EPIDEMIOLOGY AND ETIOLOGY OF
DIABETIC PERIPHERAL NEUROPATHIES*
Solomon Tesfaye, MD, FRCP†
ABSTRACT of reflexes and sensory responses to vibration,
light touch, pinprick, and the 10-g monofilament.
Although diabetic peripheral neuropathy Other tests to measure various thresholds are
(DPN) affects up to 50% of patients with diabetes, available as well.
it has in some respects been considered the Several drugs and nonpharmacologic modal-
“Cinderella” of microvascular complications of ities are available to treat painful DPN, but their
diabetes, receiving less attention than it deserves efficacy and tolerability vary. Recent evidence of
with regard to screening, early detection, treat- central nervous system involvement in DPN may
ment, and prevention. open new avenues of investigation for the devel-
The clinical manifestations of DPN are pain opment of rational therapies.
(often experienced as burning, tingling, or allody- The promise of the aldose reductase inhibitors
nia) and insensitivity, which dramatically increases in the treatment of DPN has not been fully real-
the risk for burns, injuries, and foot ulceration. ized. Newer aldose reductase inhibitors are
Risk factors for DPN include poor glycemic undergoing clinical trials. Other interventions,
control, duration of diabetes, hyperlipidemia, such as α-lipoic acid and ruboxistaurin, show
elevated albumin excretion rate, and increased promise in reducing pain. In addition, according
body mass index. DPN also is associated with to study results, ruboxistaurin appears to have
cardiovascular disease and mortality and is sec- beneficial effects on nerve impairment and pain
ond only to macroangiopathy as an independent symptoms in patients with DPN.
risk factor for mortality. (Adv Stud Med. 2004;4(10G):S1014-S1021)
The pathogenesis of DPN involves the progres-
sion of diabetes and chronic hyperglycemia to
endothelial dysfunction via increased activity in 4
possible pathways: the protein kinase C pathway
via diacylglycerol, advanced glycation end-product
formation, the polyol pathway, and oxidative stress. iabetic peripheral neuropathy (DPN), a
Endothelial dysfunction, influenced by cardiovascu- microvascular complication of diabetes, is
lar and genetic factors, progresses to microan- associated with considerable morbidity,
giopathy and then to nerve hypoxia. mortality, and diminished quality of life.
In clinical practice, detection of DPN begins Characterized by pain, paresthesia, and sensory loss, it
with a careful history of sensory and motor symp- affects up to 50% of patients with diabetes,1 with new
toms, an inspection of the feet, and an evaluation cases occurring at an annual incidence of about 2%.2
In absolute numbers, against the estimated global
prevalence of 220 million cases of diabetes by 2010,1
*Based on a presentation given by Dr Tesfaye at a sym- DPN is likely to affect as many as 110 million persons
posium held in conjunction with the 12th International
Congress of Endocrinology. worldwide—and at tremendous cost. In the United
†Consultant Physician and Diabetologist, Sheffield States alone, the total cost associated with DPN is
Teaching Hospitals, Sheffield, United Kingdom. $10.9 billion a year.3
Address correspondence to: Solomon Tesfaye, MD,
Royal Hallamshire Hospital, Glossop Road, Sheffield, S10
Diabetic peripheral neuropathy is the main predis-
2 JF, UK. E-mail: Solomon.Tesfaye@sth.nhs.uk. posing factor for foot ulceration and infection. Foot
S1014 Vol. 4 (10G) I December 2004
ulceration alone occurs in 15% of patients with dia- completely numb feet, putting them at extremely high
betes during their lifetime4 and is the most common risk for foot ulceration.
cause of hospital bed occupancy in the United States Pain and insensitivity are the 2 clinical conse-
and abroad. With an annual incidence that ranges quences of DPN. Pain symptoms, including burning,
from 1% to 3%,5,6 diabetic foot ulceration accounts for paresthesia (“pins and needles”), hyperesthesia, and
a 15-fold increase in the likelihood of lower limb allodynia (contact pain), can be extremely distressing
amputation in patients with diabetes versus nondia- and are typically worse at night.10 Pain can range from
betic people,7 and a 2-fold increase in mortality.8 Of tingling in one or more toes to severe and persistent
those patients with diabetes who have had a lower neuropathic pain. Patients commonly describe their
limb amputated, more than 50% required amputation symptoms as “sharp electric shocks that shoot up my
of the contralateral limb within 1 year.4 legs” or “stepping on broken glass.” Others complain
From these epidemiologic data, DPN and diabetic of disrupted sleep and pacing the floor during the
foot ulceration clearly are far from rare, far from night to distract themselves from the pain. Still, others
benign, and pose a major healthcare challenge to the voice frustration and depression because of the pain.11
medical profession and to society. However, in some Some patients, in fact, have committed suicide because
respects, DPN is the “Cinderella” of diabetic microvas- of intractable and persistent neuropathic pain.
cular complications, receiving less attention than it Insensitivity, or loss of pain, can lead to foot ulcer-
deserves in terms of screening, early detection, treat- ation and a host of unintentional, but serious injuries.
ment, and prevention. New and emerging treatments Patients who have lost sensation in their hands cannot
for DPN, in addition to early detection and better detect temperature and often burn themselves while
control of hyperglycemia and other risk factors, soon cooking or ironing, for example, and also have diffi-
may bring more attention to DPN. culty handling small objects. Those patients who have
lost sensation in their feet often sustain puncture
CLINICAL FEATURES OF wounds, friction wounds, and burns that can become
DIABETIC PERIPHERAL NEUROPATHY infected and/or ulcerated and lead to amputation.
However, with appropriate foot care, at least 50% of
Diabetic peripheral neuropathy, which is also ulcerations can be prevented.10
referred to as distal symmetrical polyneuropathy, is the
most common neuropathic syndrome seen in patients RISK FACTORS FOR DIABETIC PERIPHERAL NEUROPATHY
with diabetes. Less common neuropathic syndromes
include cranial mononeuropathies and focal neu- Studies in patients with type 1 or type 2 diabetes
ropathies such as proximal motor neuropathy. mellitus have shown that poor glycemic control is one
Diabetic peripheral neuropathy starts in the toes risk factor for DPN.
and gradually moves upward. Once it is well estab- The EURODIAB IDDM Complications Study,
lished in the lower limbs, DPN affects the upper which involved 3250 patients with type 1 diabetes
limbs, with sensory loss following the typical “glove mellitus from 31 centers in 16 European countries,
and stocking” pattern of distribution.9 Significant found that DPN was related to both glycemic control
motor deficits are not common in the early stages of and duration of disease.12 Although the 30% baseline
DPN,9 although magnetic resonance imaging (MRI) prevalence of DPN was significantly related to glyco-
examination of the lower limbs reveals atrophy of sylated hemoglobin (HbA1c; P <.001), the prevalence
the small muscles of the foot as an early feature. varied from 17% to 41% after data were adjusted for
Symptomatic muscle weakness, however, tends to duration of diabetes mellitus, with lower HbA1c levels
develop later in the disease course. associated with lower prevalence rates and higher lev-
Painful symptoms such as burning, tingling, and els associated with higher prevalence rates. However,
paresthesias are present early on in 30% of patients.9 even those with good glycemic control (HbA1c <5.4%,
Importantly, symptoms are not a reliable indicator of equivalent to HbA1c of 7% in the Diabetes Control
the severity of nerve damage.9 Some patients with and Complications Trial) still developed microvascular
severe pain symptoms have little sensory deficit, disease, suggesting that factors other than glycemic
whereas other patients with no painful symptoms have control and disease duration are involved.
Advanced Studies in Medicine I S1015
Follow-up data from the EURODIAB cohort of (PKC) activation. Increased activity in these path-
patients with type 1 diabetes mellitus revealed that tra- ways leads to endothelial dysfunction, which in turn
ditional markers of macrovascular disease such as cho- leads to microangiopathy and then to nerve hypoxia.
lesterol and fasting triglyceride levels, albumin The latter results in structural damage to the nerve
excretion rates, von Willebrand factor levels, and body and irreversible neuropathy or reduced nerve con-
mass index also were associated with the development duction velocity.15
of DPN. After excluding all patients with DPN at
Cardiovascular risk factors such as hyperlipidemia,
baseline and adjusting the data for HbA1c and dura- hypertension, smoking, and increased body mass
tion of diabetes mellitus, the investigators found that index are major and independent contributors to
values for all of these factors were significantly elevat- endothelial dysfunction. Genetic factors may play a
ed in all of the 276 patients who developed DPN after role as well, which may explain why some diabetic
7 years of follow-up compared with the values for the patients with good glycemic control develop microvas-
896 patients who did not develop DPN during the fol- cular complications while others with poor glycemic
low-up period. control do not. Similarly, cardiovascular risk factors
There is also evidence that DPN, a microvascular contribute to microangiopathy, which is also influ-
complication of diabetes mellitus, is associated with enced by coagulation and hematologic factors.15
cardiovascular disease and mortality. In a study of 132 Some of the vascular and metabolic interactions that
patients with type 2 diabetes mellitus, 38 died during occur in the pathogenesis of DPN are described below to
the 9-year follow-up period.14 Macroangiopathy was help define the process further and to outline the ratio-
found to be the strongest independent risk factor for nale for new and emerging therapeutic interventions.
mortality, followed in descending order
by DPN, albumin excretion rate, and
HbA1c. In addition, postmortem stud-
ies conducted at King’s College
Hospital in London have demonstrated Figure 1. Pathogenesis of Diabetic Peripheral Neuropathy
that patients with DPN die of coronary
artery disease (Personal communica-
tion, Michael Edmonds, MD, March
PATHOGENESIS OF DIABETIC PERIPHERAL
↑ AGE - RAGE ↑ Polyol ↑ Free radical
↑ PKC activity
pathway activity formation
Until recently, there were 2 schools
of thought regarding the etiology and NO quenching
↓ NO generation
pathogenesis of DPN: metabolic versus
vascular. Recent studies, however, have Genetics Endothelial dysfunction DM ↑ Lipids, ↑ BP,
shown that vascular factors and meta- ↑ BMI, Smoking
bolic interactions are involved at all
Rigid RBC Micro-
stages of DPN.15 ↑ Coagulability angiopathy DM ↑ Lipids, ↑ BP,
Current thinking on the pathogene- ↑ Platelet reactivity ↑ BMI
sis of DPN can be summarized by the
schematic shown in Figure 1. Diabetes
leads to chronic hyperglycemia, which
Structural damage ↓ Nerve conduction
appears to stimulate 4 pathways directly Irreversible neuropathy velocity
or indirectly: advanced glycation end-
product (AGE-RAGE interaction) for-
AGE = advanced glycation end-products; BP = blood pressure; BMI = body mass index; DAG = dia-
mation, polyol pathway hyperactivity, cylglycerol; DM = diabetes mellitus; HbA1c = glycosylated hemoglobin; NO = nitric oxide; PKC = pro-
oxidative stress, and protein kinase C tein kinase C; RAGE = receptors for advanced glycation end-products; RBC = red blood cells.
S1016 Vol. 4 (10G) I December 2004
In 1966, Gabbay et al proposed that sorbitol accu- blood vessels on the nerve surface—similar to the net-
mulation was a cause of DPN.16 In the polyol pathway, work of new blood vessels seen on the surface of the
glucose is converted to sorbitol by the enzyme aldose retina in patients with diabetes.
reductase, and sorbitol to fructose by the enzyme sor- A recent study examining epineurial blood flow in
bitol dehydrogenase. The key enzyme is aldose reduc- patients with painful and painless DPN found that
tase. Later studies showed that aldose reductase fluorescein levels rose more rapidly and epineurial
inhibitors were able to increase nerve conduction intravascular oxygen saturation was significantly high-
velocity in animals,17 and to a lesser extent in er in patients with painful DPN compared to patients
humans,18 and that these agents produced some with painless DPN.26 These findings may reflect
increase in nerve fiber counts in patients with shunting from the endoneurium to the epineurium in
DPN.19,20 However, there was no unequivocal improve- patients with painful DPN and also support the
ment in terms of lessening signs and symptoms. Newer premise that hemodynamic factors play a role in the
aldose reductase inhibitors are currently undergoing pathogenesis of neuropathic pain.
Nerve fiber loss is the cause of insensitivity in
DPN. As revealed by fascicular biopsy of the sural
nerve, nerve fibers in patients with diabetes but no Figure 2. Nerve Fiber Loss in Diabetes and
DPN are more numerous than in those patients with Diabetes with Diabetic Peripheral Neuropathy
diabetes with DPN (Figure 2).21 Sural nerve biopsies
also reveal microvascular defects in the endoneurial
vessels, such as gross basement membrane thickening,
endothelial cell proliferation, and hypertrophy (Figure
3),15 in addition to reduced oxygen tension22 in
patients with DPN, as compared with patients who
have diabetes but do not have DPN. Similarly, pho-
tography of surgically isolated sural nerve reveals
microvascular abnormalities in the epineurial arteries Normal diabetic Neuropathy
and veins (Figure 4),23 whereas fluorescein angiography Biopsy of the sural nerve demonstrates that nerve fibers are less numerous
reveals arteriosclerosis on the surface of the nerve and in diabetic peripheral neuropathy.
Reproduced with permission from Malik et al. Diabetologia.
impaired blood flow23 in patients with DPN compared 1993;36:454-459.21
to patients with diabetes without DPN.
Impaired blood flow adversely affects nerve con-
duction velocity, as demonstrated in exercise studies.24
Exercise to 80% maximal heart rate will increase sural Figure 3. Microvascular Defects in Endoneurial
sensory nerve conduction velocity by about 5 m per Vessels in Diabetes and Diabetes with Diabetic
second in healthy people and 4 m per second in Peripheral Neuropathy
patients with diabetes without DPN. The same exer-
cise will not increase conduction velocity in those
patients with DPN because the neuropathic nerve that
is also arteriosclerotic is unable to increase blood flow
in response to exercise.
A more recent study of changes in sural nerve
blood vessels in acute diabetic neuropathies such as
acute insulin neuritis has shown that rapid improve-
ment in blood glucose control can produce severe Normal diabetic Diabetic neuropathy
painful neuropathic symptoms.25 Acute insulin neuri- Photomicrographs of capillaries from sural nerve biopsies show an open ves-
tis, once thought to be a metabolic condition, is, in sel in the normal diabetic nerve and gross basement membrane thickening,
endothelial cell proliferation, and a closed vessel in the neuropathic nerve.
fact, associated with proliferative changes that result in Reproduced with permission from Cameron et al. Diabetologia.
the development of a fine network of “neural new” 2001;44:1973-1988.15
Advanced Studies in Medicine I S1017
Figure 4. Microvascular Abnormalities in Epineurial because inappropriate or ill-fitting shoes are the most
Vessels in Diabetes and Diabetes with Diabetic common form of trauma to the diabetic foot.
Peripheral Neuropathy Some authorities advocate the additional use of a
clinical scoring system that grades the degree of neu-
ropathy on the basis of symptom, reflex, and sensory
scores, such as the Toronto Clinical Scoring System
that is weighted to emphasize sensory symptoms
(Table).28 The scoring system has been validated and
correlates well with electrophysiology findings and
The 10-g monofilament test is now standard for
examination of the diabetic foot, is inexpensive and
easy to use, and produces rapid and reproducible
results.29 The test also predicts foot ulceration; those
Photographs of surgically isolated sural nerve show increased arterial tortu-
osity in diabetic peripheral neuropathy. In the area where the artery joins
patients who cannot feel the monofilament are 15
the vein, the vein is distended because of increased pressure and also times more likely to develop ulceration within 3 years
becomes tortuous. than are those patients who have foot sensation.30
Reproduced with permission from Tesfaye et al. Diabetologia.
1993;36:1266-1274.23 However, there is some controversy regarding which
and how many areas of the foot should be tested. Some
groups suggest testing as many as 10 areas of the foot,
In studying the effect of revascularization on but the group in Toronto that developed the scoring sys-
nerve function in patients with DPN, Young et al tem advocates testing only the dorsum of the first toe in
examined groups of patients who had and had not both feet.31 Within the 0 to 8 reference range, a com-
undergone femoral popliteal bypass. 27 The bined score of 5 or above for both feet predicts DPN.
researchers found that both foot transcutaneous oxy- Another important test is the vibration percep-
gen saturation and peroneal nerve multiconduction tion threshold that detects subclinical neuropathy.
velocity increased significantly 6 weeks after surgery. It, too, is easy to use and predicts foot ulceration,
However, in the control group of patients, there was with 0 to 5 volts indicating low risk, 16 to 25 volts
no increase in either variable, demonstrating that indicating intermediate risk, and more than 25 volts
improving blood flow to the limbs improves nerve indicating high risk.32,33 And, as demonstrated by
EARLY DETECTION OF
DIABETIC PERIPHERAL NEUROPATHY
Table. Toronto Clinical Scoring System
Early detection of DPN is extremely important as
it may result in earlier treatment and in prevention of
Symptom Scores Reflex Scores Sensory Scores
further damage. In clinical practice, early detection
0–6 0–8 0–5
begins with a careful history and an evaluation of sen-
sory and motor symptoms, an assessment of disability Foot pain Knee reflexes Pinprick
resulting from the neuropathy, and the exclusion of Numbness Ankle reflexes Temperature
conditions other than diabetes mellitus that may be Tingling (both sides) Light touch
causing the neuropathy. Weakness Vibration
The clinical examination should include a careful Ataxia Position sense
inspection of the feet; evaluation of ankle and knee
Upper limb symptoms
reflexes; a sensory examination that includes testing
for vibration, light touch, and pinprick sensations; the Maximum score = 19.
0–6 = no neuropathy; 6–8 = mild neuropathy; 9–11 = moderate neuropa-
10-g monofilament test to assess sensation in the foot; thy; ≥12 = severe neuropathy.
and an assessment of footwear, which is important Modified from Bril V, Perkins BA. Diabetes Care. 2002;25:2048-2052.28
S1018 Vol. 4 (10G) I December 2004
Coppini et al, the vibration perception threshold low-up in one study, patients reported that the stimula-
also predicts mortality, with higher death rates in tor markedly reduced both background pain and peak
diabetic patients with neuropathy than in diabetic pain when it was turned on compared to when it was
patients without neuropathy.34 turned off.38 Over all, about 80% of patients responded
Other devices to measure the vibration perception to spinal cord stimulation; 20% did not respond, leading
threshold include the biothesiometer, the vibrameter, the investigators to explore whether abnormalities in the
and the neuroasthesiometer. Efforts to standardize the spinal cord itself could explain the lack of response.
equipment used for threshold measurement are cur- Using MRI examination, the investigators found
rently under way. that the cross-sectional area of the cord was much
More sophisticated equipment also is available to smaller in patients with DPN than in control subjects
assess DPN and to detect thresholds for vibration, cool- and slightly smaller in those with DPN compared to
ing, heat, and pain.35 However, to characterize DPN patients with diabetes who did not have DPN.39
most accurately, nerve conduction velocity tests and stag- Although there was no statistically significant differ-
ing the severity of neuropathy also are necessary. ence between the diabetic patients with DPN and
without DPN, the study findings demonstrated spinal
TREATMENT OF PAINFUL DIABETIC PERIPHERAL cord involvement in DPN. More recently, a large MRI
NEUROPATHY study found significant spinal cord atrophy not only in
patients with established DPN but also in those patients
The current approach to management of painful with subclinical DPN compared to patients with dia-
DPN centers on achieving and maintaining near-nor- betes who did not have DPN.40 This study provides fur-
mal glycemia (HbA1c) levels. However, many patients ther evidence of spinal cord involvement in DPN.
with diabetes, particularly those with type 2 diabetes More recent research provides evidence of neuronal
mellitus, find this difficult. dysfunction in the thalamus as a source of severe painful
Several drugs and nonpharmacologic modalities neuropathic symptoms (Unpublished observations).
are available for pain relief, but efficacy and tolerabili- Clearly, a deeper understanding of the origins of these
ty of adverse effects vary. For example, tricyclic antide- symptoms is needed to develop more effective therapies.
pressants are effective in many patients but are
associated with numerous adverse effects. Therefore, NEW INTERVENTIONS FOR
the starting dose, to be given at bedtime, should be DIABETIC PERIPHERAL NEUROPATHY
low and can then be increased gradually or discontin-
ued, depending on how effectively pain is relieved and Although aldose reductase inhibitors, which
whether the patient is tolerating the side effects. reduce activity in the polyol pathway, generated con-
Anticonvulsants such as gabapentin are also help- siderable optimism about the treatment of DPN
ful,36 as is the opiate derivative tramadol.37 Although when first introduced in the 1980s, they have not
there may be a risk of addiction with tramadol, it is a demonstrated consistent efficacy or safety. Two of the
useful drug to have in the armamentarium. Similarly, inhibitors have been withdrawn from the market
intravenous lidocaine and oral mexiletine, which are because of lack of efficacy, and 3 have been with-
usually used to control cardiac arrhythmias, are often drawn because of allergic reactions, hepatotoxicity, or
helpful in refractory patients. renal toxicity. Only one of this class of agents, epal-
Other drugs and nonpharmacologic approaches that restat, is still on the market (in Japan), although
are helpful in some patients include capsaicin cream, newer agents, including fidarestat, are being investi-
isosorbide dinitrate nasal spray, amantadine, acupunc- gated in ongoing clinical trials.
ture, and transcutaneous electrical nerve stimulation. Antioxidants have been used to treat DPN on the
However, treatment of painful DPN is less than premise that they reduce free radical formation. In the
satisfactory, with some patients having severe adverse SYDNEY trial, α-lipoic acid 600 mg per day signifi-
effects to therapy and others failing to respond at all to cantly lowered pain scores at 4 weeks compared to
any of the above therapies. intravenous riboflavin.41 Thus far, α-lipoic acid, which
A last resort for the treatment of refractory pain in is marketed in Germany, appears to be effective in
DPN is spinal cord stimulation. After 3 years of fol- reducing pain.
Advanced Studies in Medicine I S1019
Ruboxistaurin, an inhibitor of the β isoform of
Figure 5. Effect of Treatment with Ruboxistaurin on
PKC, is being studied in clinical trials involving
Neurologic Impairment Scores and Composite Score
patients with DPN and other microvascular compli-
cations of diabetes mellitus. In a phase II study of
205 patients with DPN and significant nerve
Entire patient population, n = 205
impairment, preliminary results showed that rubox-
istaurin 32 mg administered every day for 1 year P = .049 P = .033 P = .046
markedly improved the composite score of 3 mea-
sures of nerve function from baseline compared to a
placebo (Figure 5).42 The measures were the nerve
impairment score of the lower limbs (representing
the clinical examination), the nerve impairment 0
score for the reflex examination, and the nerve Pl acebo
impairment score of the lower limbs plus 4. 32 mg QD
A preliminary subgroup analysis of 83 patients NIS (LL) NIS reflex NIS (LL) + 4
with clinically significant pain symptoms at baseline
revealed marked improvement in pain scores after 6
months of treatment with ruboxistaurin 64 mg NIS = neurologic impairment score; NIS(LL) = neurologic impairment score of
administered every day and after 1 year of treatment the lower limbs; QD = every day.
Reproduced with permission from Litchy et al. Diabetes. 2002;51(suppl 2):A197.42
with ruboxistaurin 32 mg administered every day.43
As the study results indicate, ruboxistaurin appears
to have beneficial effects on nerve impairment and
pain symptoms in patients with DPN. Two large mul- REFERENCES
ticenter studies of this drug are under way, and reports
of their results are awaited. 1. Thomas PK. Diabetic peripheral neuropathies: their cost to
patient and society and the value of knowledge of risk fac-
tors for development of interventions. Eur Neurol. 1999;
CONCLUSIONS 41(suppl 1):35-43.
2. Duby JJ, Campbell RK, Setter SM, et al. Diabetic neuropa-
thy: an intensive review. Am J Health Syst Pharm.
Diabetic peripheral neuropathy is a common dis- 2004;61:160-173.
ease, affecting up to 50% of patients with diabetes and 3. Gordois A, Scuffham P, Shearer A, et al. The health care
accounts for considerable morbidity, mortality, and costs of diabetic peripheral neuropathy in the US. Diabetes
reduced quality of life. 4. Boulton AJ. The diabetic foot: a global view. Diabetes
Glycemic control is the central component of treat- Metab Res Rev. 2000;16(suppl 1):S2-S5.
ment but is difficult to achieve for many patients. 5. Abbott CA, Carrington AL, Ashe H, et al. The North-West
Diabetes Foot Care Study: incidence of, and risk factors for,
Cardiovascular risk factors play a major role in dia- new diabetic foot ulceration in a community-based patient
betes mellitus and the pathogenesis of DPN and cohort. Diabet Med. 2002;19:377-384.
6. Reiber GE. The epidemiology of diabetic foot problems.
should also be controlled. Diabet Med. 1996;13(suppl 1):S6-S11.
Painful neuropathy is difficult to treat. Because 7. Muller IS, de Grauw WJ, van Gerwen WH, et al. Foot
available therapies for pain and nerve impairment are ulceration and lower limb amputation in type 2 diabetic
patients in Dutch primary health care. Diabetes Care.
less than satisfactory, rational therapies that address the 2002;25:570-574.
underlying pathogenesis must be developed. Evidence 8. Boyko EJ, Ahroni JH, Smith DG, Davignon D. Increased
that the central nervous system and vascular factors are mortality associated with diabetic foot ulcer. Diabet Med.
involved in DPN should open new investigations for 9. Boulton AJM, ed. Diabetic Neuropathy. Bridgewater, NJ:
the development of effective therapies. At present, Aventis Pharma; 2001.
promising interventions that appear to be beneficial 10. Boulton AJM. The diabetic foot. In: Pickup JC, Williams G,
eds. Textbook of Diabetes. 2nd ed. Oxford, UK;
are being studied, and the results of the clinical trials Cambridge, Mass: Blackwell Science; 1997.
are eagerly awaited. 11. Quattrini C, Tesfaye S. Understanding the impact of painful
diabetic neuropathy. Diabetes Metab Res Rev. 2003;
S1020 Vol. 4 (10G) I December 2004
12. Tesfaye S, Stephens LK, Stephenson JM, et al. Prevalence of 28. Bril V, Perkins BA. Validation of the Toronto Clinical Scoring
diabetic peripheral neuropathy and its relation to glycaemic System for diabetic polyneuropathy. Diabetes Care.
control and potential risk factors: the EURODIAB IDDM 2002;25:2048-2052.
Complications Study . Diabetologia. 1996;39:1377-1384. 29. Smieja M, Hunt DL, Edelman D, et al. Clinical examination for
13. Tesfaye S, Chaturvedi N, Eaton SM, et al. Cardiovascular the detection of protective sensation in the feet of diabetic
risk factors predict development of diabetic peripheral neu- patients. International Cooperative Group for Clinical
ropathy. Diabetes Res Clin Pract. 2000;50:274. Examination Research. J Gen Intern Med. 1999;14:418-424.
14. Forsblom CM, Sane T, Groop PH, et al. Risk factors for 30. Rith-Najarian SJ, Stolusky T, Gohdes DM. Identifying diabetic
mortality in Type II (non-insulin-dependent) diabetes: evi- patients at high risk for lower-extremity amputation in a primary
dence of a role for neuropathy and a protective effect of health care setting. A prospective evaluation of simple screen-
HLA-DR4. Diabetologia. 1998;41:1253-1262. ing criteria. Diabetes Care. 1992;15:1386-1389.
15. Cameron NE, Eaton SE, Cotter MA, Tesfaye S. Vascular 31. Perkins BA, Olaleye D, Zinman B, Bril V. Simple screening
factors and metabolic interactions in the pathogenesis of tests for peripheral neuropathy in the diabetes clinic.
diabetic neuropathy. Diabetologia. 2001;44:1973-1988. Diabetes Care. 2001;24:250-256.
16. Gabbay KH, Merola LO, Field RA. Sorbitol pathway: pres- 32. Young MJ, Breddy JL, Veves A, Boulton AJ. The prediction of
ence in nerve and cord with substrate accumulation in dia- diabetic neuropathic foot ulceration using vibration percep-
betes. Science. 1966;151:209-210. tion thresholds. A prospective study. Diabetes Care. 1994;
17. Tomlinson DR, Moriarity RJ, Mayer JH. Prevention and rever- 17:557-560.
sal of defective axonal transport and motor nerve conduc- 33. Abbott CA, Vileikyte L, Williamson S, et al. Multicenter
tion velocity in rats with experimental diabetes by treatment study of the incidence of and predictive risk factors for dia-
with the aldose reductase inhibitor Sorbinil. Diabetes. betic neuropathic foot ulceration. Diabetes Care.
18. Judzewitsch RG, Jaspan JB, Polonsky KS, et al. Aldose 34. Coppini DV, Bowtell PA, Weng C, et al. Showing neuropa-
reductase inhibition improves nerve conduction velocity in thy is related to increased mortality in diabetic patients: a
diabetic patients. N Engl J Med. 1983;308:119-125. survival analysis using an accelerated failure time model.
19. Sima AA, Bril V, Nathaniel V, et al. Regeneration and J Clin Epidemiol. 2000;53:519-523.
repair of myelinated fibers in sural-nerve biopsy specimens 35. Dyck PJ, Zimmerman I, Gillen DA, et al. Cool, warm, and
from patients with diabetic neuropathy treated with sorbinil. heat-pain detection thresholds: testing methods and infer-
N Engl J Med. 1988;319:548-555. ences about anatomic distribution of receptors. Neurology.
20. Greene DA, Arezzo JC, Brown MB. Effect of aldose reduc- 1993;43:1500-1508.
tase inhibition on nerve conduction and morphometry in 36. Backonja M, Beydoun A, Edwards KR, et al. Gabapentin
diabetic neuropathy. Zenarestat Study Group [published for the symptomatic treatment of painful neuropathy in
correction appears in Neurology. 2001;56:1124]. patients with diabetes mellitus: a randomized controlled
Neurology. 1999;53:580-591. trial. JAMA. 1998;280:1831-1836.
21. Malik RA, Tesfaye S, Thompson SD, et al. Endoneurial 37. Harati Y, Gooch C, Swenson M, et al. Double-blind ran-
localisation of microvascular damage in human diabetic domized trial of tramadol for the treatment of the pain of
neuropathy. Diabetologia. 1993;36:454-459. diabetic neuropathy. Neurology. 1998;50:1842-1846.
22. Newrick PG, Wilson AJ, Jakubowski J, et al. Sural nerve 38. Tesfaye S, Watt J, Benbow SJ, et al. Electrical spinal-cord
oxygen tension in diabetes. Br Med J (Clin Res Ed). stimulation for painful diabetic peripheral neuropathy.
1986;293:1053-1054. Lancet. 1996;348:1698-1701.
23. Tesfaye S, Harris N, Jakubowski JJ, et al. Impaired blood 39. Eaton SE, Harris ND, Rajbhandari SM, et al. Spinal-cord
flow and arterio-venous shunting in human diabetic neu- involvement in diabetic peripheral neuropathy. Lancet.
ropathy: a novel technique of nerve photography and fluo- 2001;358:35-36.
roscein angiography. Diabetologia. 1993;36:1266-1274. 40. Selvarajah D, Wilkinson ID, Griffiths PD, et al. Abnormal
24. Tesfaye S, Harris ND, Wilson RM, Ward JD. Exercise- thalamic resonance spectroscopy: evidence for functional
induced conduction velocity increment: a marker of impairment of the thalamus in diabetic neuropathy. Diabet
impaired peripheral nerve blood flow in diabetic neuropa- Med. 2004;21(suppl 2):abstract.
thy. Diabetologia. 1992;35:155-159. 41. Ametov AS, Barinov A, Dyck PJ, et al. The sensory symp-
25. Tesfaye S, Malik R, Harris N, et al. Arterio-venous shunting toms of diabetic polyneuropathy are improved with alpha-
and proliferating new vessels in acute painful neuropathy of lipoic acid: the SYDNEY trial [published correction appears
rapid glycaemic control (insulin neuritis). Diabetologia. in Diabetes Care. 2003;26:2227]. Diabetes Care.
26. Eaton SE, Harris ND, Ibrahim S, et al. Increased sural 42. Litchy W, Dyck W, Tesfaye S, et al. Diabetic peripheral
nerve epineurial blood flow in human subjects with painful neuropathy (DPN) assessed by neurological examination
diabetic neuropathy. Diabetologia. 2003;46:934-939. (NE) and composite scores (CS) is improved with
27. Young MJ, Veves A, Smith JV, et al. Restoring lower limb blood LY333531 treatment. Diabetes. 2002;51(suppl 2):A197.
flow improves conduction velocity in diabetic patients. 43. Vinik AI, Tesfaye S, Zhang D, et al. LY333531 treatment
Diabetologia. 1995;38:1051-1054. improves diabetic peripheral neuropathy (DPN) with symp-
toms. Diabetes. 2002;51(suppl 2):A79.
Advanced Studies in Medicine I S1021