumed thatfocal demyelination occurs secondary toischemia (108). This is supported bymore recent studies in plastic sections of-fering clearer pathological detail (109).However, this conclusion should be inter-preted with caution as acute ischemic in-

jury should result in axonal degeneration.Furthermore, pallor in staining shouldnot be interpreted as demyelination, asthis has not been confirmed using teasedfiber analysis. A recent study of oculomo-tor nerve specimens from 8 diabetic pa-tients without oculomotor nerve palsycompared with 15 nondiabetic patientshas shown subperineurial as opposed tocentrofascicular alteration. Microfascicu-lation has been demonstrated and sug-gests chronic injury due to ischemia(110). Management is expectant withstrong reassurance to the patient for re-covery. Maintaining optimal glycemiccontrol as well as minimizing the otherstronger risk factors for ischemia, includ-ing hypertension and hyperlipidemia,may aid recovery.Facial neuropathy. In most series of id-iopathic facial neuropathy or Bell’s palsy,diabetes is well represented, ranging from6% (111) to 48.8% (112). The main neu-rological findings are those of acute-onsetunilateral weakness of facial muscles,widening of the palpebral fissure, and sec-ondary corneal irritation. This is accom-panied by varying degrees of disturbancein taste and hyperacusis. The presence ofhypertension and severity of paralysis atonset, but not diabetes, determines thedegree of recovery at 1 year (113). Neu-rophysiological studies demonstrate re-duced or absent compound muscle actionpotentials (CMAPs) in the nasalis muscle,which can actually be used to determineoutcome. Thus, CMAP �30% results in90–100% recovery, CMAP 10–30% re-sults in �50% recovery, and CMAP�10% results in virtually no recovery.This is associated with prolongation of theR1 and R2 latencies of the trigeminal“blink reflex” (114). If the presentation isacute for �1 week, 7–14 days of pred-nisone may be administered but with at-tention to optimizing glycemic control.Other cranial nerves. Other cranialnerves may be affected in diabetes. How-ever, their relatively infrequent involve-ment warrants an awareness of theiroccurrence but will not be discussed indetail. Thus, olfactory and optic nerve in-volvement has been described. More re-cently, corneal confocal microscopy hasbeen used to show significant degenera-tion of small myelinated and unmyeli-nated fibers in the cornea of diabeticpatients with increasing neuropathic se-verity (115). There are also reports of anincreased frequency of trigeminal neural-

gia in diabetic patients. Hearing loss as aresult of VIII nerve involvement has alsobeen described (116). Vagal nerve in-volvement manifests as part of diabeticautonomic neuropathy. Vocal cord paral-ysis has also been attributed to recurrentlaryngeal nerve involvement.

C. Diabetic amyotrophyClinical features. Diabetic amyotrophytypically occurs in patients with type 2diabetes aged 50–60 years and presentswith severe pain and uni- or bilateralmuscle weakness and atrophy in theproximal thigh muscles (117).Pathogenesis. Factors contributing tothe development of diabetic amyotrophy(proximal motor neuropathy) (118,119)are poorly understood. Somewhat polar-ized views have evolved. One proposalhas implicated ischemia based on an earlycase report that demonstrated infarcts inthe proximal femoral, sciatic, and obtura-tor nerves and lumbosacral plexus (120);other investigators have proposed a met-abolic basis for a more subacute, symmet-rical disorder affecting the distal branchesof the proximal motor nerves (121). Aconciliatory view is derived from reportsof patients with an initial rapid evolutionwith subsequent slow progression ofsymptoms and signs over several months,indicating a combination of both vascularand metabolic factors (117,122).Neurophysiology. Needle electrodesampling reveals different responses de-pending on the stage of this condition. Inthe early stages, spontaneous fibrillationand reduced motor unit recruitment oc-cur, suggestive of denervation. In laterstages, there is an increase in amplitude ofthe motor unit potential, indicating rein-nervation via collateral sprouting. Elec-trophysiological studies demonstrate areduction in femoral NCV (123). Addi-tionally, however, femoral nerve stimula-tion produces an attenuated compoundmuscle action potential of the quadricepsmuscle, supporting the occurrence of ax-onal pathology (124).Pathology. Recent reports have demon-strated an epineurial vasculitis in the in-termediate cutaneous nerve of the thigh(ICNT) in a proportion of patients withamyotrophy (124–126). Centrofasciculardegeneration of the ICNT has been ob-served in association with an inflamma-tory infiltrate and occlusion of epineurialblood vessels (124,125). These studieshave highlighted the heterogeneous na-



ture of myelinated and unmyelinated fi-ber damage as some patients showed analmost complete loss of fibers, whereasothers demonstrated a moderate reduc-tion with regeneration both proximally inthe ICNT and distally in the sural nerve(124,125). Another study demonstrated areduction or absence of CMAPs in the tib-ial and peroneal nerves and also in thesensory action potential of the suralnerve, with relative preservation of NCVs,suggestive of axonal degeneration ratherthan demyelination (127). This was con-firmed on pathological examination of thesural nerve, which demonstrated multifo-cal fiber loss and dystrophic fibers withpredominantly axonal degeneration andsecondary demyelination associated withabortive regeneration represented by theformation of microfasciculi and perineur-ial scarring (125). One may question therelevance of distal findings in the suralnerve for a condition that affects proximallumbosacral nerve segments. Thus, incontrast, a recent study in the intermedi-ate cutaneous nerve of 15 patients withdiabetic amyotrophy has shown multifo-cal fiber loss, but teased fiber studies haveprimarily demonstrated demyelination(128).

The other major finding that has pro-vided a novel perspective on the patho-genesis of this condition has been derivedfrom immunohistological studies. In thestudy by Said et al. (126), all ICNT biop-sies demonstrated an inflammatory infil-trate composed of B- and T-cells andoccasional macrophages. In anotherstudy, which included 12 patients withdiabetic amyotrophy, the sural nervedemonstrated a mononuclear cell infil-trate in 4 patients and a perivascular infil-trate of activated T-cells expressing bothinterleukin (IL)-2 and major histocom-patibility complex class II antigens in 6patients (129). There was, however, noevidence of infiltration with B-cells orpolymorphonuclear cells. The majority ofnerves from patients also showed stainingfor tumor necrosis factor (TNF)-�, IL-6,and IL-1�. Furthermore, C3d and C5b-9complement protein was found withinendoneurial and epineurial blood vesselwalls in all patients (129). Dyck et al.(127) demonstrated epineurial vascularand perivascular mononuclear inflamma-tory infiltrates, which stained positive forleukocytes. There were also additionalfeatures of a necrotizing vasculitis (arte-riolar, venular, and capillary wall infiltra-

tion with inflammatory cells) withhemosiderin deposition. Together thesechanges suggested a microscopic vasculi-tis, and levels of IL-1� and IL-6 were in-creased in these patients (127). A recentstudy has demonstrated a polymorpho-nuclear vasculitis with transmural infil-tration of postcapillary venules with IgMdeposition along the endothelium as wellas in the endoneurium and subperineur-ial regions, indicating increased perme-ability and immune-mediated nervedamage (128). There was additional dep-osition of activated complement (C5b-9)in the same areas, indicating an active im-mune-mediated vasculitis.Management. Current therapies lack arobust evidence base to support the use ofany therapy, because the rarity of thiscondition precludes controlled clinicaltrials. The main aim of therapy is to con-trol pain, and this can be achievedthrough nonsteroidal antiinflammatoryagents (ibuprofen and naproxen), opioids(codeine phosphate and morphine elixir),or tricyclic antidepressants (amitriptylineand imipramine). Other agents that maybe useful include tramadol and gabapen-tin (see Section 5B, no. 8). Measures thatputatively affect the underlying pathologyinclude an improvement in glycemic con-trol. This has been particularly advocatedin patients on oral hypoglycemic agentswho are recommended for conversion toinsulin therapy. Based on the observa-tions of vasculitis in a proportion ofpatients, immunosuppressive therapyhas been recommended using initial in-travenous, followed by high-dose oralcorticosteroids, or intravenous immuno-globulin. Reports of a dramatic improve-ment in neurological function (130) haveresulted in the initiation of a multicenterclinical trial of immunotherapy in theU.S., due to report in 2004 (131).

D. Diabetic truncalradiculoneuropathyClinical features. Diabetic truncal ra-diculoneuropathy affects middle-aged toelderly diabetic patients and appears tohave a predilection for men. Pain is a pri-mary feature and is acute in onset but mayevolve over several months. It is aching orburning in quality, may be superimposedwith lancinating stabs, and demonstratesnocturnal exacerbation with cutaneoushyperesthesia. It occurs in a girdle-likedistribution over the lower thoracic or ab-dominal wall, usually unilateral but

sometimes bilaterally. On rare occassions,it may result in motor weakness withbulging of the abdominal wall (11). Pro-found weight loss may accompany the on-set of symptoms. Clinical examinationdemonstrates heterogeneous neurologicalfindings ranging from no abnormality tosensory loss and hyperesthesia in a com-plete dermatomal pattern, but may some-times just involve the distribution of theventral or dorsal rami (132). Resolution ofsymptoms generally occurs within 4–6months.Pathogenesis. Clinically, this conditionbares strong similarities to diabetic amy-otrophy, but due to the lack of patholog-ical studies, its pathogenesis is basedmore on inference than on actual evi-dence. The acute onset suggests a vascularcause, although its occurrence in patientswith generally poorer glycemic controlsuggests a metabolic basis.Electrophysiology. Electromyographydemonstrates denervation potentials inthe intercostal, anterior abdominal wall,and paraspinal muscles (133). There areno reported conduction studies of the in-tercostal nerves in this condition.Pathology. There are no pathologicalstudies on this condition. Therefore, onemay only infer the site of the lesion. Inthose patients demonstrating denervationof the paraspinal muscles, a lesion of thedorsal primary rami is probable (134).However, those who do not demonstratethis feature may demonstrate lesionsmore distally in the intercostal or subcos-tal nerves. From the sensory deficits, it isclear that the lesions may vary and involveposterior primary rami of the spinalnerves or intercostal nerves (132). Thecause may be ischemia, and the contigu-ous dermatomal involvement may be ex-plained by the occlusion of a singleintercostal artery supplying several trun-cal nerves.Management. There is no evidence tosupport the use of any therapy. Becausethe natural history is for spontaneous res-olution within 4–6 months, and as thereare strong similarities to diabetic amyot-rophy, the approach to management isvery similar to that for the latter. The mainaim of therapy is to control pain. Again,an improvement in glycemic control hasbeen advocated, as has immunosuppres-sive therapy with corticosteroids or intra-venous immunoglobulin.



E. CIDPDiabetic patients occasionally developclinical and electrodiagnostic featuressuggestive of CIDP (29). It is important torecognize that this subgroup, unlike thosewith diabetic polyneuropathy, is treatable(135). Many of the clinical, electrophysi-ological, and nerve biopsy criteria are notsufficiently helpful in the differential di-agnosis of these two conditions. However,when an unusually severe and progressivepolyneuropathy develops in diabetic pa-tients, one must consider CIDP.

Although electrodiagnosis is an im-portant element in the diagnosis of CIDP,current electrodiagnostic criteria aloneappear to be insufficient for definingmany cases of CIDP and therefore cer-tainly should not be relied on to differ-entiate from diabetic polyneuropathy(136).

Nerve biopsies in CIDP demonstratesegmental demyelination and remyelina-tion, onion bulbs, and inflammatory infil-trates, but they are also present in diabeticpolyneuropathy (137). The presence ofincreased numbers of macrophages indi-cating a macrophage-associated demyeli-nation may be helpful, as this is acharacteristic feature of CIDP not ob-served in diabetic polyneuropathy (137).

Treatment of CIDP requires long-term immunomodulatory therapy withcombinations of corticosteroids, azathio-prine, plasmapheresis, and intravenousimmune globulin but does produce rela-tively rapid and substantial improvementin neurological deficits and electrophysi-ology (138,139).

SECTION 5: DISTALSYMMETRICALPOLYNEUROPATHY

A. Acute sensory neuropathyAcute sensory (painful) neuropathy is a dis-tinctive variant of DPN that warrants a sep-arate discussion (23,24,140). Althoughmany of the symptoms of acute sensory andchronic sensorimotor neuropathy are simi-lar, there are clear differences in the mode ofonset, accompanying signs, and prognosis,which are summarized in Table 4 (140–142). Pain is the outstanding complaint inall patients, who also experience severeweight loss, depression, and, frequently inmales, erectile dysfunction. Common com-plaints include constant burning discom-fort (especially in the feet), severehyperesthesiae, and deep aching pain, andmany experience sudden, sharp, stabbing,or “electric shock”–like sensations in thelower limbs. All symptoms are prone tonocturnal exacerbation, with bed clothes ir-ritating hyperesthetic skin. Clinical exami-nation is usually relatively normal, withallodynia on sensory testing, a normal mo-tor exam, and occasionally reduced anklereflexes.

Acute painful neuropathy is associ-ated with poor glycemic control and mayfollow an episode of ketoacidosis; it hasalso been associated with weight loss andeating disorders (143).

Conversely, it may develop after sud-den improvement of glycemic control; theterm “insulin neuritis” is unfortunate, as itmay follow improvement of glycemiccontrol induced by oral hypoglycemicagents. Both of these observations are in

keeping with the hypothesis that bloodglucose flux is important in the genesis ofneuropathic pain (10).

Sural nerve biopsies have been per-formed in patients with acute painful neu-ropathy (141,144) and show activedegeneration of both myelinated and un-myelinated fibers. No correlations weredemonstrated between pain and either ac-tive degeneration of myelinated fibers orregenerative activity in myelinated or un-myelinated axons. Thus, it is difficult toreconcile these findings with the sugges-tion that acute painful neuropathy is anexample of a “small-fiber neuropathy.”Indeed, a recent review concluded thatpainful neuropathy is not restricted to se-lective involvement of small or large fibers(145). There is also a suggestion thatacute painful neuropathy may be relatedto neural ischemia precipitated by suddenimprovement of glycemic control. Usingin vivo epineurial vessel photography andfluorescein angiography, Tesfaye et al.(146) demonstrated severe abnormalitiesof epineurial vessels in acute painful neu-ropathy, with arteriovenous shunting andproliferating neural “new vessels” that re-sembled new vessels seen in retinopathy.One hypothesis for the genesis of neuro-pathic pain in such cases is that suddenchanges in blood glucose control result inalterations in their blood flow, leading to a“steal” effect with arteriovenous shuntingthus rendering the endoneurium ischemic.

In the management of this condition,achieving stable blood glucose control ismost important: stability may well be thekey feature as blood glucose flux (as as-sessed by the “M” value) is associated withpain (10,147). Additionally, most pa-tients require medications for their neu-ropathic pain (these medications aredetailed below). The natural history ofthis condition is very different from themuch more common chronic sensorimo-tor neuropathy: its onset is acute or sub-acute, but the severe symptoms resolve inless than a year (140,141).

B. Chronic sensorimotor neuropathy(DPN)Chronic sensorimotor neuropathy is themost common manifestation of the DNsthat is usually insidious in onset and maybe the presenting feature in people withtype 2 diabetes (10,26). Many patients areasymptomatic, and a neurological deficitmay be discovered by chance during aroutine neurological exam, although they

Table 4—Contrasts between acute sensory and chronic sensorimotor neuropathies

Acute sensory Chronic sensorimotor

Mode of onset Relatively rapid Gradual, insidiousSymptoms Severe burning pain

Aching: weight loss usualBurning pain, paresthesiae,

numbness; weight lossunusual

Symptom severity �� 0 to ��Signs Mild sensory in some:

motor unusualStocking and glove sensory loss:

absent ankle reflexesOther diabetic complications Unusual Increased prevalenceElectrophysiological

investigationsMay be normal or minor

abnormalitiesAbnormalities unusual in motor

and sensory nervesNatural history Complete recovery within

12 monthsSymptoms may persist

intermittently for years: atrisk of foot ulceration



may even present with a neuropathiccomplication such as a painless foot ulcer.It is a length-dependent process, and itssensory manifestations are most pro-nounced in the lower limbs and, in moresevere cases, in the fingers and hands.

The following subsections will focuson the clinical presentation and assess-ment of chronic sensorimotor neuropathy;methods of quantitative sensory testing,electrophysiological study, and other meth-ods of assessment; and treatments ofchronic sensorimotor neuropathy.1. Clinical presentation of chronic sen-sorimotor neuropathy (DPN).Symptoms. DPN occurs in both type 1 andtype 2 diabetes and is more common withincreasing age and duration of diabetes.These symptoms tend to be intermittentand of similar character but with lesserintensity than those described underpainful neuropathy. In a large populationsurvey, Harris et al. (148) reported that30% of type 1 diabetic patients and 36%of male and 40% of female type 2 diabeticpatients experienced neuropathic symp-toms. However, 10% of males and 12% offemales in the nondiabetic population re-ported similar symptoms.

As in acute sensory neuropathy, pain-ful symptoms tend to be more pro-nounced at night, but in addition,patients with DPN may experience “neg-ative” symptoms such as numbness or“feet feel dead.” Patients often find it dif-ficult to describe the symptoms as theyare different than the pain that they havepreviously experienced. Though not of-ten mentioned in older texts, unsteadi-ness is increasingly being recognized as amanifestation of DPN, due to disturbedproprioception and possibly abnormalmuscle sensory function (149). Such un-steadiness has been quantified (150–152)and may result in repetitive minor traumaor falls and in late complications such astrauma or Charcot’s neuroarthropathy.

Signs. On clinical examination, thereis usually a symmetrical sensory loss to allmodalities in a stocking distribution. Insevere cases, this may extend well abovethe ankle and also involve the hands. Theankle reflexes are usually reduced or ab-sent, and the knee reflexes may also beabsent in some cases. Motor weakness isunusual, although small muscle wastingin the feet and also the hands may also beseen in more advanced cases. Any pro-nounced motor signs should raise thepossibility of a nondiabetic etiology of the

neuropathy, especially if asymmetrical(1,19).

In more severe cases, with loss of pro-prioception, patients may demonstrate apositive Romberg’s sign.

As DPN is often accompanied by dis-tal (sympathetic) autonomic neuropathy(140), signs of autonomic dysfunction areoften apparent on examinations: thesemight include warm dry skin (in the ab-sence of peripheral vascular disease) andthe presence of plantar callus under pres-sure-bearing areas. The “at-risk” foot forneuropathic ulceration might also have ahigh arch (pes cavus) and clawing of thetoes (153). However, it must be empha-sized that all patients with DPN with orwithout obvious foot deformities must beconsidered as being at risk of neuropathiccomplications, such as Charcot’s neuro-arthopathy or foot ulceration (27,153).2. Clinical assessment of DPN.Symptoms. As noted above, many patientshave difficulty in describing the symp-toms of neuropathy. Pain and paresthe-siae are personal experiences, and there ismarked variation in the description ofsymptoms between individuals with sim-ilar pathological lesions. This has impor-tant implications for the assessment ofsymptoms: Huskisson (154) clearlystated that “[p]ain is a personal psycho-logical experience and an external ob-server can play no part in its directmeasurement.” When recording symp-toms in clinical practice, physicians musttherefore avoid the temptation to “inter-pret” or “translate” patient reports; in-stead, they should record the patient’sdescription verbatim.

A number of simple symptom screen-ing questionnaires are available to recordsymptom quality and severity. A simpli-fied neuropathy symptom score that wasused in the European prevalence studiescould also be useful in clinical practice(2,4). The Michigan Neuropathy Screen-ing Instrument (MNSI) is a brief 15-itemquestionnaire that can be administered topatients as a screening tool for neuropa-thy (155). Other similar symptom scoringsystems have also been described (156).

Simple visual analog or verbal de-scriptive scales may be used to follow pa-tients’ responses to treatment of theirneuropathic symptoms (156–158). How-ever, it must always be remembered thatidentification of neuropathic symptoms isnot useful as a diagnostic or screening tool

in the assessment of DN, as shown byFranse et al. (159).

It is well recognized that both symp-toms and deficits may have an adverse ef-fect on quality of life (QOL) in DN (160).The NeuroQol, a recently developed andvalidated QOL instrument, also includesa symptom checklist and may be used asan outcome measure in future clinicalstudies (161).

Signs. The use of composite scores toassess clinical signs was pioneered byDyck and colleagues (21,162), who firstdescribed the NDS and later the Neurop-athy Impairment Score (NIS). A modifiedNDS has been used in several large studies(2,4,52) and can also be used in the com-munity by a trained nonspecialist (Fig. 1).It has been shown to be the best predictorof foot ulceration and the best neuro-pathic end point in a large prospectivecommunity study (52). The maximumNDS is 10, with a score of 6 or more beingpredictive of foot ulcer risk.

Similarly, the Toronto group (163)have described a number of simplescreening tests for the diagnosis of neu-ropathy in outpatient clinics. They haverecently validated this clinical scoring sys-tem (164) and concluded that it can beused to document and monitor neuropa-thy in the clinic. Looking to the future,Dyck et al. (165) recently reported elec-tronic case-report forms for the recordingof symptoms and signs of neuropathy thatmight be useful in the longitudinal fol-low-up of neuropathic patients.

Whatever methodology is used in theassessment and documentation of neuro-pathic signs, it should be noted that theneurological exam of the lower limbs isthe important aspect in the clinical diag-nosis of DN (166).

Simple devices for clinical screening.The dividing line between simple devicesused in daily clinical practice and QST isdifficult to define. For the purposes of thisreview, QST will be defined as proceduresrequiring a power source where the inten-sity and characteristics of the stimuli arewell controlled and where the detectionthreshold is determined in parametricunits that can be compared with estab-lished “normal” values (167).

Although the simple handheldscreening devices are less sensitive thanthe more sophisticated QST devices de-scribed below, they have the advantage ofbeing relatively inexpensive, easy to oper-



ate, and easily portable; therefore, theiruse in clinical practice is increasing.

The most widely used device in clin-ical practice is the Semmes-Weinsteinmonofilament (168,169). The filamentassesses pressure perception when gentlepressure is applied to the handle sufficientto buckle the nylon filament. Althoughfilaments of many different sizes are avail-able, it is the one that exerts 10 g of pres-sure, the value most commonly used toassess pressure sensation in the diabeticfoot. It is also referred to as the 5.07monofilament because, during calibra-tion, the filaments are calibrated to exert aforce measured in grams that is 10 logof the force exerted at the tip; hence, 5.07exerts 10 g of force.

A number of cross-sectional studieshave assessed the sensitivity of the 10-gmonofilament to identify feet at risk ofulceration. Sensitivities vary from 86 to100% (170–172), although there is noconsensus as to how many sites should betested. The most common algorithm rec-ommends four sites per foot: generally thehallux and metatarsal heads 1, 3, and 5(169). However, the most recent study(172) suggested that there is little advan-tage gained from multiple site assess-ments. There is also no universalagreement as to what constitutes an ab-normal result (i.e., one, two, three, or fourabnormal results from the sites tested).

Despite these problems, the 10-g mono-filament is widely used for the clinical as-sessment of neuropathy.

A final caution on the use of the fila-ments: Booth and Young (173) identifiedthat filaments manufactured by certaincompanies do not actually buckle at 10 gof force. Indeed, several tested filamentsbuckled at �8 g. Thus, care must be takenwhen selecting suppliers of filaments.

The graduated Rydel-Seiffer tuningfork is used in some centers to assess neu-ropathy (174,175). This fork uses a visualoptical illusion to allow the assessor todetermine the intensity of residual vibra-tion on a 0–8 scale at the point of thresh-old (disappearance of sensation). Hilz etal. (174) reported that results with thisinstrument correlated well with otherQST measures.

The tactile circumferential discrimi-nator assesses the perception of calibratedchange in the circumference of a probe (avariation of two-point discrimination).Vileikyte et al. (176) reported a 100%sensitivity in the identification of patientsat risk of foot ulceration. Similarly, thisdevice also demonstrated good agreementwith other measures of QST.

Finally, the recently reported Neuro-pen is a clinical device that assesses painusing both a Neurotip at one end of the“pen” and a 10-g monofilament at theother end. This was shown to be a sensi-

tive device for assessing nerve functionwhen compared with the simplified NDS(177).3. QST. The progressive loss, or change,in sensation is the hallmark of DPN. QSTmeasures can be used to identify the sen-sory modalities affected and to estimatethe magnitude of the deficit. In the dia-betic population, vibration, thermal, andpain thresholds have proven valuable inthe detection of subclinical neuropathy(178,179), in tracking the progression ofneuropathy in large cohorts (180,181),and in predicting patients “at risk” for footulceration (182,183). In addition, QSTmeasures have played a key role as pri-mary efficacy end points in a series ofmulticenter clinical trials evaluating theprevention or treatment of diabetic poly-neuropathy (74,91).

The strengths of QST are well docu-mented (rev. in 167) and include 1) theaccurate control of stimulus characteris-tics; 2) the ability to assess multiple mo-dalities; 3) the use of well-establishedpsychophysical procedures to enhancesensitivity; 4) the capacity to measurefunction over a wide dynamic range ofintensities, thus supporting the evalua-tion of multiple degrees of neuropathy; 5)the ability to measure sensation at multi-ple anatomical sites, enabling the explo-ration of a potential distal-to-proximalgradient of sensory loss; and 6) for most

Figure 1—The modified NDS.



measures, the availability of data fromlarge, age-matched, “normal” comparisongroups. The limitations of QST are alsoclear. No matter what the instrument orprocedure used, QST is only a semiobjec-tive measure, affected by the subject’s at-tention, motivation, and cooperation, aswell as by anthropometric variables suchas age, sex, body mass, and history ofsmoking and alcohol consumption(184,185). Expectancy and subject biasare additional factors that can exert apowerful influence on QST findings(186). Further, QST is sensitive tochanges in structure or function along theentire neuroaxis from nerve to cortex; it isnot a specific measure of peripheral nervefunction (167).

There have been several reviews ofQST procedures (167,187–189) andseveral “consensus expert panels” have con-sidered the value of QST as a method ofassessing sensory neuropathy (20,30,33,34). A discussion of the merits of specificinstruments or testing algorithms is beyondthe scope of this review. However, it is note-worthy that a recent study comparing VPTsusing two very different instruments andprocedures reported similar sensitivity tomild DPN and consistent correlation ofeach VPT measure with sural NCV (190).

Recently, a consensus subcommitteeof the American Academy of Neurology(34) stated “QST testing for vibratory andcooling thresholds receives a Class II ratingas a diagnostic test. Further, QST is desig-nated as safe, effective and established, witha type B strength of recommendation. How-ever, QST is unacceptable as the sole criteriato define diabetic neuropathy.”

Vibration thresholds. The relationshipbetween elevated VPT and DN has beendocumented for almost 100 years. Whentested in the 50- to 300-Hz range, VPTreflects the activation of mechanorecep-tors (i.e., Pacinian and Meissner corpus-cles), conduction in large-diametermyelinated peripheral axons, and trans-mission through the dorsal column spinalpathways.

Multiple studies have documentedthe relation between loss of vibration sen-sation and the progression of a variety ofindicators of DPN (191,192). Dyck et al.(193) used computer-assisted QST toevaluate three large cohorts and identifieda “strong and consistent correlation” be-tween sensory loss and other markers ofDN. These studies confirmed that vibra-tion thresholds are especially sensitive to

mild or subclinical neuropathy. Davis etal. (194) also demonstrated that vibratorythresholds can detect subclinical neurop-athy in children and adolescents withtype 1 diabetes. Boulton et al. (195) doc-umented that vibration thresholds pro-vided a strong indication of “risk” forfuture ulceration across a wide range ofages and durations of diabetes. In a 4-yearprospective study (182), patients withbaseline threshold elevated above a fixedvalue (i.e., 25 V with the biosthesiometer)were seven times more likely to developfoot ulcers. This observation is supportedby the recent evaluation of 187 type 2diabetic patients that used multivariatelogistic regression to document that an el-evated VPT score was the strongest pre-dictor of foot ulceration (i.e., relative riskof 25.4) (183). The strength of the rela-tionship between elevated VPT and footulceration is illustrated by the finding, in1,035 type 1 and type 2 diabetic patients,that each 1-unit increase in vibrationthreshold (voltage scale) at baseline in-creased the hazard of foot ulceration by5.6% over a 1-year study period (196).

Thermal thresholds. Although mostmechanoreceptors and free nerve endingscan be stimulated by thermal energy, truecutaneous thermoreceptors are orders ofmagnitude more sensitive to shifts in tem-perature. Separate cold and warm ther-moreceptors have been identified (197)and generally characterized by small re-ceptor fields. Thermal energy is con-ducted in thinly myelinated A orunmyelinated C fibers and is principallytransmitted in the crossed anterolateraltracts of the spinal cord. The sensation ofpain can also be driven by high-intensitystimulation of thermoreceptors, espe-cially those sensitive to warming; this ac-tivation can be assessed by measuringheat-pain thresholds (198).

As is the case with vibration, alteredthermal thresholds have been well docu-mented in patients with DN defined byother criteria (179,191,193), and their el-evation has been associated with progres-sion of neuropathy and ultimately withfoot ulceration (199). Abnormal thermalthresholds have been reported in 75% ofsubjects with moderate-to-severe DPN,and elevated heat-pain thresholds weredetected in 39% of these subjects (200).Generally, there is a high correlation be-tween elevated thermal and vibrationthresholds, but these measures can be dis-sociated, suggesting a predominant

small- or large-fiber neuropathy in indi-vidual patients. The symptoms of neuro-pathic pain have been associated withaltered thermal thresholds (201), but, asstated earlier, painful neuropathy likelyinvolves both small- and large-diameterneurons (145). Lowered heat-painthresholds have been reported in patientswith DN, and this condition may be animportant indication of hypersensitivityassociated with early changes in distalnerve segments (193).

It is technically more challenging tomeasure thermal thresholds comparedwith vibration thresholds; the evaluationgenerally takes longer and the smallestdetectable difference has been reported asapproximately double that of vibration(201). Computer-assisted proceduresmay be especially valuable in examiningthermal thresholds (202).4. Electrophysiology. Whole nerveelectrophysiologic procedures (e.g.,NCV, F-waves, sensory, and/or motoramplitudes) have emerged as an impor-tant method of tracing the onset and pro-gression of DPN (203). Mult ipleconsensus panels have recommended theinclusion of electrophysiology in the eval-uation of DPN, as well as the use of theseprocedures as surrogate measures in mul-ticenter clinical trials (20,33). These pro-cedures have also been used extensivelyto explore the mechanisms of dysfunctionand the value of various therapeutic inter-ventions in chemical and genetic animalmodels of hyperglycemia.

An appropriate battery of electro-physiologic tests supports the measure-ment of the speed of both sensory andmotor conduction, the amplitude of thepropagating neural signal, the density andsynchrony of muscle fibers activated bymaximal nerve stimulation, and the integ-rity of neuromuscular transmission (rev.in 204,205). These are objective, para-metric, noninvasive, and highly reliablemeasures. However, “standard” proce-dures, such as maximal NCV, reflect onlya limited aspect of neural activity and thenonly in a small subset of large-diameterand heavily myelinated axons. Even inlarge-diameter fibers, NCV is insensitiveto many pathologic changes known to beassociated with DPN. For example, thereis strong evidence linking DPN with a re-duction in Na�/K� adenosine triphos-phatase activity (206). This deficit wouldprimarily diminish the ability of neuronsto rapidly reestablish appropriate trans-



membrane ion gradients after activation.Standard NCV, which essentially evalu-ates single pulses, could be unaffected at atime point when assessment of refractorycycles and axonal recovery would docu-ment altered function (207).

A key role for electrophysiological as-sessment is to rule out other causes ofneuropathy or to identify neuropathiessuperimposed on DPN. Unilateral condi-tions, such as entrapments, are far morecommon in diabetic patients (208).Sharma et al. (209) report that the odds ofoccurrence of CIDP were 11 times higheramong diabetic than nondiabetic pa-tients. The symmetry of electrophysiolog-ical measures, and the nature andmagnitude of the deficits, can help iden-tify additional causes for neurological def-icits and can be valuable in selectingappropriate subjects for clinical trials.

Mechanisms of NCV slowing. The prin-cipal factors that influence the speed ofNCV are 1) the integrity and degree ofmyelination of the largest diameter fibers,2) the mean cross-sectional diameter ofthe responding axons, 3) the representa-tive internodal distance in the segmentunder study, and 4) the microenviron-ment at the nodes, including the distribu-tion of ion channels.

Functional deficits. The responsivenessof the neuron and its ability to propagateneuroelectric signals are ultimately de-pendent on the distribution of transmem-brane ion channels. In myelinated axons,the nodal region is characterized by a highdensity of voltage-sensitive channels per-meable to Na�, K�, or both, as well as anonspecific channel likely responsible fornodal “leakage currents.” Factors thatmay contribute to the anchoring of spe-cific Na� channels to the nodal region in-clude suppression of Na� channels in theinternodal regions or select binding ofchannels to portions of the extracellularmatrix (rev. in 210). DPN has been re-ported to alter the nodal/paranodal distri-bution of ion channels and to widen thenodal gap. These changes may reflect“functional” deficits that could underlieacute and rapidly reversible slowing ofNCV (200,201).

Early structural deficits. Demyelina-tion can have a profound effect on NCV,but this mechanism appears to be only aminor factor in slowing of NCV in DPN(127). A recent study of 57 patients withdiabetes reported an ampl i tude-independent slowing of NCV in interme-

diate, but not distal, nerve segments,consistent with some contribution of de-myelination (211). The initial structuraldeficit responsible for slowing of NCV inDPN is likely a diminished “length con-stant” of large-diameter axons due to al-tered cross-sectional volume (i.e., earlystages of a distal axonopathy). This earlystructural pathology could reasonably bedue to a diminished production of en-doskeletal and growth-associated pro-teins (212).

Chronic deficits . The impact ofchanges in ion distributions and axonaldiameters continues to be present inchronic DPN, but in addition, NCV is al-tered by Wallerian degeneration, a conse-quent reduction in axon density, and agradual but relentless shift in the fiber di-ameter histogram toward smaller-diameter fibers.Specific electrophysiologic measuresin DPN.NCV. In the past 5 years, there have beenmore than 100 published articles discuss-ing the link between NCV and DPN, andthis builds on decades of previous re-search. A thorough discussion of this lit-erature is beyond the scope of this review;however, several key findings haveemerged:

● NCV is only gradually diminished byDPN, with estimates of a loss of �0.5 m� s�1 � year�1 (204). In a 10-year naturalhistory study of 133 patients withnewly diagnosed type 2 diabetes, NCVdeteriorated in all six nerve segmentsevaluated, but the largest deficit was 3.9m/s for the sural nerve (i.e., 48.3 to 44.4m/s); peroneal motor NCV was de-creased by 3.0 m/s over the same period(8). A similar slow progression ofchange in NCV was detected in the Di-abetes Control and Complications Trial(DCCT) (7), in which the sural and per-oneal nerve velocities in the conven-tionally treated group diminished by2.8 and 2.7 m/s, respectively, over the5-year study period.

● NCV provides a sensitive but nonspe-cific index on the onset of DPN and canbe valuable in detecting subclinical def-icits. The earliest reports of altered NCVin patients without clinical symptomsor signs of DPN date back for more than40 years and have been confirmed inrecent studies (204,205).

● NCV can trace the progression of DPNand can provide a valuable measure of

the severity of DPN and “quality of liferelated to peripheral nerve involve-ment” (213).

● Changes in NCV are related to glycemiccontrol (214). In the DCCT, subjectswho were “free of confirmed neuropa-thy at baseline” had a 40.2% incidenceof abnormal NCV in the conventionallytreated group and only 16.5% in thegroup receiving intensive therapy aftera period of 5 years (7). This was associ-ated with a between-group difference of4.0 m/s for the peroneal nerve and 3.9m/s for the sural nerve. A previousstudy in 45 type 1 diabetic patients uti-lized a regression analysis to documentthat a 1% change in HbA1 was associ-ated with a 1.3 m/s change in maximalnerve conduction (215).

● Changes in NCV can reflect underlyingstructural pathology in large-diameteraxons, including atrophy, demyelina-tion, and loss of fiber density (205).

● NCV can improve with effective ther-apy (51) or with transplantation (216).

Amplitudes, area, and duration. Peakamplitude of either the SNAP or theCMAP driven by maximal stimulation re-flects the number of responding fibersand the synchrony of their activity. Thereis a strong correlation (r � 0.74; P �0.001) between myelinated fiber densityand whole-nerve sural amplitude (217) inDPN. Russell et al. (218) calculated that achange of 1.0 V in sural nerve SNAPamplitude is associated with a decrease of�150 fibers/mm2, while a loss of 200 fi-bers/mm2 is associated with an approxi-mate 1.0-mV reduction in the meanamplitude of the CMAP from the ulnar,peroneal, and tibial nerves. Longitudinalstudies suggest an average loss of SNAPamplitude at a rate of �5% per year inDPN over a 10-year period (8).

Measuring the total area of the SNAPand CMAP has been suggested as a meansof assessing the contribution of slowerconducting fibers, but these measures areseverely limited by variability. Area alone,or in association with peak amplitude,can also be used to estimate the degree oftemporal dispersion and conductionblock.

F-waves. F-waves reflect the anti-dromic conduction of the compoundneural volley to the ventral spinal cord,the activation of a subpopulation of spinalmotor neurons, the orthodromic conduc-tion of the newly established volley, and



the postsynaptic activation of a portion ofthe muscle fibers in the innervated mus-cle. Because of its “long-loop” nature, thismeasure is sensitive to factors that alterthe speed of conduction, especially thosewidely distributed along the nerve. A sub-tle change affecting each node may not bedetected in measures focused on an iso-lated distal segment, but may accumulateand become evident in the long latencyF-wave response.

F-wave procedures have been re-ported as a sensitive and reliable tool inpatients with axonal polyneuropathy(219). However, changes limited to thedistal segment of the axon, including pos-sible therapeutic benefits, may be poorlyrepresented in F-wave measures. Minimallatency is the most frequent measure ofF-wave activity. However, the addition ofchronodispersion, duration, persistence,and amplitude can add sensitivity toslower conducting axons (219).

Distribution of velocities. Several pro-cedures have been developed to analyzethe distribution of conduction velocitiesas a means of measuring activity in small-diameter axons (220). The original stud-ies of an al tered dis tr ibut ion ofconduction velocities in DPN are approx-imately two decades old, but the tech-nique is rarely used in current clinicalstudies. The fusion of a collision tech-nique and an analysis of the distributionof velocities is promising as a practicaland valuable electrophysiologic proce-dure to explore the effects of DPN onslower fibers. Caccia et al. (221) demon-strated that a computer-assisted collisionprocedure was both capable of examiningvelocities in slower conducting fibers andsensitive to the presence of subclinicalneuropathy in insulin-dependent diabeticsubjects. Bertora et al. (222) examined138 patients with subclinical DPN and re-ported sensory nerve deficits were de-tected in 58% of the subjects using thedistribution of conduction velocities datacompared with only 11% of subjects us-ing standard procedures.

Excitability. In addition to measuringthe speed of conduction or the size of theactivated signal, the magnitude and na-ture of the current necessary to establishthe electrophysiologic response can be animportant parameter in assessing neurop-athy (rev. in 223). For instance, one com-puter-assisted measure of excitability,termed “threshold electrotonus,” exam-ines the effects of prolonged hyperpolar-

izing and depolarizing subthresholdcurrents (224) to explore differences inexcitability between sensory and motoraxons and fluctuation of excitability dur-ing the refractory period. Excitation stud-ies have indicated that the diabetic nervehas less accommodation to hyperpolar-ization (i.e., inward rectification), whichmay limit its ability to follow rapid stim-ulus trains (225). At the cellular level,changes in excitability may be related toalterations in intracellular levels of cyclicadenosine monophosphate (cAMP) thathave been reported with DPN (225).5. Other methods of assessment. Themajority the methods included in this sec-tion are relatively invasive, requiring bi-opsy of a whole nerve or fascicle or a skinbiopsy to assess small-fiber structure.More recent noninvasive techniques in-clude magnetic resonance imaging (MRI)and corneal confocal microscopy.

Nerve biopsy. The nerve biopsy, typi-cally of the sural nerve posterior to thelateral malleolus, has been used for manyyears in the study of peripheral neuropa-thy (226–228). When undertaken at acenter with sufficient expertise, it is a use-ful diagnostic procedure in patients withneuropathy of a known origin or in dia-betic patients with atypical neuropathies(228). However, this is an invasive proce-dure with recognized sequelae that mightinclude persistent pain at the biopsy site,cold intolerance, unpleasant though mildmechanically elicited sensory symptoms,and sensory deficits in the sural distribu-tion (229,230). These prolonged sensorysymptoms and sensory loss appear to oc-cur more commonly in diabetic than innondiabetic subjects (229). Thus, withthe widespread availability of accurateQST and electrophsiological techniques,biopsies are rarely required for the routinediagnosis of DPN. For clinical diagnosticpurposes, a fascicular or subtotal biopsyshould suffice; if the nerve is left in con-tinuity, a greater possibility of regenera-tion across the gap exists (228).

The use of morphological measures ofneuropathy from biopsies as end points intrials of potential pharmacological thera-pies for DPN is a more controversial area.Whereas several trials of ARIs have usedmyelinated fiber density and other mor-phological measures as end points(51,53), concerns have been expressedabout the use of such measures as endpoints (228). These concerns not only in-clude the invasive nature of having two

nerve biopsies on separate occasions, butalso relate to a number of uncertainties inrelation to the interpretation of these find-ings (228,231,232). The PeripheralNerve Society (PNS) consensus report onDPN in controlled clinical trials suggestedthat the use of biopsy findings in assessingresponse to therapy needs further valida-tion (33). In addition to the above list ofconcerns, the PNS also suggested thatthere is insufficient information as to howwell neuropathological measures predictthe severity and course of neuropathy andquestions the validity of such assessmentsas axonal atrophy and axo-glial dysjunc-tion, which require electron microscopy.The PNS did report that, of all the patho-logical measures, the myelinated fiberdensity is probably most useful as it cor-relates with the clinical deficit and elec-trophysiological findings (33,217).

In addition to assessing responses totherapy, nerve biopsies have also beenused to help determine the etiopathogen-esis of neuropathy. Examples of this in-clude studies of diabetic amyotrophy(125) and the importance of glycemiccontrol in DPN (233).

Nerve exposure. A number of pub-lished studies investigating the pathogen-esis of neuropathy have studied the suralnerve in vivo without actually biopsyingit. These have included using microelec-trodes to measure endoneurial oxygentension (234) and the use of epineurialvessel photography and fluorescein an-giography to study the neural microvas-culature (235). More recently, the samegroup used a new minimally invasivetechnique of microlight-guide spectro-photomety to measure blood flow and ox-ygen saturation in the sural nerve (236).However, these techniques are only usedin specialist research units investigatingthe etiopathogenesis of DPN.

Skin biopsy. The significance andusefulness of immunohistochemicallyquantitated cutaneous nerves in themorphological assessment of DPNis increasingly being recognized (237,238). It was the discovery of the panax-onal marker, protein gene product 9.5,that allowed the direct visualization ofepidermal nerve fibers. This technique,though still invasive, only requires a3-mm skin biopsy and enables a directstudy of small nerve fibers, which aredifficult to assess electrophysiologi-cally (238). Much of this work has beendeveloped by researchers who use the



technique to study three main groups ofpatients: DN, HIV-associated neuropa-thy, and idiopathic small-fiber sensoryneuropathy (238) . Recent ly thismethod was used to assess early neuro-pathic changes in diabetes and IGT(38). The assessment of cutaneousnerve pathology, including nerve regen-eration, has also been recently advo-cated by other groups (239).

Noninvasive assessment. MRI has beenused to assess involvement of the spinalcord in neuropathy. In an exploratorystudy, Eaton et al. (240) used MRI of thecord and demonstrated that patients withDPN had a lower cross-sectional cord areathan healthy control subjects in the cervi-cal and thoracic regions, leading them tosuggest that DPN is not simply a disease ofthe peripheral nerves.

More recently, Malik et al. (115) re-ported the technique of corneal confocalmicroscopy in the assessment of DPN.This is a completely noninvasive tech-nique that offers the future potential ofassessing nerve structure “in vivo” with-out the need for biopsy. This technique isable to accurately define the extent of cor-neal nerve damage and repair, which seemsto correlate with peripheral nerve function.This may provide the opportunity in futurestudies to act as a structural surrogate mea-sure of nerve function in diabetes.6. Epidemiology and natural history ofDPN.Background. This section will specificallyexamine data that pertain to manifesta-tions of DPN. Some liberty has been takento use the DPN “label” since this term is notused in some of the referenced studies.

A number of studies have producedextensive data regarding epidemiologicaspects of DPN. Despite this, there are stillareas in which knowledge is deficient.This seeming paradox is at least partiallyexplained by the pathological complexityof DPN. The number of peripheral nervesthat can be affected, their differing com-positions of sensory and motor fibers, andthe varying extent of pathology of thenerve fibers account for this complexity.Thus, although there are similarities inclinical presentations, the manifestationsof DPN can be quite heterogeneous.

This heterogeneity has led to multiplekinds of assessments and multiple endpoints in studies of DPN, including symp-toms, signs, QST, and electrophysiology.Because epidemiologic studies have useda variety of these assessments, singly or in

combination, it is difficult to evaluatethese studies for consistency of findingsand to draw firm conclusions. This is evenfurther complicated by differences incharacteristics of the study populations.

Given the above considerations, thisreview will examine epidemiologic stud-ies of DPN according to the specific as-sessments and end points that have beenused. This discussion will at times drawfrom findings from clinical trials, sincethese studies can contribute useful epide-miologic information pertaining to DPN.

Positive sensory symptoms and painfulneuropathy. The importance of distin-guishing positive sensory symptoms fromnegative sensory symptoms has been em-phasized (32). Positive sensory symptomsarise spontaneously or as a response tostimuli. In contrast, negative sensorysymptoms represent decreased respon-siveness to stimuli. There is an abundanceof types of positive sensory symptoms,and it has been suggested that they shouldbe divided into painful and nonpainfulcategories (32). This classification issomewhat arbitrary, since there is littleevidence that it relates specifically to neu-ropathology; however, for certain pur-poses, such a classification may be useful.

Table 5 presents a listing of positivesensory symptoms compiled by a com-mittee that examined end points for pain-ful neuropathy (32). Although painfuland nonpainful symptoms were sepa-rated, “prickling” and “tingling” appearedin both categories. This overlap underliesthe difficulty in developing symptom cri-teria for painful neuropathy.

The discussion below will mostly fo-cus on painful neuropathy. However, itshould be emphasized that studies ofpainful neuropathy have often requiredevidence of DPN from other neurologicassessments for the inclusion of individu-als. Although this strategy helps to con-fine studies only to individuals who trulyhave painful neuropathy, it carries the im-plicit assumption that pain in itself doesnot occur as a sole manifestation of DPN.Also, as discussed above, it is difficult todetermine the specific symptoms thatshould constitute painful neuropathy. Itis clear that care must be taken in inter-preting studies of painful neuropathy.

Prevalence. There have been few epi-demiologic studies that have specificallyexamined the prevalence of painful neu-ropathy. The large population-basedstudy of Harris et al. (148) utilized ques-tionnaires of both diabetic and nondia-betic individuals to ascertain informationabout sensory symptoms. Symptomswere categorized according to pain andtingling, numbness, and the inability tofeel hot or cold. Of interest was an appre-ciable prevalence of painful symptoms inthe nondiabetic individuals. The basis fortheir symptoms is unclear, but the differ-ences in the prevalence estimates betweenthose with and those without diabetes canbe used as an estimate of the prevalence ofsymptoms due to diabetes. The overallprevalence estimate for painful symptomsin the diabetic individuals was 27%, butthe difference in the prevalence rates be-tween those with and without diabeteswas smaller and tended to decrease withage.

Other studies (8,241,242) have ex-amined the prevalence of painful symp-toms in clinical settings, and estimateshave varied from 3 to �20%. This varia-tion probably is a function of the differingcriteria used for painful neuropathy andthe characteristics of those studied.

Because some studies suggest thatpainful neuropathy can remit (see below),its prevalence could be much lower thanits cumulative incidence over the fullcourse of diabetes. Unfortunately, thereare no such cumulative incidence dataavailable.

Natural history. There is limited infor-mation regarding the natural history ofpainful neuropathy. A decrease in the in-tensity of painful neuropathy with wors-ening of quantitative measures of sensoryfunction has been observed (243). The

Table 5—Descriptions of positive neuro-pathic sensory symptoms

Nonpainful Painful

Thick PricklingStiff TinglingAsleep Knife-likePrickling Electric shock-likeTingling Squeezing

ConstrictingHurtingBurningFreezingThrobbingAllodyniaHyperalgesia

*Allodynia: the perception of pain from a nonnox-ious stimulus. Adapted from ref. 32.



findings of this study are consistent withthe hypothesis that pain can decreasewith the pathological progression of DPN.It should be noted that a number of thestudy participants were on treatment forpainful symptoms, and this could haveaffected the findings. The observed im-provement of pain with decreasing sen-sory function in that study appears tocontrast with another study (147) thatfound a coupling of improvement of painwith improvement of sensory function inindividuals treated with continuous sub-cutaneous insulin infusions.

There have been inconsistent data re-garding how commonly painful neuropa-thy remits. Studies in this area mustcarefully differentiate those with chronicpainful symptoms from those who de-velop painful symptoms more acutelyearly in the course of diabetes. These pat-terns may represent different pathologicalprocesses. In a study of 36 diabetic pa-tients with chronic, painful symptomswho were followed for an average periodof 4.7 years, there was no overall changein the severity of pain scores over time,and there were no full remissions in any ofthose followed (244). However, otherstudies have observed an appreciable oc-currence of remissions (243,245).

Risk factors. Despite the large numberof studies that have examined risk factorsfor DPN, there are few that have ad-dressed painful neuropathy per se. Inanalyses of risk factors in their study, Har-ris et al. (148) combined positive and neg-ative sensory symptoms together as endpoints. They found that the combinedsensory symptoms were related to yearssince diagnosis, reported degree of hyper-glycemia and glycosuria, and hyperten-sion. In a study of a clinic population,painful neuropathy was found to be asso-ciated with diabetes duration but not withHbA1c levels (8). In an uncontrolled in-tervention study, pain levels improved inindividuals treated with continuous sub-cutaneous insulin infusions (147).Negative sensory symptoms and hy-poesthetic neuropathy. Much more in-format ion has been accrued forhypoesthetic neuropathy than for positivesensory symptoms and painful neuropa-thy, possibly as a result of the ability tomore objectively quantify sensory func-tion through QST. Because it is well rec-ognized clinically that patients tend tounderestimate their degree of insensitiv-ity, there has been more reliance on QST

than on symptomatology in studies of hy-poesthesia. Thus, the discussion belowwill mostly pertain to QST assessmentsrather than those based on symptomatol-ogy. Although studies have examined dif-ferent sensory modalities with a numberof quantitative sensory methodologies,much of the available data pertain toVPTs. The sensory modalities and themethodologies utilized for their assess-ments should be considered in interpret-ing the data presented below, since nervefibers may be differentially affected, andas indicated above, techniques can varyconsiderably. Also, it should be empha-sized that quantitative sensory measure-ments are not fully objective because theyare dependent on the understanding andcooperation of the individuals studied.

Prevalence. The prevalence of hypoes-thesia that has been reported in studiesvaries greatly according to the above con-siderations and to the criteria used to de-fine abnormality. In a study that utilizedthree quantitative sensory measurementsin the same individuals, the prevalence ofabnormalities varied from 8 to 34%(246). The location of measurement isalso an important factor. The presence ofhypoesthesia increases substantially asmeasurements become more distal (247).Also, the prevalence can vary greatly ac-cording to the characteristics of the pop-ulation under study. For example, agerequires careful consideration, since nor-mal values increase markedly with age(248).

Relatively few studies using QST haveprovided specific estimates of the preva-lence of decreased sensory function alone.However, perhaps because of the aboveconsiderations, there appears to be arather wide variation (242,246,247,249).In the Wisconsin Epidemiologic Study ofDiabetic Retinopathy (WESDR), a pro-spective cohort study, the 10-year inci-dence of the symptoms of loss of tactilesensation and loss of temperature sensi-tivity varied from 19 to 25% and from 11to 19%, respectively, depending on theage of onset and the use of insulin (250).

Natural history. Quantitative sensorymeasurements lend themselves to studiesof natural history. In a study of childrenand adolescents, there was a small but sta-tistically significant elevation of VPTs(251). A recent study of adults revealedthat the presence of decreased sensationat diagnosis appears to vary according todiagnostic criteria for diabetes (252).

Those who fulfilled the �126 mg/dl fast-ing glucose criterion, on average, had el-evated vibration and thermal thresholdsat screening for diabetes. In contrast,those who fulfilled the �200 mg/dl 2-horal glucose tolerance test criterion, butnot the �126 mg/dl fasting glucose crite-rion, had normal thresholds.

In a study of type 1 diabetic patientswho were followed from diagnosis, ther-mal thresholds tended to increase morethan vibration thresholds in the first 5years after diagnosis. The increase in ther-mal thresholds was particularly evident inthose with higher glucose levels. Nerveconduction was also found to be affectedin that interval (253).

The progression of hypoesthesia hasbeen examined early in the course of type2 diabetes. In one study, individuals werefound to have small but statistically sig-nificant increases of vibration and ther-mal thresholds over an average follow-upinterval of �2 years (254). Findings froma study that utilized 10-g monofilamenttesting suggested that there is little pro-gression of neuropathy in the first fewyears of diabetes (255). However, the10-g monofilament may not be suffi-ciently sensitive to detect a small changein loss of sensation.

Progression appears to be more rapidonce decreased sensation appears. In astudy that followed a control group par-ticipating in a clinical neuropathy trial for18 months, there was marked worseningof vibration and thermal thresholds(256). A similar rate of progression wasobserved over a 2-year period in anotherstudy of individuals with an appreciabledegree of hypoesthesia at baseline (257).In this study, the changes of the varioussensory modalities paralleled each other.In another study that followed individualsinitially considered to have normal vibra-tion thresholds for an average interval of12 years, there was marked progression inthose who eventually developed de-creased sensation (181).

Risk factors. Diminished sensory func-tion was consistently related to diabetesdurat ion in a number of studies(242,247,249,255). Decreases in sensorythresholds have been found to be relatedto the degree of hyperglycemia (181,242,249,253) and height (181,242,247,249,255) in both cross-sectional and fol-low-up studies. In the WESDR, the devel-opment of symptoms of the loss of tactileand temperature sensation tended to be



related to HbA1c (250). Studies have re-vealed variable findings for the relationsof hypoesthesia with alcohol consump-tion (247,249,255) and cigarette smok-ing (249,255).Combined assessments. Included inthis section are studies that have utilizedvarious combinations of positive symp-toms, negative symptoms, QST, abnor-malities of the neurological exam, andelectrophysiological testing for end pointsof neuropathy. Although substantial in-formation has been obtained from largeepidemiologic studies and clinical trialsthat have utilized such combined assess-ments as criteria for neuropathy, thesestudies can be difficult to interpret inpathophysiological terms. For example, itis problematic to discern the specificfunctions that are actually related to therisk factors identified with this approach.The interpretation of findings is also com-plicated by differences between studies inthe choice of the combined assessments.

Prevalence and incidence. In a large ep-idemiologic study that utilized combinedend points, Pirart (258) followed a num-ber of patients for the development ofneuropathy over many years. The criteriawere quite broad and some of those con-sidered neuropathic could have had otherforms of DN instead of DPN. The findingsrevealed that the prevalence of neuropa-thy was �40% after 25 years of knownduration.

The EURODIAB IDDM Complica-tions Study (9) has examined a large num-ber of participants from clinical centersover a broad geographical area. The over-all prevalence of neuropathy was 28%;however, among the 27 centers includedin the study, the prevalence ranged from�20% in several centers to �50% in twocenters. In the Epidemiology of DiabetesComplications (EDC) Study (259), a pro-spective study of patients with type 1 di-abetes, there was an overall prevalence ofDPN at baseline of 37% in those �18years of age with substantial variation ac-cording to age (18% for those 18 –29years and 58% for those who were older).After an average length of follow-up of 5.3years for that cohort, the incidence ratefor DPN was 2.8 per 100 person-yearswith a cumulative probability of 29%(260). In the San Luis Valley DiabetesStudy (SLVDS) (261), a population-basedstudy of type 2 diabetic patients, therewas an overall prevalence of 28%. In ananalysis of baseline data from the DCCT

(262), “clinically detectable” neuropathywas found in 39% of the participants.

Natural history. There is a lack of dataon the progression and regression of DPNin studies of combined end points. Thisinformation is difficult to ascertain sincesuch end points are not optimal for quan-titative assessments of severity. However,these types of studies could still be usefulfor identifying individuals with a substan-tial change in neuropathic status.

Risk factors. The Pirart study (258) re-vealed strong associations of neuropathywith duration of diabetes and with thedegree of hyperglycemia (according tocertain clinical indicators); however, onlyunivariate associations were reported andno other risk factors were discussed. TheEURODIAB IDDM study (9) identified as-sociations of neuropathy with a numberof factors. Although the degree of associ-ation was to some extent dependent onthe multivariate modeling, neuropathywas consistently related to age, duration,HbA1c, and severe ketoacidosis. Factorsfor which statistical significance was modeldependent were weight, height, and currentcigarette smoking. The prevalence of neu-ropathy was related to elevated diastolicblood pressure, triglyceride, and decreasedHDL cholesterol.

The DCCT utilized combined endpoints in its baseline prevalence estimateand in the trial itself (7). However, in itsbaseline report, the DCCT specifically ex-amined associations of nerve conductionindexes with characteristics of partici-pants. Nerve conduction impairment wasfound to have some association with age,diabetes duration, HbA1c, male sex, andC-peptide deficiency. The DCCT defini-tively showed a decrease in the rate of de-velopment of DPN with intensivetreatment of hyperglycemia.

In the EDC Study (259), DPN at base-line was observed to be related to a num-ber of variables with varying degrees ofassociation according to age. In the over-all analysis, DPN was associated with di-abetes duration, HbA1, HDL cholesterol,hypertension, and cigarette smoking.Prospective data revealed associations ofDPN with duration, height, HbA1, ciga-rette smoking, and hypertension.

The SLVDS (261) found that DPNwas related to age, diabetes duration,HbA1c, and insulin use. Several studiesobserved associations of DPN with othercomplications of diabetes (9,259,261).

In a case-control study, DPN was ob-

served to be associated with lifetime ciga-rette smoking in individuals with type 1diabetes, but not in those with type 2 di-abetes (263).

Conclusion. This section has discussedthe epidemiology of DPN according to theend points that were utilized in studiesrather than according to risk factors. Al-though it might appear that this is a te-dious approach, the end points should beconsidered separately in order to makeetiologic and pathogenetic sense of the lit-erature. Also, this approach can help toexplain the seeming inconsistencies offindings in studies.

Despite the many different endpoints, study populations, and methodol-ogies described, some definitive conclu-sions can still be made. First, it is clearthat DPN is a very common condition.Although the prevalence estimates of itsvarious manifestations are clearly studydependent, it appears that at least onemanifestation of peripheral neuropathy ispresent in well over 20% of individualswith diabetes. However, the prevalence ofpainful neuropathy appears to be appre-ciably lower.

Second, regarding natural history,studies suggest that some evidence of ab-normality can occur relatively early in thecourse of diabetes. Studies also show thatonce DPN is present, there is a tendencytoward rapid pathological progression.The frequency of the remission of painfulneuropathy and other manifestations ofneuropathy is still unclear.

Finally, DPN is clearly associatedwith certain risk factors. The degree ofhyperglycemia has been identified as arisk factor in both epidemiologic studiesand clinical trials. Diabetes duration hasalso been a consistent risk factor. Heighthas been observed to be a risk factor in anumber of studies, but it may be depen-dent on location and the sensory modalityassessed. Certain conventional cardiovas-cular disease risk factors, including lipidand blood pressure indexes, have beenidentified as risk factors for DPN. Severalstudies have observed associations ofDPN with other complications of diabe-tes. Other risk factors such as alcohol con-sumption and cigarette smoking havebeen less consistent in their associationswith DPN.

The definitive risk factors that havebeen identified have biological plausibil-ity for involvement in the pathogenesis ofDPN. Duration and degree of hyperglyce-



mia can be viewed as being indicative ofthe extent of overall exposure to hyper-glycemia. Height, as a proxy for nervelength, appears to be an across-individualexpression of the intraindividual depen-dence of nerve length for the occurrenceof DPN. A hypothesis can be entertainedthat longer nerves are more susceptible tothe metabolic consequence of diabetes. In-deed, one study has shown an association ofvibration perception with an interaction ofheight and degree of hyperglycemia (254).Blood pressure and lipid indexes could in-dicate that vascular abnormalities contrib-ute to the development of DPN. Finally, therelation between DPN and other complica-tions could mean that they have commonpathogenetic pathways.

It is clear that epidemiologic findingsalready provide some clues for mechanis-tic research; yet, there is a potential for theaccrual of much more epidemiologic in-formation with regard to occurrence, eti-ology, and natural history. This will beespecially enhanced with attention to theend points and methodology utilized,both with regard to designing new studiesand the interpretation of studies alreadyperformed.7. Pathogenetic treatments and preven-tion. This section will discuss thosetreatments that may prevent the onset ormodify the natural history of DPN by tar-geting known pathogenetic mechanisms.In general, treatments in this category donot treat symptoms and are mostly exper-imental and not therefore available forclinical usage at present. The discussionof most of these approaches will be brief,as mention is also made in SECTION 3:PATHOGENESIS OF DIABETIC NEUROPATHY.

Near normoglycemia. In addition tothe DCCT (7), three much smaller butlong-term prospective studies have con-firmed that maintained near-normal glyce-mia prevents the development and retards

the progression of DPN as assessed electro-physiologically. These include the Stock-holm Diabetes Intervention Study (7.5[264] and 10 [265] years), the Oslo Study(8 years) (266), and, in type 2 diabetes, theKumamato Study (6 years) (267).

The most reliable method of achievingand maintaining near-normal glycemia isby pancreatic or islet cell transplantation.However, as in most published series thatassess nerve function, the transplant was incombination with renal transplants, and therecipients generally had long duration of di-abetes and established neuropathy. Thus, inthe series from Minneapolis, Minnesota(40,268), only modest improvements inmeasures of neuropathy were seen after sev-eral years of normoglycemia. Nevertheless,achieving near-normoglycemia should bethe aim in both the prevention of and thefirst step of managing DPN.

ARIs. The first clinical trials of ARIs inDN took place 25 years ago, and currentlyonly one agent is available in one country(Epalrestat in Japan) (269). Most of theearly trials can be summarized as:

● Too small. The effect of the drug wasinadequate in terms of inhibiting nervesorbitol accumulation.

● Too few. Inadequate numbers of sub-jects were included.

● Too short. Many trials were for onlyweeks or months for a chronic diseaseof many years’ duration.

● Too late. No drug targeting a pathoge-netic mechanism is likely to be effec-tive when the complication is wellestablished.

A summary of some of the drugs thathave been studied in clinical trials arelisted in Table 6: further details are pro-vided in Section 3B.

Antioxidants. As discussed in section3, there is accumulating evidence to sup-

port the role of oxidative stress in thepathogenesis of neuropathy. Studies withthe antioxidant �-LA have provided evi-dence of potential efficacy for this agent,which may well be beneficial for bothneuropathic symptoms and modifyingthe natural history of DPN (62–64). Twolarge North American/European clinicaltrials of the efficacy of �-LA are inprogress and should report in 2005.

�-LA. �-LA (GLA) is a component ofevening primrose oil and can prevent ab-normalities present in diabetes and in essen-tial fatty acid and prostanoid metabolism(273). GLA treatment for 1 year in a ran-domized trial resulted in improvement inelectrophysiology and deficits (276).

The use of a conjugate of LA and GLAwas proved to be effective in improvingboth electrophysiological and neuro-chemical correlates of experimental DN(275). To date, this conjugate has notbeen sited in clinical neuropathy.

Neutrophins. As a result of contradic-tory results from clinical trials, the clinicaldevelopment of NGF was halted, and nofurther studies are planned at the time ofwriting (276).

Inhibitors of glycation. Studies of ami-noguanidine, which inhibits the forma-tion of AGEs, have mainly focused onnephropathy (54). Few data are availableon aminoguanidine or other inhibitors ofAGE formation in clinical neuropathy(277).

PKC inhibition. Intracellular hypergly-cemia increases DAG levels, which acti-vates PKC formation, leading to multiplepathogenetic consequences including al-tered expression of endothelial nitric ox-ide synthetase and VEGF. Preliminarydata suggest that treatment with a PKC-�inhibitor might ameliorate measures ofnerve function in DPN (278). Multicentertrials are currently in progress and shouldreport in 2004 or 2005.

Vasodilators. Treatment with ACE in-hibitors has been shown to improve elec-tophysiological measures of nervefunction in mild neuropathy (68). Theshort-acting vasodilator isosorbide dini-trate has been shown to improve painfulsymptoms, but its effect on deficits andelectrophysiology are unknown (279).8. Symptomatic management of DPN.This section will discuss the managementof neuropathic symptoms. Most of thepharmacological and other interventionsthat will be described have no effect onthe natural history of neuropathy, which

Table 6—Trials of ARIs

Drug Results Status Ref.

Alrestatin Minor benefits (Sy) Withdrawn (toxicity) 270Sorbinil Benefits (Sy, Ep, M) Withdrawn (toxicity) 43Tolrestat Minor benefits (Sy, Ep) Withdrawn (toxicity) 271Ponalrestat No efficacy Withdrawn 272Zenarestat Minor benefits (Ep, M) Withdrawn (toxicity) 53Epalrestat Minor benefits (Sy, Ep) Marketed (Japan) 269Fidarestat Minor benefits (Sy, Ep) Under investigation 54

Ep, electrophysiology; M, motor; Sy, symptoms.



is of progressive loss of nerve function.The initial management of patients withsymptomatic neuropathy is summarizedin Table 7.

Control of hyperglycemia. A number ofsmall open-label uncontrolled studieshave suggested that achieving stable near-normoglycemic control is helpful in themanagement of painful neuropathicsymptoms. In one such study (147), pa-tients with painful neuropathy weretreated with continuous subcutaneous in-sulin infusion for a period of 4 months. Aswell as resulting in relief of neuropathicsymptoms, improvements were noted inQSTs and electrophysiological investiga-tions. Improvement of glycemic controlwas assessed by glycated hemoglobin aswell as regular home blood glucose mon-itoring. The fact that blood glucose fluxwas reduced in this early study might ex-plain the symptomatic benefit of thistreatment in light of more recent observa-tions (10). In this later study, when pa-tients with painful neuropathy werecompared with those with painless neu-ropathy, those with painful symptomshad poorer control, more excursions tohyper- and hypoglycemic levels, andgreater blood glucose flux as assessed by anumber of measures. Thus, it may be thestability of glycemic control that is equallyimportant to the level of achieved control.Despite the lack of appropriately de-signed controlled studies in this area, it isgenerally accepted that intensive diabetestherapy aimed at near normoglycemiashould be the first step in the treatment ofany form of DN.

Pharmacotherapy. A large number oftherapeutic agents have been used in themanagement of painful symptoms; someof the more commonly used ones arelisted in Table 8. Although some have ad-vocated the use of nonsteroidal anti-inflammatory drugs in symptomaticneuropathy, there is little evidence to sup-port their use. Moreover, these agentsshould be used with caution in neuro-pathic diabetic patients, many of whommay have renal impairment, a contraindi-cation to nonsteroidal drugs in mostcases.

Tricyclic drugs. Several randomizedclinical trials have supported the use ofthese agents in the management of neuro-pathic pains. Putative mechanisms bywhich these drugs relieve pain include in-hibition of norepinephrine and/or seroto-nin reuptake at synapses of centraldescending pain control systems and,more recently, the antagonism of N-methyl-D-aspartate receptors, which me-diate hyperalgesia and allodynia (280).The rapid onset of pain relief with theseagents, together with the fact that theyseem to be equally effective in relievingpain in patients with normal and de-pressed moods, suggests a mode of actionthat is not primarily relief of depression.Although these agents remain the first-line treatment for symptomatic neuropa-thy in most centers, their use is restrictedbecause of the frequency and severity ofside effects.

Most experience has been achievedwith amitriptyline and imipramine. Thedosage of either one of these two drugsrequired for symptomatic relief is similar(25–150 mg daily); to avoid unduedrowsiness, the dose can be taken once a

day in the evening. Desipramine is also auseful drug that may be better toleratedthan amitriptyline in many patients(280). The usefulness of these agents wasconfirmed in a systematic review per-formed by McQuay et al. (281). As notedabove, the major problem remains the fre-quency of side effects, which are predict-able. Although drowsiness and lethargyare common, the anticholinergic side ef-fects, particularly dry mouth, are the mosttroublesome.

In cases of very severe painful neu-ropathy that are partially resistant to tri-cyclic drugs, a combination of thetricyclics with other agents, such as majortranquilizers, may be useful (282). Morerecently, the combination of amitriptylineand transcutaneous electrotherapy hasbeen described in those who failed on tri-cyclic monotherapy. In a controlled trial,this combination was superior to that oftricyclic monotherapy plus sham electro-therapy (283).

Selective serotonin-reuptake inhibitors.Selective serotonin-reuptake inhibitors(SSRIs) inhibit presynaptic reuptake ofserotonin but not norepinephrine. Stud-ies suggest that treatment with paroxetine(284) but not fluoxetine (280) is associ-ated with significant pain relief. Similarly,citalopram 40 mg/day was confirmed tobe efficacious in relieving neuropathicpain, but was less effective than imipra-mine (285). These drugs should, how-ever, be used with caution in diabeticpatients who may be on other medica-tions, as there is a suggestion that SSRIsmight increase the risk of upper-gastrointestinal bleeding (286). However,troublesome side effects are in generallyless common with SSRIs.

Table 7—Initial management of symptom-atic neuropathy

1) Exclude nondiabetic causes● Malignant disease (e.g., bronchogenic

carcinoma)● Metabolic● Toxic (e.g., alcohol)● Infective (e.g., HIV infection)● Iatrogenic (e.g., isoniazid, vinca

alkaloids)● Medication related (chemotherapy, HIV

treatment)2) Explanation, support, and practical

measures (e.g., bed cradle to lift bed,clothes off hyperesthetic skin)

3) Assess level of blood glucose controlprofiles

4) Aim for optimal stable control5) Consider pharmacological therapy

Table 8—Oral symptomatic therapy of painful neuropathy

Drug class Drug Daily dose (mg) Side effects Ref.

Tricyclics Amitriptyline 25–150 �� 280,281Imipramine 25–150 �� 280,281

SSRIs Paroxitene 40 �� 284Citalopram 40 �� 285

Anticonvulsants Gabapentin 900–1,800 �� 290,291Lamotrigine 200–400 �� 292Carbamazepine Up to 800 �� 289

Antiarrhythmics* Mexilitene Up to 450 �� 293,294Opioids Tramadol 50–400 �� 295,296

Oxycodone CR† 10–60 �� 297,298

All medications in the table have demonstrated efficacy in randomized controlled studies. *Mexiliteneshould be used with caution and with regular EKG monitoring; †oxycodone CR may be useful as an add-ontherapy in severe symptomatic neuropathy.



Anticonvulsants. Anticonvulsants havebeen used in the management of neuro-pathic pain for many years (287,288).Limited evidence exists for the efficacy ofphenytoin and carbamazepine for DN(287,289). Gabapentin is now widelyused for neuropathic symptoms (thisagent is structurally related to the neuro-transmitter �-aminobutyric acid [GABA])and was introduced some years ago as ananticonvulsant for complex partial sei-zures. In a large controlled trial of gabap-entin in symptomatic neuropathy,significant pain relief together with re-duced sleep disturbance was reported us-ing dosages of 900–3,600 mg daily (290).In a recent review of all the trials of gaba-pentin for neuropathic pain, it was con-cluded that dosages of 1,800–3,600 mgper day of this agent were effective; theside-effect profile also seems superior tothat of the tricyclic drugs (291).

Lamotrigine is an antiepileptic agentwith at least two antinociceptive proper-ties. In a randomized placebo controlledstudy, Eisenberg et al. (292) confirmedthe efficacy of this agent in patients withneuropathic pain.

Antiarrhythmics. Mexilitine is a class1B antiarrhythmic agent and a structuralanalog of lignocaine. Its efficacy in neuro-pathic pain has been confirmed in con-trolled trials and reviewed by Dejgard etal. (293) and Jarvis and Coukell (294).The dosage used in trials (up to 450 mgdaily) is lower than that usually used forthe treatment of cardiac arrhythmias;however, regular electrocardiogram(ECG) monitoring is necessary, and thelong-term use of mexilitine cannot berecommended.

Other agents. Tramadol is an opioid-like, centrally acting, synthetic nonnar-cotic analgesic. Its efficacy in themanagement of patients with painful neu-ropathy was confirmed in a randomizedcontrolled trial (295). Although this firsttrial was only of 6 weeks’ duration, a sub-sequent follow-up study suggested thatsymptomatic relief could be maintainedfor at least 6 months (296). Side effects,however, are relatively common and sim-ilar to other opioid-like drugs. Similarly,two randomized trials have confirmed theefficacy of controlled-release oxycodonefor neuropathic pain in diabetes(297,298). Opioids such as oxycodonemay be considered as add-on therapies forpatients failing to respond to nonopioidmedications.

Topical and physical treatment.Topical nitrate. A recent controlled studysuggested that the local application to thefeet of isosorbide dinitrate spray was ef-fective in relieving overall pain and burn-ing discomfor t and the burningdiscomfort of DN (279). If confirmed bylarger randomized studies, this could of-fer a very useful alternative and localpharmacological treatment for relievingneuropathic symptoms.

Capsaicin. This alkaloid, which isfound in red pepper, depletes tissue ofsubstance P and reduces chemically in-duced pain. Several controlled studiescombined in meta-analyses seem to pro-vide some evidence of efficacy in diabeticneuropathic pain (299). However, trueblinding of these studies has been ques-tioned because of the local hyperalgesia ex-perienced when applying the active drug.Its use is only recommended for up to 8weeks of treatment, and it seems to be mostuseful in those with localized discomfort.

Acupuncture. A number of unmaskedstudies support the use of acupuncture.In the most recent published report, ben-efits of acupuncture lasted for up to 6months, and reduced use of other analge-sics was reported (300). The conduct ofpotential blinded studies of acupunctureis problematic; although a placebo re-sponse is possible with acupuncture, thisresponse should not detract from its use,which is generally without side effects.

Other physical therapies. Many otherphysical therapies have been proposed.Controlled evidence has been providedfor the use of percutaneous nerve stimu-lation (301) and, most recently, staticmagnetic field therapy (302).

Electrical spinal cord stimulation. Acase series of patients with severe painfulneuropathy unresponsive to conventionaltherapy suggested efficacy of using an im-planted spinal cord stimulator (303).However, this cannot be generally recom-mended except in very resistant cases, asit is invasive, expensive, and unproven incontrolled studies.

SECTION 6: NEUROPATHYAND ITS LATE SEQUELAE — Thelate sequelae of DPN are recognized to befoot ulceration (which may occasionallyresult in amputation) and, less com-monly, Charcot ’s neuroarthropathy(27,153,304). The importance of DPN inthe etiopathogenesis of foot ulceration hasbeen confirmed in several prospective

studies (50,182,196). Both large- andsmall-fiber somatic as well as sympatheticautonomic dysfunction have been impli-cated in the pathway to ulceration(153,182,196,305). However, it must beremembered that the neuropathic footdoes not ulcerate spontaneously; it is thecombination of neuropathy with eitherextrinsic factors (e.g., ill-fitting shoe gearor foreign body in shoe) or intrinsic fac-tors (e.g., high foot pressures or plantarcallus) that results in ulceration. In an ob-servational study, Reiber et al. (306) ap-plied Rothman’s model of causation to thepathogenesis of foot ulceration and re-ported that the most common pathway todiabetic foot ulceration comprised thecombination of neuropathy, trauma, andfoot deformity. Although it is generallybelieved that education and preventativefoot care should reduce the risk of ulcer-ation in high-risk individuals, a recentsystematic review could find few data tosupport this contention (307). However,one large randomized study of a screeningand protection program reported a non-significant trend to reduced ulceration;significantly, those in the interventiongroup who developed ulcers were lesslikely to proceed to amputation (308).This suggests a potential benefit of screen-ing and education. For a more extensivediscussion on the relation between neu-ropathy and foot ulceration, please con-sult the technical review on this topic(153).

Charcot neuroarthropathy is a rareand disabling condition affecting thebones and joints of the foot. It particularlyaffects patients with both somatic and au-tonomic neuropathy who have intact pe-ripheral circulation (27). Although theoverall prevalence of Charcot neuroar-thropathy in the diabetic population islow, a study of a randomly selected neu-ropathic population reported radiologicalevidence of Charcot neuroarthropathy in16% of patients (309), suggesting a keyrole of neuropathy in the pathogenesis ofthis condition. For further discussion ofCharcot neuroarthropathy, consult thereview by Sanders and Frykberg (310).

SECTION 7: CONCLUSIONS —DPN, which may be asymptomatic in upto 50% of cases, is one of the most com-mon complications of diabetes. Every di-abetic patient, regardless of type, shouldundergo a careful clinical examination ofthe lower extremities and feet at least once



a year (18). A number of simple screeningmethods that are applicable to clinicalpractice, including the MNSI (153), the10-g monofilament (165), and the modi-fied NDS (Fig. 1) (52), have been devel-oped. Of these screening tests, themodified NDS (Fig. 1) has been proven ina large prospective study to be predictiveof insensate foot ulceration: those withNDS �6 have a sixfold increased risk ofdeveloping an ulcer (52). Those patientswith foot ulcer risk require more frequentreview, regular podiatric care, and foot-care education, as there is a suggestionthat these steps might result in earlier pre-sentation when ulcers develop (308). Inthose patients with atypical presentations(e.g., rapidly progressing motor deficits),alternative diagnoses, such as CIDP,should be considered.

Rarer somatic neuropathies associatedwith diabetes comprise the focal and multi-focal neuropathies, including amyotrophy.

A number of therapeutic choices areavailable for the management of symp-tomatic DPN, although few if any of thesewill influence the natural history of neu-ropathy. Several pathogenetic therapiesare currently under investigation. Surro-gate end points for such trials includeelectrophysiology and QST of large- andsmall-fiber function. It is anticipated thatnewer, noninvasive techniques to assessdirectly nerve fiber damage will be devel-oped and that these will replace biopsiesof nerve or skin.

References1. Dyck PJ, Katz KM, Karnes JL, Litchy WJ,

Klein R, Pach JM, Wilson DM, O’BrienPC, Melton LJ, Service FJ: The preva-lence by staged severity of various typesof diabetic neuropathy, retinopathy andnephropathy in a population-based co-hort: the Rochester Diabetic NeuropathyStudy. Neurology 43:817–824, 1993

2. Young MJ, Boulton AJM, McLeod AF,Williams DRR, Sonksen PH: A multicen-tre study of the prevalence of diabeticperipheral neuropathy in the UK hospi-tal clinic population. Diabetologia36:150–156, 1993

3. Kumar S, Ashe HC, Parnell LN, Fer-nando DJ, Tsigos C, Young RJ, Ward JD,Boulton AJM: The prevalence of foot ul-ceration and its correlates in type 2 dia-betes: a population-based study. DiabetMed 11:480–484, 1994

4. Cabezas-Cerrato J: The prevalence of di-abetic neuropathy in Spain: a study inprimary care and hospital clinic groups.

Diabetologia 41:1263–1269, 19985. Vinik AI, Maser RE, Mitchell B, Freeman

R: Diabetic autonomic neuropathy: atechnical review. Diabetes Care 26:1553–1579, 2003

6. Carrington AL, Abbott CA, Shaw JE, Vil-eikyte L, Van Schie CHM, Boulton AJM:Can motor nerve conduction velocitypredict foot problems in diabetic neu-ropathy over a 6-year outcome period?Diabetes Care 25:2010–2015, 2002

7. DCCT Research Group: The effect of in-tensive diabetes therapy on the develop-ment and progression of neuropathy.Ann Int Med 122:561–568, 1995

8. Partanen J, Niskanen L, Lehtinen J, Mer-vaala E, Siitonen O, Uusitupa M: Naturalhistory of peripheral neuropathy in pa-tients with non-insulin dependent dia-betes. New Engl J Med 333:39–84, 1995

9. Tesfaye S, Stevens LK, Stephenson JMand the Eurodiab IDDM study group:Prevalence of diabetic peripheral neu-ropathy and its relation to glycaemiccontrol and potential risk factors: theEurodiab IDDM complication study.Diabetologia 39:1377–1386, 1996

10. Oyibo S, Prasad YD, Jackson NJ, JudeEB, Boulton AJM: The relationship be-tween blood glucose excursions andpainful diabetic peripheral neuropathy:a pilot study. Diabet Med 19:870–873,2002

11. Boulton AJM, Malik RA: Diabetic neu-ropathy. Med Clin N Am 82:909–929,1998

12. Boulton AJM: Neuropathy and diabetes.Diabet Rev 7:235–410, 1999

13. Mendell JR, Sahenk Z: Painful sensoryneuropathy. N Engl J Med 248:1243–1255, 2003

14. Vinik AI, Park TS, Stansberry KB, Pit-tenger GL: Diabetic neuropathies. Diabe-tologia 43:957–973, 2000

15. Ziegler D, Luft D: Clinical trials for drugsagainst diabetic neuropathy: can wecombine scientific needs with clinicalpracticalities? Int Rev Neurobiol 50:431–463, 2002

16. Feldman EL: Oxidative stress and dia-betic neuropathy: a new understandingof an old problem. J Clin Invest 111:431–433, 2003

17. Boulton AJM: Treatment of symptomaticdiabetic neuropathy. Diabete Metab ResRev 19:S16–S21, 2003

18. Spruce MC, Potter J, Coppini DV: Thepathogenesis and management of pain-ful diabetic neuropathy. Diabet Med 20:88–98, 2003

19. Boulton AJM, Gries FA, Jervell JA:Guidelines for the diagnosis and outpa-tient management of diabetic peripheralneuropathy. Diabet Med 15:508–514,1998

20. American Diabetes Association, Ameri-

can Academy of Neurology: Report andrecommendations of the San AntonioConference on Diabetic Neuropathy(Consensus Statement). Diabetes Care11:592–597, 1988

21. Dyck PJ: Severity and staging of diabeticpolyneuropathy. In Textbook of DiabeticNeuropathy. Gries FA, Cameron NE,Low PA, Ziegler D, Eds. Stuttgart, Thi-eme, 2003, p. 170–175

22. Boulton AJM, Ward JD: Diabetic neu-ropathies and pain. Clin EndocrinolMetab 15:917–932, 1986

23. Thomas PK: Classification, differentialdiagnosis and staging of diabetic periph-eral neuropathy. Diabetes 46 (Suppl. 2):S54–S57, 1997

24. Thomas PK: Classification of the diabeticneuropathies. In Textbook of DiabeticNeuropathy. Gries FA, Cameron NE,Low PA, Ziegler D, Eds. Stuttgart, Thi-eme, 2003, p. 175–177

25. Said G: Different patterns of neuropa-thies in diabetic patients. In DiabeticNeuropathy. Boulton AJM, Ed. Cologne,Aventis, Academy Press, 2001, p. 16–41

26. UKPDS: Intensive blood glucose withsulphonylureas or insulin comparedwith conventional treatment and risk ofcomplications in patients with type 2 di-abetes. Lancet 352:837–853, 1998

27. Jude EB, Boulton AJM: End-stage com-plications of diabetic neuropathy. Diabe-tes Rev 7:395–410, 1999

28. Karpitskaya Y, Novak CB, MackinnonSE: Prevalence of smoking, obesity, dia-betes mellitus and thyroid disease in pa-tients with carpal tunnel syndrome. AmPlast Surg 48:269–273, 2002

29. Stewart JD, McKelvey R, Durcan L, Car-penter S, Karpati G: Chronic inflamma-tory demyelinating polyneuropathy(CIDP) in diabetics. J Neurol Sci 142:59–64, 1996

30. American Diabetes Association: Pro-ceedings of a consensus developmentconference on standardized measures indiabetic neuropathy. Diabetes Care 15:1079–1107, 1992

31. Dyck PJ, Dyck PJB: Diabetic polyneu-ropathy: section III. In Diabetic Neurop-athy. 2nd ed. Dyck PJ, Thomas PK, Eds.Philadelphia, W.B. Saunders. 1999, p.255–278

32. Apfel SC, Asbury AK, Bril V, Burns TM,Campbell JN, Chalk CH, Dyck PJ, DyckJB, Feldman EL, Fields HL, Grant IA,Griffin JW, Klein CJ, Lindblom U, LitchyWJ, Low PA, Melanson M, Mendell JR,Merren MD, O’Brien PC, Rendel M,Rizza RA, Service FJ, Thomas PK, WalkD, Wang AK, Wessel K, Windebank AJ,Ziegler D, Zochodne DW: Positive neu-ropathic sensory symptoms as endpointsin diabetic neuropathy trials. J NeurologSci 189:3–5, 2001



33. Peripheral Nerve Society: Diabetic poly-neuropathy in controlled clinical trials:consensus report of the peripheral nervesociety. Am Neurol 38:478–482, 1995

34. Shy ME, Frohman EM, So YT ArezzoJC, Cornblath DC, Giuliani MJ, Sub-committee of the American Academy ofNeurology: quantitative sensory testing.Neurology 602:898–906, 2003

35. Nishikawa T, Edelstein D, Du XL, Yama-gishi S, Matsumura T, Kaneda Y, YorekMA, Beebe D, Oates PJ, Hammes HP, Gi-ardino I, Brownlee M: Normalizing mito-chondrial superoxide production blocksthree pathways of hyperglycemic dam-age. Nature 404:787–790, 2000

36. Dyck PJ, Davies JL, Wilson DM, ServiceFJ, Melton LJ 3rd, O’Brien PC: Risk fac-tors for severity of diabetic polyneurop-athy: intensive longitudinal assessmentof the Rochester Diabetic NeuropathyStudy cohort. Diabetes Care 22:1479–1486, 1999

37. Sundkvist G, Dahlin LB, Nilsson H,Eriksson KF, Lindgarde F, Rosen I, Lat-timer SA, Sima AA, Sullivan K, GreeneDA: Sorbitol and myo-inositol levels andmorphology of sural nerve in relation toperipheral nerve function and clinicalneuropathy in men with diabetic, im-paired, and normal glucose tolerance.Diabet Med 17:259–268, 2000

38. Smith AG, Ramachandran P, Tripp S,Singleton JR: Epidermal nerve innerva-tion in impaired glucose tolerance anddiabetes-associated neuropathy. Neurol-ogy 13:1701–1704, 2001

39. Sumner CJ, Sheth S, Griffin JW, Corn-blath DR, Polydefkis M: The spectrum ofneuropathy in diabetes and impairedglucose tolerance. Neurology 60:108–111, 2003

40. Navarro X, Sutherland DE, KennedyWR: Long-term effects of pancreatictransplantation on diabetic neuropathy.Ann Neurol 42:727–736, 1997

41. Azad N, Emanuele NV, Abraira C, Hen-derson WG, Colwell J, Levin SR, NuttallFQ, Comstock JP, Sawin CT, Silbert C,Rubino FA: The effects of intensive gly-cemic control on neuropathy in the VACooperative Study on Type II DiabetesMellitus (VA CSDM). J Diabetes Compl13:307–313, 1999

42. Gaede P, Vedel P, Larsen N, Jensen GV,Parving HH, Pedersen O: Multifactorialintervention and cardiovascular diseasein patients with type 2 diabetes. N EnglJ Med 348:383–393, 2003

43. Oates PJ: Polyol pathway and diabeticperipheral neuropathy. Int Rev Neurobiol50:325–392, 2002

44. Kasajima H, Yamagishi S, Sugai S, Yagi-hashi N, Yagihashi S: Enhanced in situexpression of aldose reductase in pe-ripheral nerve and renal glomeruli in di-

abetic patients. Virchows Arch 439:46–54, 2001

45. Dyck PJ, Sherman WR, Hallcher LM,Service FJ, O’Brien PC, Grina LA,Palumbo PJ, Swanson CJ: Human dia-betic endoneurial sorbitol, fructose, andmyo-inositol related to sural nerve mor-phometry. Ann Neurol 8:590–596, 1980

46. Mayhew JA, Gillon KR, Hawthorne JN:Free and lipid inositol, sorbitol and sug-ars in sciatic nerve obtained post-mor-tem from diabetic patients and controlsubjects. Diabetologia 24:13–15, 1983

47. Hale PJ, Nattrass M, Silverman SH, Sen-nit C, Perkins CM, Uden A, Sundkvist G:Peripheral nerve concentrations of glu-cose, fructose, sorbitol and myoinositolin diabetic and non-diabetic patients.Diabetologia 30:464–467, 1987

48. Dyck PJ, Zimmerman BR, Vilen TH,Minnerath SR, Karnes JL, Yao JK, Po-duslo JF: Nerve glucose, fructose, sorbi-tol, myo-inositol, and fiber degenerationand regeneration in diabetic neuropa-thy. N Engl J Med 319:542–548, 1988

49. Shimizu H, Ohtani KI, Tsuchiya T, SatoN, Tanaka Y, Takahashi H, Uehara Y,Inukai T, Mori M: Aldose reductasemRNA expression is associated withrapid development of diabetic microan-giopathy in Japanese type 2 diabetic(T2DM) patients. Diabetes Nutr Metab13:75–79, 2000

50. Demaine AG: Polymorphisms of the al-dose reductase gene and susceptibility todiabetic microvascular complications.Curr Med Chem 10:1389–1398, 2003

51. Airey M, Bennett C, Nicolucci A, Wil-liams R: Aldose reductase inhibitors forthe prevention and treatment of diabeticperipheral neuropathy. Cochrane Data-base Syst Rev 2:CD002182, 2000

52. Abbott CA, Carrington AL, Ashe H, BathS, Every LC, Griffiths J, Hann AW, Hus-sein A, Jackson N, Johnson KE, RyderCH, Torkington R, Van Ross ER, Whal-ley AM, Widdows P, Williamson S,Boulton AJM, North-West Diabetes FootCare Study: The North-West DiabetesFoot Care Study: incidence of, and riskfactors for, new diabetic foot ulcerationin a community-based patient cohort.Diabet Med 19:377–384, 2002

53. Greene DA, Arezzo JC, Brown MB: Effectof aldose reductase inhibition on nerveconduction and morphometry in dia-betic neuropathy: Zenarestat StudyGroup. Neurology 53:580–591, 1999

54. Hotta N, Toyota T, Matsuoka K, ShigetaY, Kikkawa R, Kaneko T, Takahashi A,Sugimura K, Koike Y, Ishii J, SakamotoN, the SNK-860 Diabetic NeuropathyStudy Group: Clinical efficacy of fidares-tat, a novel aldose reductase inhibitor,for diabetic peripheral neuropathy: a 52-week multicenter placebo-controlled

double-blind parallel group study. Dia-betes Care 24:1776–1782, 2001

55. King RH: The role of glycation in thepathogenesis of diabetic polyneurop-athy. Mol Pathol 54:400–408, 2001

56. McLennan SV, Martell SK, Yue DK: Ef-fects of mesangium glycation on matrixmetalloproteinase activities: possiblerole in diabetic nephropathy. Diabetes51:2612–2618, 2002

57. Portero-Otin M, Pamplona R, BellmuntMJ, Ruiz MC, Prat J, Salvayre R, Negre-Salvayre A: Advanced glycation endproduct precursors impair epidermalgrowth factor receptor signalling. Diabe-tes 51:1535–1542, 2002

58. Ryle C, Donaghy M: Non-enzymatic gly-cation of peripheral nerve proteins inhuman diabetics. J Neurol Sci 129:62–68, 1995

59. Sugimoto K, Nishizawa Y, Horiuchi S,Yagihashi S: Localization in human dia-betic peripheral nerve of N (epsilon)-carboxymethyllysine-protein adducts,an advanced glycation end product. Dia-betologia 40:1380–1387, 1997

60. Amano S, Kaji Y, Oshika T, Oka T,Machinami R, Nagai R, Horiuchi S: Ad-vanced glycation end products in hu-man optic nerve head. Br J Ophthalmol85:52–55, 2001

61. Zotova EV, Chistiakov DA, Savost’ianovKV, Bursa TR, Galeev IV, Strokov IA,Nosikov VV: Association of the SOD2Ala(-9)Val and SOD3 Arg213Gly poly-morphisms with diabetic polyneurop-athy in patients with diabetes mellitustype 1. Mol Biol (Mosk) 37:404–408,2003

62. Reljanovic M, Reichel G, Rett K, LobischM, Schuette K, Moller W, Tritschler HJ,Mehnert H: Treatment of diabetic poly-neuropathy with the antioxidant thiocticacid (alpha-lipoic acid): a two year mul-ticentre randomized double-blind pla-cebo-controlled trial (ALADIN II). FreeRadic Res 31:171–179, 1999

63. Ziegler D, Hanefeld M, Ruhnau KJ,Hasche H, Lobisch M, Schutte K, KerumG, Malessa R: Treatment of symptomaticdiabetic polyneuropathy with the anti-oxidant alpha-lipoic acid: a 7-monthmulticentre randomized controlled trial(ALADIN III Study). ALADIN III StudyGroup. Diabetes Care 22:1296–301,1999

64. Ametov AS, Barinov A, Dyck PJ, HermannR, Kozlova N, Litchy WJ, Low PA, Nehr-dich D, Novosadova M, O’Brien PC, Rel-janovic M, Samigullin R, Schuette K,Strokov I, Tritschler HJ, Wessel K,Yakhno N, Ziegler D, SYDNEY TrialStudy Group: The sensory symptoms ofdiabetic polyneuropathy are improvedwith �-lipoic acid: the SYDNEY trial.Diabetes Care 26:770–776, 2003



65. Young MJ, Veves A, Walker MG, Boul-ton AJM: Correlations between nervefunction and tissue oxygenation in dia-betic patients: further clues to the aetiol-ogy of diabetic neuropathy? Diabetologia35:1146–1150, 1992

66. Veves A, Donaghue VM, Sarnow MR, Gi-urini JM, Campbell DR, LoGerfo FW:The impact of reversal of hypoxia by re-vascularization on the peripheral nervefunction of diabetic patients. Diabetolo-gia 39:344–348, 1996

67. Akbari CM, Gibbons GW, HabershawGM, LoGerfo FW, Veves A: The effect ofarterial reconstruction on the naturalhistory of diabetic neuropathy. Arch-Surg 132:148–152, 1997

68. Malik RA, Williamson S, Abbott CA,Carrington AL, Iqbal J, Schady W Boul-ton AJM: Effect of angiotensin-convert-ing enzyme (ACE) inhibitor trandaloprilon human diabetic neuropathy: ran-domised double-blind controlled trial.Lancet 352:1978–1981, 1998

69. Estacio RO, Jeffers BW, Gifford N,Schrier RW: Effect of blood pressurecontrol on diabetic microvascular com-plications in patients with hypertensionand type 2 diabetes. Diabetes Care 23(Suppl. 2):B54–B64, 2000

70. Eichberg J: Protein kinase C changes indiabetes: is the concept relevant to neu-ropathy? Int Rev Neurobiol 50:61–82,2002

71. Cameron NE, Cotter MA: Effects of pro-tein kinase C beta inhibition on neuro-vascular dysfunction in diabetic rats:interaction with oxidative stress and es-sential fatty acid dysmetabolism. DiabeteMetab Res Rev 18:315–323, 2002

72. Litchy W, Dyck PJ, Tesfaye S, Zhang D,Bastyr E, The MBBQ Study Group: Dia-betic peripheral neuropathy (DPN) as-sessed by neurological examination(NE) and composite scores (CS) is im-proved with LY333531 treatment. Dia-betes 45 (Suppl. 2):S197, 2002

73. The EURODIAB Prospective Complica-tions Study (PCS) Group: Cardiovascularrisk factors predict diabetic peripheralneuropathy in type 1 subjects in Europe.Diabetologia 42:A50–A181, 1999

74. Okamoto T, Yamagishi SI, Inagaki Y,Amano S, Koga K, Abe R, Takeuchi M,Ohno S, Yoshimura A, Makita Z: Angio-genesis induced by advanced glycationend products and its prevention by cer-ivastatin. FASEB J 16:1928–1930, 2002

75. Nangle MR, Cotter MA, Cameron NE:Effects of rosuvastatin on nitric oxide-dependent function in aorta and corpuscavernosum of diabetic mice: relation-ship to cholesterol biosynthesis pathwayinhibition and lipid lowering. Diabetes52:2396–2402, 2003

76. Fried LF, Forrest KY, Ellis D, Chang Y,

Silvers N, Orchard TJ: Lipid modulationin insulin-dependent diabetes mellitus:effect on microvascular outcomes. J Di-abetes Compl 15:113–119, 2001

77. Backes JM, Howard PA: Association ofHMG-CoA reductase inhibitors withneuropathy. Ann Pharmacother 37:274–278, 2003

78. Anand P, Terenghi G, Warner G, Kopel-man P, Williams-Chestnut RE, SinicropiDV: The role of endogenous nervegrowth factor in human diabetic neu-ropathy. Nat Med 2:703–707, 1996

79. Diemel LT, Cai F, Anand P, Warner G,Kopelman PG, Fernyhough P, Tomlin-son DR: Increased nerve growth factormRNA in lateral calf skin biopsies fromdiabetic patients. Diabet Med 16:113–118, 1999

80. Kennedy AJ, Wellmer A, Facer P, Sal-danha G, Kopelman P, Lindsay RM,Anand P: Neurotrophin-3 is increased inskin in human diabetic neuropathy.J Neurol Neurosurg Psychiatry65:393–395, 1998

81. Lee DA, Gross L, Wittrock DA, Winde-bank AJ: Localization and expression ofciliary neurotrophic factor (CNTF) inpostmortem sciatic nerve from patientswith motor neuron disease and diabeticneuropathy. J Neuropathol Exp Neurol55:915–923, 1996

82. Terenghi G, Mann D, Kopelman PG,Anand P: trkA and trkC expression isincreased in human diabetic skin. Neu-rosci Lett 228:33–36, 1997

83. Apfel SC, Kessler JA, Adornato BT,Litchy WJ, Sanders C, Rask CA: Recom-binant human nerve growth factor in thetreatment of diabetic polyneuropathy:NGF Study Group. Neurology 51:695–702, 1998

84. Apfel SC, Schwartz S, Adornato BT,Freeman R, Biton V, Rendell M, Vinik A,Giuliani M, Stevens JC, Barbano R, DyckPJ: Efficacy and safety of recombinanthuman nerve growth factor in patientswith diabetic polyneuropathy: a ran-domized controlled trial. JAMA 284:2215–2221, 2000

85. Wellmer A, Misra VP, Sharief MK, Ko-pelman PG, Anand P: A double-blindplacebo-controlled clinical trial of re-combinant human brain-derived neuro-trophic factor (rhBDNF) in diabeticpolyneuropathy. J Peripher Nerv Syst6:204–210, 2001

86. Delaney CL, Russell JW, Cheng HL,Feldman EL: Insulin-like growth factor-Iand over-expression of Bcl-xL preventglucose-mediated apoptosis in Schwanncells. J Neuropathol Exp Neurol 60:147–160, 2001

87. Schmidt RE, Dorsey DA, Beaudet LN,Peterson RG: Analysis of the Zucker Di-abetic Fatty (ZDF) type 2 diabetic rat

model suggests a neurotrophic role forinsulin/IGF-I in diabetic autonomic neu-ropathy. Am J Pathol 163:21–28, 2003

88. Grandis M, Nobbio L, Abbruzzese M,Banchi L, Minuto F, Barreca A, GarroneS, Mancardi GL, Schenone A: Insulintreatment enhances expression of IGF-Iin sural nerves of diabetic patients. Mus-cle Nerve 24:622–629, 2001

89. Sima AA: C-peptide and diabetic neu-ropathy. Expert Opin Investig Drugs 12:1471–1488, 2003

90. Cotter MA, Ekberg K, Wahren J, Cam-eron NE: Effects of proinsulin C-peptidein experimental diabetic neuropathy:vascular actions and modulation by ni-tric oxide synthase inhibition. Diabetes52:1812–1817, 2003

91. Ekberg K, Brismar T, Johansson BL, Jons-son B, Lindstrom P, Wahren J: Amelio-ration of sensory nerve dysfunction byC-peptide in patients with type 1 diabe-tes. Diabetes 52:536–541, 2003

92. Carmeliet P, Storkebaum E: Vascularand neuronal effects of VEGF in the ner-vous system: implications for neurolog-ical disorders. Semin Cell Dev Biol 13:39–53, 2002

93. Schratzberger P, Walter DH, Rittig K,Bahlmann FH, Pola R, Curry C, Silver M,Krainin JG, Weinberg DH, Ropper AH,Isner JM: Reversal of experimental dia-betic neuropathy by VEGF gene transfer.J Clin Invest 107:1083–1092, 2001

94. Samii A, Unger J, Lange W: Vascular en-dothelial growth factor expression in pe-ripheral nerves and dorsal root ganglia indiabetic neuropathy in rats. Neurosci Lett262:159–162, 1999

95. Isner JM, Ropper A, Hirst K: VEGF genetransfer for diabetic neuropathy. HumanGene Ther 12:1593–1594, 2001

96. Srinivasan S, Stevens MJ, Sheng H, HallKE, Wiley JW: Serum from patients withtype 2 diabetes with neuropathy inducescomplement-independent, calcium-de-pendent apoptosis in cultured neuronalcells. J Clin Invest 102:1454–1462, 1998

97. Pittenger GL, Malik RA, Burcus N, Boul-ton AJ, Vinik AI: Specific fiber deficits insensorimotor diabetic polyneuropathycorrespond to cytotoxicity against neu-roblastoma cells of sera from patientswith diabetes. Diabetes Care 22:1839–1844, 1999

98. Wilbourn AJ: Diabetic entrapment andcompression neuropathies. In DiabeticNeuropathy. Dyck PJ, Thomas PK, Eds.Philadelphia, WB Saunders, 1999, p.481–508

99. Schady W, Abuaisha B, Boulton AJM:Observations on severe ulnar neuropa-thy in diabetes. J Diabetes Compl 12:128–132, 1998

100. Goldberg BJ, Light TR, Blair SJ: Ulnarneuropathy at the elbow: results of me-



dial epicondylectomy. J Hand Surg 1419:182–188, 1989

101. Brown WF, Watson BF: AAEM case re-port #27: acute retro humeral radialneuropathies. Muscle Nerve 16:706–711, 1993

102. Garland H, Moorehouse D: Compressivelesions of the external popliteal (com-mon peroneal) nerve. Br Med J 21:1373–1378, 1952

103. Katirji MB, Wilbourn AJ: Common per-oneal neuropathy: a clinical and electro-physiological study of 116 cases.Neurology 38:1723–1728, 1988

104. Mulder DW, Lambert EH, Bastron JA:The neuropathies associated with diabe-tes mellitus. Neurology 13:1021–1030,1961

105. Watanabe K, Hagura R, Akanuma Y,Takasu T, Kajinuma H, Kuzuya N, IrieM: Characteristics of cranial nerve pal-sies in diabetic patients. Diabetes Res ClinPract 10:19–27, 1990

106. Richards BW, Jones FR, Younge BR:Causes and prognosis in 4,278 cases ofparalysis of the oculomotor, trochlearand abducens cranial nerves. Am J Oph-thalmol 113:489–496, 1992

107. Goldstein JE, Cogan DG: Diabetic oph-thalmoplegia with special reference tothe pupil. Arch Ophthalmol 64:592–598,1960

108. Asbury AK, Aldredge H, Hershberg R,Fisher CM: Oculomotor palsy in diabetesmellitus: a clinico-pathological study.Brain 93:955–966, 1970

109. Usui Y, Mukoyama M, Hashizume M,Takahashi A: Diabetic ophthalmoplegia:a clinicopathological study of the firstcase in Japan. Clin Neurol 29:442–449,1989

110. Smith BE, Dyck PJ: Subclinical his-topathological changes in the oculomo-tor nerve in diabetes mellitus. AnnNeurol 32:376–385, 1992

111. Aminoff MJ, Miller AL: The prevalenceof diabetes mellitus in patients withBell’s palsy. Acta Neurol Scand 48:381–384, 1972

112. Abraham-Inpijn L, Devriese PP, HartAA: Predisposing factors in Bell’s palsy: aclinical study with reference to diabetesmellitus, hypertension, clotting mecha-nisms and lipid disturbance. Clin Otolar-yngol 7:99–105, 1982

113. Abraham-Inpijn L, Oosting J, Hart AA:Bell’s palsy: Factors affecting prognosisin 200 patients with reference to hyper-tension and diabetes mellitus. Clin Oto-laryngol 12:349–355, 1987

114. Stevens JC, Smith BE: Cranial reflexes.In Clinical Neurophysiology. Daube JR,Ed. Philadelphia, FA Davis, 1996, p.321–335

115. Malik RA, Kallinikos P, Abbott CA, vanSchie CHM, Morgan P, O’Donnell C,

Efron N, Boulton AJM: Corneal confocalmicroscopy: a non-invasive surrogate ofnerve fibre damage and repair in diabeticpatients. Diabetologia 46:683–688, 2003

116. Smith BE: Cranial neuropathy in diabe-tes mellitus. In Diabetic Neuropathy.Dyck PJ, Thomas PK, Eds. Philadelphia,WB Saunders, 1999, p. 457–467

117. Coppack SW, Watkins PJ: The naturalhistory of diabetic femoral neuropathy.Q J Med 79:307–313, 1991

118. Asbury AK: Proximal diabetic neuropa-thy. Ann Neurol 2:179–180, 1977

119. Said G, Thomas PK: Proximal diabeticneuropathy. In Diabetic Neuropathy.Dyck PJ, Thomas PK, Eds. Philadelphia,WB Saunders, 1999, p. 474–480

120. Raff MC, Asbury AK: Ischaemic mono-neuropathy and mononeuropathy mul-tiplex in diabetes mellitus. N Engl J Med279:17–22, 1968

121. Chokroverty S: AAEE case report 13: di-abetic amyotrophy. Muscle Nerve 10:679–684, 1987

122. Pascoe MK, Low PA, Windebank AJ:Subacute diabetic proximal neuropathy.Mayo Clin Proc 72:1123–1132, 1997

123. Subramony SH, Wilbourn AJ: Diabeticproximal neuropathy: clinical and elec-tromyographic studies. J Neurol Sci 53:293–304, 1982

124. Said G, Elgrably F, Lacroix C, Plante V,Talamon C, Adams D, Tager M, Slama G:Painful proximal diabetic neuropathy:inflammatory nerve lesions and sponta-neous favourable outcome. Ann Neurol41:762–770, 1997

125. Said G, Goulon-Goeau C, Lacroix C,Moulonguet A: Nerve biopsy findings indifferent patterns of proximal diabeticneuropathy. Ann Neurol 35:559–569,1994

126. Llewelyn JG, Thomas PK, King RHM:Epineurial microvasculitis in proximaldiabetic neuropathy. J Neurol 245:159–165, 1998

127. Dyck PJB, Norell JE, Dyck PJ: Microvas-culitis and ischemia in diabetic lumbo-sacral radiculoplexus neuropathy.Neurology 53:2113–2121, 1999

128. Kelkar P, Masood M, Parry GJ: Distinc-tive pathologic findings in proximaldiabetic neuropathy [diabetic amyotro-phy]. Neurology 55:83–88, 2000

129. Younger DS, Rosoklija G, Hays AP, Tro-jaborg W, Latov N: Diabetic peripheralneuropathy: a clinical and immunohis-tological analysis of sural nerve biopsies.Muscle Nerve 19:722–727, 1996

130. Ogawa T, Taguchi T, Tanaka Y, IkeguchiK, Nakano I: Intravenous immunoglob-ulin therapy for diabetic amyotrophy.Intern Med 40:349–352, 2001

131. Dyck PJ, Windebank AJ: Diabetic andnondiabetic lumbosacral radiculoplexusneuropathies: new insights into patho-

physiology and treatment. Muscle Nerve25:477–491, 2002

132. Stewart JD: Diabetic truncal neuropathy:topography of the sensory deficit. AnnNeurol 25:233–238, 1989

133. Kitka DG, Breuer AC, Wilbourn AJ:Thoracic root pain in diabetes: the spec-trum of clinical and electromyographicfindings. Ann Neurol 11:80–91, 1983

134. Sun SF, Streib EW: Diabetic thoracoab-dominal neuropathy: clinical and elect-rodiagnostic features. Ann Neurol 9:75–79, 1981

135. Haq RU, Pendlebury WW, Fries TJ, Tan-dan R: Chronic inflammatory demyelinat-ing polyradiculoneuropathy in diabeticpatients. Muscle Nerve 27:465–470, 2003

136. Wilson JR, Park Y, Fisher MA: Electrodi-agnostic criteria in CIDP: comparisonwith diabetic neuropathy. ElectromyogrClin Neurophysiol 40:181–185, 2000

137. Vital C, Vital A, Lagueny A, Ferrer X,Fontan D, Barat M, Gbikpi-Benissan G,Orgogozo JM, Henry P, BrechenmacherC, Bredin A, Desbordes P, Ribiere-Bach-elier C, Latinville D, Julien J, Petry KG:Chronic inflammatory demyelinatingpolyneuropathy: immunopathologicaland ultrastructural study of peripheralnerve biopsy in 42 cases. UltrastructPathol 24:363–369, 2000

138. Cocito D, Ciaramitaro P, Isoardo G, Bar-bero P, Migliaretti G, Pipieri A, Proto G,Quadri R, Bergamasco B, Durelli L: In-travenous immunoglobulin as first treat-ment in diabetics with concomitantdistal symmetric axonal polyneuropathyand CIDP. J Neurol 249:719–722, 2002

139. Sharma KR, Cross J, Ayyar DR, Mar-tinez-Arizala A, Bradley WG: Diabeticdemyelinating polyneuropathy respon-sive to intravenous immunoglobulintherapy. Arch Neurol 59:751–757, 2002

140. Eaton S, Tesfaye S: Clinical manifesta-tions and measurement of neuropathy.Diabetes Rev 7:312–325, 1999

141. Archer AG, Watkins PJ, Thomas PK,Sharma AK, Payan J: The natural historyof acute painful neuropathy is diabetes.J Neurol Neurosurg Psych 48:491–499,1983

142. Watkins PJ: Pain and diabetic neuropa-thy. Br Med J 288:168–169, 1984

143. Steel JM, Young RJ, Lloyd GG, ClarkeBF: Clinically apparent eating disordersin young diabetic women: associationswith painful neuropathies and othercomplications. Br Med J 296:859–802,1987

144. Llewelyn G, Gilbey SG, Thomas PK,King RH, Muddle JR, Watkins PJ: Suralnerve morphomety in diabetic auto-nomic and painful sensory neuropathy:a clinicopathological study. Brain 114:867–892, 1991

145. Otto M, Bak S, Bach FW, Jensen TS, Sin-



drup SH: Pain phenomena and possiblemechanism in patients with painfulpolyneuropathy. Pain 101:187–192,2003

146. Tesfaye S, Malik RA, Harris N,Jakubowski JJ, Mody C, Rennie IG,Ward JD: Arterio-venous shunting andproliferating new vessels in acute painfulneuropathy of rapid glycemic control(insulin neuritis). Diabetologia 39:329–335, 1996

147. Boulton AJM, Drury J, Clarke B, WardJD: Continuous subcutaneous insulininfusion in the management of painfuldiabetic neuropathy. Diabetes Care 5:386–390, 1982

148. Harris MI, Eastman R, Cowie C: Symp-toms of sensory neuropathy in adultswith NIDDM in the U.S. population. Di-abetes Care 16:1446–1452, 1993

149. Van Deursen RW, Sanchez MM, Ul-brecht JS, Cavanagh PR: The role ofmuscle spindles in ankle movement per-ception in human subjects with diabeticneuropathy. Exp Brain Res 120:1–8,1998

150. Cavanagh PR, Simoneau GG, UlbrechtJS: Ulceration, unsteadiness and uncer-tainty: the biomechanical consequencesof diabetes mellitus. J Biomech 26 (Suppl.1):23–46, 1993

151. Katoulis EC, Ebdon-Parry M, Hollis S,Harrison AJ, Vileikyte L, Kulkani J, Boul-ton AJM: Postural instability in diabeticpatients at risk of foot ulceration. DiabetMed 14:296–300, 1997

152. Katoulis EC, Ebdon-Parry M, Lansham-mar H, Vileikyte L, Kulkani J, BoultonAJM: Gait abnormalities in diabetic neu-ropathy. Diabetes Care 20:1904–1907,1997

153. Mayfield JA, Reiber GE, Sanders LJ,Janisse D, Pogach LM: Preventive footcare in people with diabetes. DiabetesCare 21:2161–2177, 1998

154. Huskisson EC: Measurement of pain.Lancet ii:1127–1131, 1976

155. Feldman EL, Stevens MJ, Thomas PK,Brown MB, Canal N, Greene DA: A prac-tical two-step quantitative clinical andelectrophysiological assessment for thediagnosis and staging of diabetic neu-ropathy. Diabetes Care 17:1281–1289,1996

156. Meijer JW, Smit AJ, Sondersen EV,Groothoff JW, Eisma WH, Links TP:Symptom scoring systems to diagnosedistal polyneuropathy in diabetes: theDiabetic Neuropathy Symptom Score.Diabet Med 19:962–965, 2002

157. Scott J, Huskisson EC: Graphic repre-sentation of pain. Pain 2:175–186, 1976

158. Ziegler D: Treatment of neuropathicpain. In Textbook of Diabetic Neuropathy.Gris FA, Cameron NE, Low PA, Ziegler

D, Eds. Stuttgart, Thieme, 2003, p. 211–226

159. Franse LV, Valk GD, Dekker JH, HeineRJ, Van Eijk JTM: “Numbness of the feet”is a poor indicator for polyneuropathy intype 2 diabetic patients. Diabetes Care17:105–110, 2000

160. Vileikyte L: Psychological aspects of di-abetic peripheral neuropathy. DiabetesRev 7:387–394, 1999

161. Vileikyte L, Peyrot M, Bundy C, RubinPR, Leventhal H, Mora P, Shaw JE, BakerP, Boulton AJM: The development andvalidation of a neuropathy and foot ulcerspecific quality of life rate. Diabetes Care26:2549–2555, 2003

162. Dyck PJ, Melton LJ, O’Brien PC, ServiceFJ: Approaches to improve epidemiolog-ical studies of diabetic neuropathy. Dia-betes 46 (Suppl. 2):S5–S13, 1997

163. Perkins BA, Olaleye D, Zinman B, Bril V:Simple screening tests for peripheralneuropathy in the diabetes clinic. Diabe-tes Care 24:250–256, 2001

164. Bril V, Perkins BA: Validation of the To-ronto clinical scoring system for diabeticpolyneuropathy.DiabetesCare25:2048–2052, 2002

165. Dyck PJ, Turner DW, Davies JL, O’BrienPC, Dyck PJ, Rask CA: Electronic case-report forms of symptoms and impair-ments of peripheral neuropathy. CanJ Neurol Sci 29:258–266, 2002

166. Valk GD, Nauta JJP, Strijem RLM, Ber-telsmann FW: Clinical examination ver-sus neurophysiological examination inthe diagnosis of diabetic polyneurop-athy. Diabet Med 9:716–721, 1992

167. Arezzo JC: Quantitative sensory testing.In Textbook of Diabetic Neuropathy. GrisFA, Cameron NE, Low PA, Ziegler D,Eds. Stuttgart, Thieme, 2003, p. 184–189

168. Valk GD, de Sonnaville JJ, van HoutumWH, Heine RJ, van Eijk JT, Bouter LM,Bertelsmann FW: The assessment of di-abetic polyneuropathy in daily practice:reproducibility and validity of Semmes-Weinstein monofilaments and clinicalneurological examination. Muscle Nerve20:116–118, 1997

169. Mayfield JA, Sugarman JR: The use ofSemmes-Weinstein monofilament andother threshold tests for preventing footulceration and amputation in peoplewith diabetes. J Fam Pract 49 (Suppl.):517–529, 2000

170. Kumar S, Fernando DJS, Veves A,Knowles EA, Young MJ, Boulton AJM:Semmes-Weinstein monofilaments: asimple effective and inexpensive screen-ing device for identifying diabetic pa-tients at risk of foot ulceration. DiabetesRes Clin Pract 13:63–68, 1991

171. Armstrong DG, Lavery LA, Vela SA,Quebedeaux TC, Fleischli JG: Choosing

a practical screening instrument to iden-tify patients at risk of diabetic foot ulcer-ation. Arch Int Med 153:289–292, 1998

172. Miranda-Palma B, Basu S, Mizel MD, So-senko JM, Boulton AJM: The monofila-ment as the gold standard for foot ulcerrisk screening: a reappraisal (Abstract).Diabetes 52 (Suppl. 1):A63, 2003

173. Booth J, Young MJ: Differences in theperformance of commercially availablemonofilaments. Diabetes Care 23:984–988, 2000

174. Hilz MJ, Axelrod FB, Hermann K, HaertlU, Duetsh M, Neundorfer B: Normativevalues of vibratory perception in 530children, juveniles and adults aged 3–79years. J Neurol Sci 159:219–225, 1998

175. Shin JB, Seong YJ, Lee HJ, Kim SH, ParkJR: Foot screening technique in diabeticpopulations. J Korean Med Sci 15:78–82,2000

176. Vileikyte L, Hutchings G, Hollis S, Boul-ton AJM: The tactile circumferential dis-criminator: a new simple screeningdevice to identify diabetic patients at riskof foot ulceration. Diabetes Care 20:623–626, 1997

177. Paisley AN, Abbott CA, van Schie CHM,Boulton AJM: A comparison of the Neu-ropen against standard quantitative sen-sory threshold measures for assessingperipheral nerve function. Diabet Med19:400–405, 2002

178. Dyck PJ, Karnes JL, O’Brien PC, LitchyWJ, Low PA, Melton LJ 3rd: The Roch-ester Diabetic Neuropathy Study: reas-sessment of tests and criteria fordiagnosis and staged severity. Neurol 42:1164–1170, 1992

179. Abad F, Diaz-Gomez NM, Rodriguez I,Perez R, Delgado JA: Subclinical painand thermal sensory dysfunction in chil-dren and adolescents with type 1 diabe-tes mellitus. Diabet Med 19:827–831,2002

180. Dyck PJ, Davies JL, Litchy WJ, O’BrienPC: Longitudinal assessment of diabeticpolyneuropathy using a composite scorein the Rochester Diabetic NeuropathyStudy cohort. Neurol 49:229–239, 1997

181. Coppini DV, Wellmer A, Weng C,Young PJ, Anand P, Sonksen PH: Thenatural history of diabetic peripheralneuropathy determined by a 12 yearprospective study using vibration per-ception thresholds. J Clin Neurosci 8:520–524, 2001

182. Young MJ, Breddy JL, Veves A, BoultonAJM: The prediction of diabetic foot ul-ceration using vibration perceptionthresholds: a prospective study. DiabetesCare 17:557–560, 1994

183. Kastenbauer T, Sauseng S, Sokol G, Au-inger M, Irsigler K: A prospective studyof predictors for foot ulceration in type 2



diabetes. J Am Podiatr Med Assoc 91:343–350, 2001

184. Gerr F, Letz R: Covariates of human pe-ripheral function: vibrotactile and ther-mal thresholds. II. Neurotoxicol Teratol16:105–112, 1994

185. Gelber DA, Pfeifer MA, Broadstone VL:Components of variance for vibratoryand thermal thresholds testing in normaland diabetic subjects. J Diabetes Compli-cations 9:170–176, 1995

186. Dyck PJ, Kennedy WR, Kesserwani H,Melanson M, Ochoa J, Shy M, StevensJC, Suarez GA, O’Brien PC: Limitationsof quantitative sensory testing when pa-tients are biased toward a bad outcome.Neurol 50:1213, 1998

187. Vinik AI, Suwanwalailom S, StansberryKB, Holland MT, McNitt PM, Colen IE:Quantitative measurement of cutaneousperception in diabetic neuropathy. Mus-cle Nerve 18:574–584, 1995

188. Zaslansky R, Tarnitsky D: Clinical appli-cations of quantitative sensory testing(QST). J Neurol Sci 153:215–238, 1998

189. Dyck PJ, O’Brien PC: Quantitative sen-sation testing in epidemiological andtherapeutic studies of peripheral neu-ropathy. Muscle Nerve 22:659–662,1999

190. Bril V, Perkins BA: Comparison of vibra-tion perception thresholds obtainedwith the Neurothesiometer and theCASE IV and relationship to nerve con-duction studies. Diabet Med 19:661–666, 2002

191. Ziegler D, Mayer P, Wiefels K, Gries FA:Evaluation of thermal, pain, and vibra-tion sensation thresholds in newly diag-nosed type 1 diabetic patients. J NeurolNeurosurg Psychiatry 11:1420–1424,1988

192. Dyck PJ, Karnes J, O’Brien PC, Zimmer-man IR: Detection thresholds of cutane-ous sensation in humans. In PeripheralNeuropathy. Dyck PJ, Thomas PK, Grif-fin JW, Low PA, Poduslo JF, Eds. Phila-delphia, W. B. Saunders, 1993, p. 706–728

193. Dyck PJ, Dyck PJB, Velosa JA, Larson TS,O’Brien PC, The Nerve Growth FactorsStudy Group: Patterns of quantitativesensation testing of hypoesthesia andhyperalgesia are predictive of diabeticpolyneuropathy: a study of three co-horts. Diabetes Care 23:510–517, 2000

194. Davis EA, Walsh P, Jones TW, ByrneGC: The use of biosthesiometry to detectneuropathy in children and adolescentswith IDDM. Diabetes Care 20:1448–1453, 1997

195. Boulton AJM, Kubrusly DB, Bowker JH,Skyler JS, Sosenko JM: Impaired vibra-tory perception and diabetic foot ulcer-ation. Diabet Med 3:335–337, 1986

196. Abbott CA, Vileikyte L, Williamson S,

Carrington AL, Boulton AJM: Multi-center study of the incidence of and pre-dictive risk factors for diabeticneuropathic foot ulceration. DiabetesCare 7:1071–1075, 1998

197. Iggo A: Cutaneous thermoreceptors. InSomatosensory Mechanisms. von Euler C,Franzen O, Lindblom U, Ottoson D,Eds. New York, Plenum Press, 1984, p.261–72

198. Yarnitsky D, Sprecher E, Zaslansky R,Hemli JA: Heat pain thresholds: norma-tive data and repeatability. Pain 60:329–332, 1995

199. Sosenko JM, Kato M, Soto R, Bild DE:Comparison of quantitative sensory-threshold measures for their associationwith foot ulceration in diabetic patients.Diabetes Care 13:1057–1061, 1990

200. Navarro X, Kennedy WR: Evaluation ofthermal and pain sensitivity in type 1diabetic patients. J Neurol Neurosurg Psy-chiat 54:60–64, 1991

201. Valk GD, Grootenhuis PA, van Eijk JT,Bouter LM, Bertelsmann FW: Methodsfor assessing diabetic polyneuropathy:validity and reproducibility of the mea-surement of sensory symptom severityand nerve function tests. Diabetes ResClin Pract 47:87–95, 2000

202. Dyck PJ, Zimmerman IR, Johnson DM,Gillen d, Hokanson JL, Kar JL, GruenerG, O’Brien PC: A standard test of heat-pain responses using CASE IV. J NeurolSci 136:54–63, 1996

203. Bril V: Electrophysiologic testing. InTextbook of Diabetic Neuropathy. GriesFA, Cameron NE, Low PA, Ziegler D,Eds. Stuttgart, Thieme, 2003, p. 177–184

204. Arezzo JC: The use of electrophysiologyfor the assessment of diabetic neuropa-thy. Neurosci Res Comm 21:13–22, 1997

205. Arezzo JC, Zotova E: Electrophysiologicmeasures of diabetic neuropathy: mech-anism and meaning. International RevNeurobiol 50:229–255, 2002

206. Hohman TC, Cotter, MA, Cameron NE:ATP-sensitive K(�) channel effects onnerve function, Na(�), K(�) ATPase,and glutathione in diabetic rats. EurJ Pharmacol 397:335–341, 2000

207. Mackel R, Brink E: Conduction of neuralimpulses in diabetic neuropathy. ClinNeurophysiol 114:248–255, 2003

208. Perkins BA, Olaleye D, Bril V: Carpaltunnel syndrome in patients with dia-betic polyneuroapthy. Diabetes Care 25:565–569, 2002

209. Sharma KR, Cross J, Farronay O, AyyarDR, Shebert RT, Bradley WG: Demyeli-nating neuropathy in diabetes mellitus.Arch Neurol 59:758–765, 2002

210. Waxman SG: Voltage-gated ion chan-nels in axons: localization, function, anddevelopment. In The Axon: Structure,

Function and Pathophysiology. WaxmanSG, Kocsis JD, Stys PK, Eds. New York,Oxford Press, 1995, p. 218–243

211. Herrmann DN, Ferguson ML, LogigianEL: Conduction slowing in diabetic dis-tal polyneuropathy. Muscle Nerve 26:232–237, 2002

212. Fernyhough P, Gallagher A, Averill SA:Aberrant neurofilament phosphoryla-tion in sensory neurons of rats with dia-betic neuropathy. Diabetes 48:881–889,1999

213. Padua L, Saponara C, Ghirlanda R,Padua R, Aprile I, Caliandro P, Tonali P:Lower limb nerve impairment in dia-betic patients: multiperspective assess-ment. Eur J Neurol 9:69–73, 2002

213. Tkac I, Bril V: Glycemic control is relatedto the electrophysiologic severity of dia-betic peripheral sensorimotor polyneu-ropathy. Diabetes Care 21:1749–1752,1998

215. Amthor KF, Dahl-Jorgensen K, Berg TJ,Heier MS, Sandvik L, Aagenaes O, Hans-sen KF: The effect of 8 years of strictglycaemia control on peripheral nervefunction in IDDM patients: the OsloStudy. Diabetologia 37:579–784, 1994

216. Muller-Felber W, Landgraf R, Scheuer R,Wagner S, Reimers CD, Nusser J, Abend-rotho D, Illner WD, Land W: Diabeticneuropathy 3 years after successful pan-creas and kidney transplantation. Diabe-tes 42:1482–1486, 1993

217. Veves A, Malik RA, Lye, RH, Masson EA,Sharma AK, Schady W, Boulton AJM:The relationship between sural nervemorphometric findings and measures ofperipheral nerve function in mild dia-betic neuropathy. Diabet Med 8:917–921, 1991

218. Russell JW, Karnes JL, Dyck PJ: Suralnerve myelinated fiber density differ-ences associated with meaningfulchanges in clinical and electrophysi-ologic measurements. J Neurol Sci 135:114–117, 1996

219. Kohara N, Kimura J, Kaji R, Goto Y, IshiiJ, Takiguchi M, Nakai M: F-wave latencyserves as the most reproducible measurein nerve conduction studies of diabeticpolyneuropathy: multicentre analysis inhealthy subjects and patients with dia-betic polyneuropathy. Diabetologia 43:915–921, 2000

220. Wells MD, Gozan SN: A method to im-prove the estimation of conduction ve-locity distribution over a short segmentof nerve. IEEE Trans Biomed Eng 46:1107–1120, 1999

221. Caccia MR, Salvaggi A, Dezuanni E: Anelectrophysiological method to assessthe distribution of the sensory propaga-tion velocity of the digital nerve innormal and diabetic subjects. Electroen-cephal Clin Neurophysiol 89:88–94, 1993



222. Bertora P, Valla P, Dezuanni E: Preva-lence of subclinical neuropathy in dia-betic patients: assessment by study ofconduction velocity distribution withinmotor and sensory nerve fibres. J Neurol245:81–86, 1998

223. Burke D, Kiernan MC, Bostock H: Excit-ability of human axons. Clin Neuro-physiol 112:1575–1585, 2001

224. Kiernan MC, Lin CS-Y, Andersen KV,Murray NM, Bostock H: Clinical evalua-tion of excitability measures in sensorynerve. Muscle Nerve 24:883–892, 2001

225. Yang Q, Kaji R, Takagi T, Kohara N, Mu-rase N, Yamada Y, Seino Y, Bostock H:Abnormal axonal inward rectifier instreptozocin-induced experimental dia-betic neuropathy. Brain 124:1149–1155, 2001

226. Sima AAF, Brown MB, Prashar A,Chakrabati S, Laudadio C, Greene DA:The reproducibility and sensitivity ofsural nerve morphometry in the assess-ment of diabetic peripheral neuropathy.Diabetologia 35:560–569, 1992

227. Sima AAF: Diabetic neuropathy: the util-ity of nerve biopsy. ElectroencephalogClin Neurophysiol Suppl 50:525–533,1999

228. Thomas PK. Nerve biopsy. Diabet Med16:351–352, 1997

229. Dahlin LB, Erikson KF, Sundkvist G:Persistent postoperative complaints afterwhole nerve sural nerve giopsies in dia-betic and non-diabetic subjects. DiabetMed 14:353–356, 1997

230. Theriault M, Dort J, Sutherland G, Zo-chodne DW: A prospective qualitativestudy of sensory deficits after wholenerve sural nerve biopsies in diabeticand non-diabetic patients: surgical ap-proach and the role of collateral sprout-ing. Neurology 50:480–484, 1998

231. Nukada H: Drug trials: to be biopsied ornot? In Recent Advances in Clinical Neu-rophysiology. Kimura J, Shibasaki H, Eds.Amsterdam, Elsevier Sciences, 1996, p.794–797

232. Thomas PK: The assessment of diabeticpolyneuropathy for drug trials. In RecentAdvances in Clinical Neurophysiology.Kimura J, Shibasaki H, Eds. Amsterdam,Elsevier Sciences, 1996, p. 787–793

233. Perkins BA, Greene DA, Bril V: Glycemiccontrol is related to the morphologicalseverity of diabetic sensorimotor poly-neuropathy. Diabetes Care 24:748–752,2001

234. Newrick PG, Wilson AJ, Jakubowski JJ,Boulton AJM, Ward JD: Sural nerve ox-ygen tension in diabetes. Br Med J 293:1053–1054, 1986

235. Tesfaye S, Malik RA, Harris N, Jaku-bowski JJ, Modie C, Rennie IG, Ward JD:Arterio-venous shunting and proliferat-ing new vessels in acute painful neurop-

athy of rapid glycaemic control (insulinneuritis). Diabetologia 39:329–335,1996

236. Ibrahim S, Harris ND, Radatz M, SelmiF, Raj-bhandari S, Brady L, JakubowskiJJ, Ward JD: A new minimally-invasivetechnique to show nerve ischaemia indiabetic neuropathy. Diabetologia 42:737–742, 1999

237. Hirai A, Yasuda H, Joko M, Maeda T,Kikkawa R: Evaluation of diabetic neu-ropathy through the quantitation of cu-taneous nerves. J Neurolog Sci 172:55–62, 2000

238. Polydefkis M, Hauer P, Griffin JW,McArthur JC: Skin biopsy as a tool toassess distal small fiber innervation indiabetic neuropathy. Diabet Technol Ther3:23–28, 2001

239. Yaneda H, Tereda M, Maeda K, Koga-was, Sanada M, Haneda M, Kashiwagi A,Kikkawa R: Diabetic neuropathy andnerve regeneration. Prog Neurobiol 69:229–285, 2003

240. Eaton SE, Harris ND, Rajbhandan SM,Greenwood P, Wilkinson ID, Ward JD,Griffiths PD, Tesfaye S: Spinal-cord in-volvement in diabetic peripheral neu-ropathy. Lancet 358:35–36, 2001

241. Boulton AJM, Knight G, Drury J, WardJD: The prevalence of symptomatic, di-abetic neuropathy in an insulin-treatedpopulation. Diabetes Care 8:125–128,1985

242. Sorensen L, Molyneaux L, Yue DK:Insensate versus painful diabetic neu-ropathy: the effects of height, gender,ethnicity and glycaemic control. Diabe-tes Res Clin Pract 57:45–51, 2002

243. Benbow SJ, Chan AW, Bowsher D, Mac-Farlane IA, Williams G: A prospectivestudy of painful symptoms, small-fibrefunction and peripheral vascular diseasein chronic painful diabetic neuropathy.Diabet Med 11:17–21, 1994

244. Boulton AJM, Scarpello JHB, ArmstrongWD, Ward JD: The natural history ofpainful diabetic neuropathy: a 4-yearstudy. Postgrad Med J 59:556–559, 1983

245. Young RJ, Ewing DJ, Clarke BF: Chronicand remitting painful diabetic neuropa-thy. Diabetes Care 11:34–40, 1988

246. Cheng WY, Jian YD, Chuang LM, HuangCN, Heng LT, Wu HP, Tai TY, Lin BJ:Quantitative sensory testing and risk fac-tors of diabetic sensory neuropathy.J Neurol 246:394–398, 1999

247. Sosenko JM, Gadia MT, Fournier AM,O’Connell MT, Aguiar MC, Skyler JS:Body stature as a risk factor for diabeticsensory neuropathy. Am J Med 80:1031–1034, 1986

248. Wiles PG, Pearce SM, Rice PJS, MitchellJMO: Vibration perception threshold:influence of age, height, sex, and smok-ing, and calculation of accurate percen-

tile values. Diabet Med 8:157–161, 1991249. Adler AI, Boyko EJ, Ahroni JH, Stensel

V, Forsberg RC, Smith DG: Risk factorsfor diabetic peripheral sensory neuropa-thy. Diabetes Care 20:1162–1167, 1997

250. Klein R, Klein BEK, Moss SE: Relation ofglycemic control to diabetic microvascu-lar complications in diabetes mellitus.Ann Intern Med 124:90–96, 1996

251. Sosenko JM, Boulton AJM, Kubrusly DB,Jagdish K, Weintraub BA, Skyler JS: Thevibratory perception threshold in youngdiabetic patients: associations with gly-cemia and puberty. Diabetes Care 8:605–607, 1985

252. Sosenko JM, Kato M, Goldberg RB: Sen-sory function and albumin excretion ac-cording to diagnostic criteria for diabetes.Diabetes Care. In press

253. Ziegler D, Mayer P, Muhlen H, Gries FA:The natural history of somatosensoryand autonomic nerve dysfunction in re-lation to glycaemic control during thefirst 5 years after diagnosis of type 1 (in-sulin dependent) diabetes mellitus. Dia-betologia 34:822–829, 1991

254. Sosenko JM, Kato M, Soto R, GoldbergRB: Sensory function at diagnosis and inearly stages of NIDDM in patients de-tected through screening. Diabetes Care15:847–852, 1992

255. Sosenko JM, Sparling YH, Hu, D, WeltyT, Howard BV, Lee E, Robbins DC: Useof the Semmes-Weinstein monofilamentin the Strong Heart Study. Diabetes Care20:1715–1720, 1999

256. Laudadio C, Sima AAF, The PonalrestatStudy Group: Progression rates of dia-betic neuropathy in placebo patients inan 18-month clinical trial. J DiabetesComp 12:121–127, 1998

257. Sosenko JM, Kato M, Soto R, Bild DE: Aprospective study of sensory function inpatients with type 2 diabetes. Diabet Med10:110–114, 1993

258. Pirart J: Diabetes mellitus and its degen-erative complications: a prospectivestudy of 4,400 patients observed be-tween 1947 and 1973. Diabetes Care1:168–188, 1978

259. Maser RE, Steenkiste AR, Dorman JS,Nielsen VK, Bass EB, Manjoo Q, DrashAL, Becker DJ, Kuller LH, Greene DA,Orchard TJ: Epidemiological correlatesof diabetic neuropathy: report fromPittsburgh Epidemiology of DiabetesComplications Study. Diabetes 38:1456–1461, 1989

260. Forrest KYZ, Maser RE, Pambianco G,Becker DJ, Orchard TJ: Hypertension asa risk factor for diabetic neuropathy. Di-abetes 46:665–670, 1997

261. Franklin GM, Shetterly SM, Cohen JA,Baxter J, Hamman RF: Risk factors for



distal symmetric neuropathy in NIDDM.Diabetes Care 17:1172–1177, 1994

262. The DCCT Research Group: Factors indevelopment of diabetic neuropathy. Di-abetes 37:476–481, 1988

263. Mitchell BD, Hawthorne VM, Vinik AI:Cigarette smoking and neuropathy in di-abetic patients. Diabetes Care 13:434–437, 1990

264. Reichard P, Nilsson BY, Rosenqvist CL:The effect of long-term intensified insu-lin treatment on the development of mi-crovascular complications of diabetes.N Engl J Med 329:304–309, 1993

265. Reichard P, Pihl M, Rosenqvist U, Sule J:Complications of IDDM are caused byelevated blood glucose levels, the Stock-holm Diabetes Intervention study at 10year follow-up. Diabetologia 39:1383–1488, 1996

266. Amthor KF, Dahl-Jorgensen K, Berg TJ,Sandvik L, Hanssen KF: The effect of 8years of strict glycaemic control on pe-ripheral nerve function in IDDM pa-tients: the Oslo study. Diabetologia 37:579–586, 1994

267. Ohkubo Y, Kishikawa H, Araki E, MiyataT, Isami S, Motoyoshi S, Kojima Y, Fu-ruyoshi N, Shichiri M: Intensive insulintherapy prevents the progression of dia-betic microvascular complications inJapanese patients with non-insulin-de-pendent diabetes: a randomized pro-spective 6-year study. Diabetes Res ClinPract 28:103–117, 1995

268. Kennedy WR, Navarro X, Goetz FC,Sutherland DE, Najarian JS: Effects ofpancreas transplantation on diabeticneuropathy. N Engl J Med 322:1031–1037, 1990

269. Hotta N, Sakamoto N, Shigeta Y,Kikkawa R, Goto Y: Clinical investiga-tion of Epalrestat, an aldose reductaseinhibitor, on diabetic neuropathy in Ja-pan. J Diabetes Compl 10:168–172, 1996

270. Handelsman DJ, Turtle JR: Clinical trialof an aldose reductase inhibitor in dia-betic neuropathy. Diabetes 30:459–464,1981

271. Boulton AJM, Levin S, Comstock J: Amulticenter trial of the aldose reductaseinhibitor Tolrestat in patients withsymptomatic diabetic neuropathy. Dia-betologia 33:431–437, 1990

272. Krentz AJ, Honigsberger L, Ellis SH,Hardman M, Nattrass M: A 12-monthrandomized controlled study of the al-dose reductase inhibitor ponalrestat inpatients with chronic symptomatic dia-betic neuropathy. Diabet Med 9:463–468, 1992

273. Horrobin DF: Essential fatty acids in themanagement of impaired nerve functionin diabetes. Diabetes 46 (Suppl. 2):S90–S93, 1997

274. �-Linolenic Acid Multicenter Trial

Group: Treatment of diabetic neuropa-thy with �-linolenic acid. Diabetes Care16:8–15, 1993

275. Hounsom L, Horrobin DF, TritschlerH, Corder R, Tomlinson DR: A lipoicacid-gamma linolenic acid conjugate iseffective against multiple indices of ex-perimental diabetic neuropathy. Diabe-tologia 41:839–843, 1998

276. Apfel SC: Nerve growth factor for thetreatment of diabetic neuropathy: whatwent wrong, what went right, and whatdoes the future hold? Int Rev Neurobiol50:393–413, 2002

277. Brownlee M: Biochemistry and molecu-lar cell biology of diabetic complica-tions. Nature 414:813–820, 2001

278. Vinik A, Tesfaye S, Zhang D, Bastyr E:LY333531 treatment improves diabeticperipheral neuropathy with symptoms(Abstract). Diabetes 51 (Suppl. 2):A79,2002

279. Yuen KC, Baker NR, Rayman G: Treat-ment of chronic painful diabetic neurop-athy with isosorbide dinitrate spray: adouble-blind placebo-controlled cross-over study. Diabetes Care 25:1699–1703, 2002

280. Max MB, Lynch SA, Muir J, Shaof SE,Smoller B, Dubner R: Effects of desipra-mine, amitriptyline and fluoxetine onpain relief in diabetic in diabetic neurop-athy. N Engl J Med 326:1250–1256,1992

281. McQuay H, Tramer M, Nye BA: A sys-tematic review of antidepressants inneuropathic pain. Pain 68:217–227,1996

282. Young RJ, Clarke BF: Pain relief in dia-betic neuropathy: the effectiveness ofimipramine and related drugs. DiabetMed 2:363–366, 1984

283. Kumar D, Alvaro Ms, Julka IS: Diabeticperipheral neuropathy: effectiveness ofelectrotherapy and amitriptyline forsymptomatic relief. Diabetes Care 21:1322–1325, 1998

284. Sindrup SH, Gram LF, Brosen K, EshojO, Mogensen BI: The SSRI paroxetine iseffective in the treatment of diabeticneuropathy symptoms. Pain 42:135–144, 1990

285. Sindrup SH, Bjerre U, Dejgaard A,Brosen K, Aaes-Jorgensen T, Gam LF:The selective serotonin reuptake inhibi-tor citalopram relieves the symptoms ofdiabetic neuropathy. Clin PharmacolTher 53:547–552, 1992

286. Rodriguez LA, Montero D: Associationbetween SSRIs and upper gastrointesti-nal bleeding: a population based case-control study. Br Med J 319:1106–1109,1999

287. McQuay H, Carroll D, Jadad AR, WiffenP, Moore A: Anticonvulsant drugs for

the management of pain: a systematic re-view. BMJ 311:1047–1052, 1995

288. Jensen TS: Anticonvulsants in neuro-pathic pain: rationale and clinical evi-dence. Eur J Pain 6 (Suppl. A):61–68,2002

289. Boulton AJM: Current and emergingtreatments for the diabetic neuropathies.Diabetes Rev 7:279–386, 1999

290. Backonja M, Beydoun A, Edwards KR,Schwartz SL, Fonseca V, Hes M, LaMo-reaux L, Garofaldo E: Gabapentin for thesymptomatic treatment of painful neu-ropathy in patients with diabetes melli-tus: a randomized controlled trial. JAMA280:1831–1836, 1998

291. Backonja M, Glanzman RL: Gabapentindosing for neuropathic pain: evidencefrom randomized placebo controlledclinical trials. Clin Ther 25:81–104,2003

292. Eisenberg E, Luri Y, Braker C, Daoud D,Ishay A: Lamotrigine reduces painful di-abetic neuropathy: a randomized, con-trolled study. Neurology 57:505–509,2001

293. Dejgard A, Petersen P, Kastrup J: Mexili-tene for the treatment of chronic painfuldiabetic neuropathy. Lancet 1:9–11,1988

294. Jarvis B, Coukell AJ: Mexilitene: a reviewof its therapeutic use in painful diabeticneuropathy. Drugs 56:691–707, 1998

295. Harati Y, Gooch C, Swenson M, Edel-man S, Greene D, Raskin P, Donofrio P,Cornblath D, Sachdeo R, Siu CO, KaminM: Double-blind randomized trial of tra-madol for the treatment of the pain ofdiabetic neuropathy. Neurology 50:1841–1846, 1998

296. Harati Y, Gooch C, Swenson M, Edel-man SV, Greene D, Raskin P, Donofrio P,Cornblath D, Olson WH, Kamin M:Maintenance of the long-term effective-ness of Tramadol in treatment of thepain of diabetic neuropathy. J DiabetesCompl 14:65–70, 2000

297. Gimbel JS, Richards P, Portenoy RK:Controlled-release oxycodone for painin diabetic neuropathy: a randomizedcontrolled trial. Neurology 60:927–934,2003

298. Watson CPN, Moulin D, Watt-Watson J,Gordon A, Eisenhoffer J: Controlled-re-lease oxycodone relieves neuropathicpain: a randomized controlled trial inpainful diabetic neuropathy. Pain 105:71–78, 2003

299. Zhang WY, Wan Po AL: The effective-ness of topically applied capsaicin: ameta-analysis. Eur J Clin Pharmacol 45:517–522, 1994

300. Abusaisha BB, Constanzi JB, BoultonAJM: Acupuncture for the treatment of



chronic painful diabetic neuropathy: along-term study. Diabetes Res Clin Pract39:115–121, 1998

301. Hamza MA, White PF, Craig WF, Gho-name ES, Ahmed HE, Proctor TJ, NoeCE, Gajraj N: Percutaneous electricalnerve stimulation: a novel analgesictherapy for diabetic neuropathic pain.Diabetes Care 23:365–370, 2000

302. Weintraub MI, Wolfe GI, Barohn RA,Cole SP, Parry GJ, Hayat G, Cohen JA,Page JC, Bromberg MB, Schwartz SL:Static magnetic field therapy for symp-tomatic diabetic neuropathy; a random-ized, double-blind, placebo-controlledtrial. Arch Phys Med Rehabil 86:736–746, 2003

303. Tesfaye S, Watt J, Benbow SJ, Pang KA,Miles J, MacFarlane IA: Electrical spinal-

cord stimulation for painful diabetic pe-ripheral neuropathy. Lancet 348:1696–1701, 1996

304. Boulton AJM: The pathway to ulcer-ation. In The Foot in Diabetes. BoultonAJM, Connor H, Cavanagh PR, Eds.Chichester, John Wiley, 2000, p.19 –31

305. Litzelman DK, Marriott DJ, Vinicor F:Independent physiological predictorsof foot lesions in patients withNIDDM. Diabetes Care 20:1273–1278,1997

306. Reiber GE, Vileikyte L, Boyko EJ, DelAguila M, Smith DG, Lavery LA, BoultonAJM: Causal pathways for incident low-er-extremity ulcers in patients with dia-betes from two settings. Diabetes Care22:157–162, 1999

307. Valk GD, Kriegsman DM, Assendelft WJ:Patient education for preventing dia-betic foot ulceration: a systematic re-view. Endocrinol Metab Clin N Am31:633–658, 2002

308. McCabe CJ, Stevenson RC, Dolan AM:Evaluation of a diabetic foot screeningand prevention programme. Diabet Med15:80–84, 1998

309. Cavanagh PR, Young MJ, Adams JE,Vickers KL, Boulton AJM: Radiographicabnormalities in the feet of patients withdiabetic neuropathy. Diabetes Care 17:201–209, 1994

310. Sanders LJ, Frykberg RG: Charcot neu-roarthropathy of the foot. In Levin andO’Neal’s The Diabetic Foot. 6th ed.Bowker JH, Pfeifer MA, Eds. St. Louis,CV Mosby, 2001, p. 439–466



Documents

Diabetic Somatic Neuropathies - Diabetes Carecare.diabetesjournals.org/content/diacare/27/6/1458.full.pdfAcute sensory neuropathy is a distinct variety of the symmetrical polyneuropa-thies