DIABETES/METABOLISM RESEARCH AND REVIEWS REVIEW Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/dmrr.2239

ARTICLE

Advances in the epidemiology, pathogenesis and management of diabetic peripheral neuropathy

Solomon Tesfaye* Dinesh Selvarajah Sheffield Teaching Hospitals, Sheffield, UK *Correspondence to: Solomon Tesfaye M.D., Room Q26, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, UK. E-mail: [email protected].

SUMMARY Diabetic peripheral neuropathy (DPN) affects up to 50% of patients with diabetes and is a major cause of morbidity and increased mortality. Its clinical manifestations include painful neuropathic symptoms and insensitivity, which increases the risk for burns, injuries and foot ulceration. Several recent studies have implicated poor glycaemic control, duration of diabetes, hyperlipidaemia (particularly hypertryglyceridaemia), elevated albumin excretion rates and obesity as risk factors for the development of DPN. Although there is now strong evidence for the importance of nerve microvascular disease in the pathogenesis of DPN, the risk factors for painful DPN are not known. However, emerging evidence regarding the central correlates of painful DPN is now afforded by brain imaging. The diagnosis of DPN begins with a careful history of sensory and motor symptoms. The quality and severity of neuropathic pain if present should be assessed using a suitable scale. Clinical examination should include inspection of the feet and evaluation of reflexes and sensory responses to vibration, light touch, pinprick and the 10-g monofilament. Glycaemic control and addressing cardiovascular risk is now considered important in the overall management of the neuropathic patient. Pharmacological treatment of painful DPN includes tricyclic compounds, serotonin–norepinephrine reuptake inhibitors (e.g. duloxetine), anticonvulsants (e.g. pregabalin), opiates, membrane stabilizers, the antioxidant alpha lipoic acid and others. Over the past 7 years, new agents with perhaps less side effect profiles have immerged. Management of patients with painful neuropathy must be tailored to individual requirements and will depend on the presence of other co-morbidities. There is limited literature with regard to combination treatment. Copyright © 2012 John Wiley & Sons, Ltd. Keywords diabetic neuropathy; diabetic peripheral neuropathy; painful diabetic neuropathy; pathogenesis of diabetic neuropathy; MRI

Received: 6 September 2011 Revised: 12 September 2011 Accepted: 13 October 2011

Copyright © 2012 John Wiley & Sons, Ltd.

Diabetic peripheral neuropathy (DPN) is associated with considerable morbidity, mortality and diminished quality of life [1]. Characterized by pain, paraesthesia and sensory loss, it affects up to 50% of people with diabetes [1]. In absolute numbers, against the estimated global prevalence of diabetes of 472 million by 2030 [2], DPN is likely to affect as many as 236 million persons worldwide and at a tremendous cost. In the United States alone, the total cost associated with DPN is $10.9 billion a year [3]. Thus, from these epidemiologic data, it is clear that DPN and the associated foot ulceration and neuropathic pain are far from rare and far from benign, posing a major healthcare challenge to the medical profession and to the society.

Advances in Epidemiology, Pathogenesis and Management of DPN

Clinical features of DPN Diabetic peripheral neuropathy is the most common neuropathic syndrome seen in persons with diabetes. The Toronto Consensus Panel on Diabetic Neuropathy recently defined DPN as a ‘symmetrical, length-dependent sensorimotor polyneuropathy attributable to metabolic and microvessel alterations as a result of chronic hyperglycaemia exposure and cardiovascular risk covariates. An abnormality of nerve conduction tests, which is frequently subclinical, appears to be the first objective quantitative indication of the condition’ [4]. Less common neuropathic syndromes include cranial mononeuropathies and focal neuropathies such as proximal motor neuropathy. DPN starts in the toes and gradually moves proximally. Once it is well established in the lower limbs, it affects the upper limbs, with sensory loss following the typical ‘glove and stocking’ pattern of distribution [1]. Significant motor deficits are not common in the early stages of DPN [1]. The patient does not typically complain of weakness, but when symptoms are present, they tend to be sensory in nature. Symptomatic muscle weakness tends to develop later in the disease course. Painful symptoms such as burning, tingling (‘pins and needles’ or paraesthesia), shooting (like electric shock) or lancing (stabbing) are present in around a third of patients with DPN and around 20% of all diabetic patients [1,5]. These symptoms are generally worse at night and disturb sleep [6]. Together with painful symptoms during the day, this often leads to a reduction in individual’s ability to perform daily activities [6]. The burden of painful DPN was reported to be considerable in one study, which resulted in a persistent discomfort despite polypharmacy and high resource use, and led to limitations in daily activities and poor satisfaction with treatments that were often deemed to be inappropriate [7]. Chronic persistently painful DPN can be extremely distressing and might be associated with profound depression together with anxiety [7]. Importantly, symptoms are not a reliable indicator of the severity of the nerve damage. Some patients with severe pain symptoms have little sensory deficit, whereas others with no painful symptoms have completely numb feet, putting them at extremely high risk for foot ulceration. Insensitivity, or loss of pain, can lead to foot ulceration and a host of unintentional but serious injuries. Patients who have lost feeling in their hands cannot sense temperature and often burn themselves while, for example, cooking or ironing, and also have difficulty handling small objects. Those who have lost sensation in their feet often sustain puncture wounds, friction wounds and burns that can become infected and/or ulcerated and lead to amputation. However, with appropriate foot care, a significant number of ulcerations can be prevented [8].

Risk factors for DPN Studies in patients with type 1 or type 2 diabetes have shown that poor glycaemic control is a risk factor for Copyright © 2012 John Wiley & Sons, Ltd.

9

DPN, but other risk factors are involved as well. The EURODIAB IDDM Complications Study, which involved 3250 patients with type 1 diabetes from 31 centres in 16 European countries, found that DPN was related to both glycaemic control and duration of disease [9]. Although the 28% baseline prevalence of DPN was significantly related to glycosylated haemoglobin (HbA1c) (p < 0.001), the prevalence varied from 17 to 41% after data were adjusted for duration of diabetes, with lower HbA1c levels associated with lower prevalence rates and higher levels associated with higher prevalence rates. However, even those with good glycaemic control (HbA1c < 5.4%, equivalent to Diabetes Control and Complications Trial HbA1c of 7%) still developed microvascular disease, suggesting that factors other than glycaemic control and disease duration are involved [9]. Follow-up data from the EURODIAB cohort of patients with type 1 diabetes revealed that over a 7-year period, approximately one-quarter of type 1 diabetic patients developed DPN; with age, duration of diabetes and poor glycaemic control being major factors [10]. The development of DPN was also associated with potentially modifiable cardiovascular risk factors, such as hypertension, hyperlipidaemia, obesity and cigarette smoking (Figure 1) [10]. Recently, other studies have also implicated cardiovascular risk factors, such as obesity [11] and triglycerides [12] in the pathogenesis of DPN. Moreover, Wiggin et al. found that elevated tryglycerides correlated with myelinated fibre loss independent of disease duration, age and diabetes control [13]. These data support the evolving concept that hyperlipidaemia might be instrumental in the progression of DPN. There is also evidence that DPN is associated with cardiovascular disease and mortality. In a study of 132 patients with type 2 diabetes, 38 died during the 9-year follow-up period [14]. Macroangiopathy was found to be the strongest independent risk factor for mortality, followed in descending order by DPN, albumin excretion

Figure 1. Risk factors for incident neuropathy. The EURODIAB Prospective Complications Study showing odds ratios for the various risk factors for diabetic peripheral neuropathy in a cohort of 1101 type 1 diabetes mellitus patients followed for 7.3  0.6 years. BMI, body mass index; CVD, cardiovascular disease Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr

10

rate and HbA1c. Elevated vibration threshold has also been found to be a risk factor for mortality in diabetic patients [15].

Pathogenesis of DPN Until recently, there were two schools of thought regarding the aetiology and pathogenesis of DPN: metabolic versus vascular. Recent studies, however, have shown that both vascular factors and metabolic interactions are involved at all stages of DPN [16]. Nerve fibre loss is the cause of insensitivity in DPN. As revealed by fascicular biopsy of the sural nerve, nerve fibres in patients with diabetes but no DPN are more numerous than in those with DPN [16]. Sural nerve biopsies also reveal microvascular defects in the endoneurial vessels, such as gross basement membrane thickening, endothelial cell proliferation and hypertrophy [17] as well as reduced oxygen tension [18] in patients with DPN compared with those who have diabetes but do not have DPN. Similarly, photography of surgically exposed sural nerve in vivo reveals microvascular abnormalities in its epineurial arteries and veins (Figure 2) [19], whereas fluorescein angiography reveals arteriosclerosis on the surface of the nerve and impaired blood flow [19] in patients with DPN compared with those with diabetes but no DPN.

Mechanisms of neuropathic pain in diabetes The exact pathophysiological mechanisms of neuropathic pain in diabetes remain enigmatic although several mechanisms including neurostructural correlates for painful neuropathy have been postulated (Table 1) [20]. Other potential mechanisms include the association of increased blood

Figure 2. Microvascular abnormalities in epineurial vessels in diabetes and diabetes with DPN. The patient with DPN has epineurial arterial attenuation/tortuosity and an arteriovenous shunt leading to increased venous pressure and tortuosity Copyright © 2012 John Wiley & Sons, Ltd.

S. Tesfaye and D. Selvarajah Table 1. Mechanisms of neuropathic pain (adapted from Reference [20]) Peripheral mechanisms

Central mechanisms

Changes in sodium channel distribution and expression Changes in calcium channel distribution and expression Altered neuropeptide expression

Central sensitization Ab-fibre sprouting into lamina II of the dorsal horn Reduced inhibition via descending pathways

Sympathetic sprouting Peripheral sensitization Altered peripheral blood flow Axonal atrophy, degeneration or regeneration Damage to small fibres Glycaemic flux

glucose instability in the genesis of neuropathic pain [21], an increase in peripheral nerve epineurial blood flow [22], altered foot skin microcirculation [23], reduced intraepidermal nerve fibre density in the context of early neuropathy [24], an increase in thalamic vascularity [25] and autonomic dysfunction [26].

Imaging studies of the central nervous system and DPN There is now increasing evidence that the impact of diabetes may be at all levels of the nervous system – after all, diabetes is a metabolic disorder and brain micro/ macrovascular disease in diabetes appears to be associated with cognitive decline and brain atrophy [27], and involvement of the central nervous system (CNS) in DPN is also increasingly being recognized.

Spinal cord involvement in DPN Involvement of the spinal cord has been reported in postmortem studies. However, many of these studies did not examine patients with DPN specifically and were conducted at a time when diabetes treatments were less than optimal and other neurological diseases of the cord (e.g. syphilis) were more prevalent [28]. The possibility of spinal cord involvement in DPN came to light when it was observed that electrical spinal cord stimulation, which can alleviate neuropathic pain, was ineffective in subjects with severe loss of vibration and joint position sense [29]. Using a non-invasive magnetic resonance (MR) imaging technique, Eaton et al. reported a significantly lower cord area in the cervical and upper thoracic regions in subjects with established DPN compared with healthy nondiabetic control subjects [30]. Diabetic subjects with no DPN had intermediate cord area measurements between nondiabetic controls and established DPN [30]. To confirm these findings and examine if these changes occur early in the natural history of DPN, a larger, adequately powered study was conducted. In this study, Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr

11

Advances in Epidemiology, Pathogenesis and Management of DPN

subjects with type 1 diabetes were subdivided into three subgroups (no DPN, subclinical DPN and established DPN), and spinal cord area measurements were performed at the C2/C3 level [31]. The study revealed that there was a significant shrinkage of the spinal cord even in subject with early subclinical DPN. Significant correlations were also found between cord area and neurophysiological parameters of DPN severity. Thus, this study showed that the neuropathic process in diabetes is not confined to the peripheral nerve and appears to involve the spinal cord. Of concern is that this occurs early in the neuropathic process [31].

Brain MR spectroscopy in DPN Our findings of involvement of the spinal cord in DPN made us question whether the brain too may be involved. Ascending sensory pathways of the spinal cord terminate within the ventroposterior lateral (VPL) thalamic subnucleus before high-order sensory projections are sent to the cortex [32]. The thalamus plays an important modulatory role of sensory information that is presented to the cortex [33]. We used proton MR spectroscopy (H-MRS), a non-invasive MR technique that can provide metabolic information from different body tissues, to investigate if thalamic neuronal function is affected in DPN. H-MRS produces spectra that contain several resonances or peaks. In brain parenchyma, the three major peaks detected are due to N-acetyl groups, total creatine and choline-containing compounds [28]. Immunohistochemical studies have suggested that N-acetyl aspartate (NAA), the major constituent of the N-acetyl group resonance at long echo time (TE), is localized exclusively in neurons and their processes throughout the CNS [28]. Thus, NAA resonance on H-MRS can provide a useful marker for brain neuronal and axonal integrity in vivo. In a preliminary study, subjects with type 1 diabetes (no DPN and established DPN) and age-matched and sexmatched nondiabetic healthy controls underwent H-MRS of the thalamus. The study showed a significantly lower long TE (135 ms) thalamic NAA/choline ratio in the group of patients with DPN compared with patients with no DPN and healthy volunteer controls [34]. The data also demonstrated significant correlations between short TE (20 ms) signal from NAA and neurophysiological markers (overall neuropathy composite score and individual nerve function tests) of DPN severity [34]. These findings may reflect thalamic neuronal dysfunction in DPN, implicating the brain in the neuropathic process. However, the mechanism of thalamic involvement is unclear. One possible explanation for thalamic neuronal dysfunction in DPN may be that loss of afferent input, as a result of peripheral nerve damage. Correlations observed between NAA acquired at short TE, duration of diabetes and severity of neuropathy are supportive of this suggestion. It may also be equally likely that the observed changes in the thalamus may be occurring concomitantly to the changes seen in the peripheral nervous system. Further studies Copyright © 2012 John Wiley & Sons, Ltd.

utilizing these techniques on a much larger numbers of subjects including a subgroup of patients with painful DPN are now clearly necessary.

MR perfusion imaging in DPN Although the pathogenesis of thalamic involvement on H-MRS in DPN is unknown [34], it is likely that both vascular and metabolic etiological factors that have been postulated in the pathogenesis of DPN and other microvascular complications of diabetes (retinopathy and nephropathy) may be involved. A further MR perfusion imaging study at 1.5 T was undertaken involving subjects with type 1 diabetes (no DPN, painful DPN and painless DPN) and a group of nondiabetic healthy volunteers. Exogenous perfusion contrast (intravenous bolus of gadolinium chelate) was used to examine the microvascular perfusion characteristics of both the thalamus and caudate nucleus (control region). Established markers of cerebral microvascular perfusion used as study endpoints were relative cerebral blood volume (rCBV), flow and bolus transit time. Figure 3 shows the composite mean relative concentration-time profile in the thalamus for the four groups. Group comparisons showed that painless DPN had lower thalamic rCBV compared with healthy volunteers and no DPN, whereas painful DPN had higher thalamic rCBV (p = 0.04, Chi squared = 8.3) [34]. Hence, painful DPN is accompanied by increased thalamic vascularity, whereas painless DPN is associated with greater thalamic microvascular impairment. Similar changes were not demonstrated in the caudate nucleus, which was a control area (not involved in somatosensory perception) [25]. Similar thalamic microvascular abnormalities have been demonstrated in other chronic pain states and may thus be important in the pathogenesis of painful DPN. However, what is causing thalamic hyperperfusion is unclear; possible reasons include impaired autoregulation 40 No DPN

35

HV

30

Painless DN

25

Painful DPN

20 15 10 5 0 0

10

20

30

40

50

60

70

80

90 100 110 120 130

Figure 3. Composite concentration-time profiles of the bolus passage of exogenous contrast agent (gadolinium-diethylenetriamine pentaacetic acid) through the thalamus in each subgroup. DPN, diabetic peripheral neuropathy; HV, healthy volunteer. Adapted from Reference [28] Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr

12

because of C-fibre-mediated/endothelial dysfunction or aberrant spontaneous thalamic activity. A recent study in the streptozotocin diabetic rat model of neuropathic pain, thalamic VPL neurons were found be hyperexcitable, with increased responses to phasic brush, press and pinch stimuli applied to identified peripheral receptive field [35]. VPL neurons from diabetic rats also displayed enhanced spontaneous activity, independent of ascending afferent barrage and enlarged receptive fields. The authors suggested that thalamic neurons act as central generators or amplifiers of pain in diabetes, and this could be associated with increased thalamic vascularity [36].

Functional MR imaging and painful DPN Functional MR imaging (fMRI) is one scientific development that has led to recent advances in our understanding of the function of the human brain [37]. The technique relies on mapping localized changes in magnetic susceptibility that occurs following the hemodynamic response to neuronal activity. Small susceptibility changes (which depend on alteration of the localized ratio of oxyhaemoglobin to deoxyhaemoglobin, hence the acronym BOLD or bloodoxygen-level-dependent fMRI) lead to small signal changes on susceptibility-weighted imaging at high-temporal resolution [28]. These are detected because of small differences in contrast to noise, following multiple dataaveraging strategies and statistical analysis. Brain areas whose signal signatures significantly correlate with the stimulus are those that are defined as being ‘active’ during a task or presentation of a stimulus. Studies using fMRI have investigated changes in brain activity in response to various experimental stimuli inducing pain. The results suggested that multiple cortical and subcortical regions are activated during painful coetaneous heat stimulation in healthy subjects. Activation in the insula, anterior cingulated cortex (ACC), prefrontal cortex (PFC), thalamus, primary and secondary somatosensory cortices (S1 and S2) and the basal ganglia is seen across the groups [38]. This led to the characterization of a network of brain areas that consistently activate in response to pain, forming a ‘pain matrix’[39]. These regions are primarily responsible for discriminating location and intensity of painful stimuli together with affective pain processing [28]. Thus, pain is no longer a purely subjective phenomenon – fMRI studies are now trying to establish objective radiological correlates to the pain experience in different chronic pain contexts. The majority of studies to date, however, have been performed mainly in healthy volunteers following acute pain stimulation, and changes in the brain associated with chronic pain have been less thoroughly investigated [28]. It is important that the pain matrix should be viewed as a flexible, integrative entity in pain perception, which may be modulated by other aspects of the pain experience, including context, mood and attention [40]. Many of the previous fMRI studies have tended to group different aetiologies of neuropathic pain together, and hence, the exact pathophysiological CNS mechanisms of painful Copyright © 2012 John Wiley & Sons, Ltd.

S. Tesfaye and D. Selvarajah

DPN remain unknown. Furthermore, given the unique pathophysiology of diabetes, involving both metabolic and vascular abnormalities, it is likely that different pathophysiological mechanisms may be involved. Utilizing fMRI, Wilkinson et al. performed a preliminary study comprising type 1 diabetic subjects (no DPN, painful DPN and painless DPN) to test the feasibility of monitoring the brain’s response to the presentation of heat pain in the context of DPN [41]. The results show that subjects with no DPN had greater BOLD response than those with painless DPN. Subjects with painful DPN showed significantly greater response than those with painless DPN. The primary somatosensory cortex, lateral frontal and cerebellar regions demonstrated greatest involvement. This may be explained by the reduced ascending nociceptive input as a result of neuropathy. However, subjects with painful and painless DPN had comparable neuropathy impairment scores based on detailed neurophysiological assessments. We also found significant negative correlation between BOLD fMRI response and overall neuropathy score in both the thalamus and left parietal lobe. One component of neuropathic pain is allodynia, which is pain elicited by normally nonpainful stimuli, and is frequently associated with spontaneous pain in subjects with diabetes. Schweinhardt et al. found that the magnitude of activation in the caudal anterior insula correlates with the perceived intensity of allodynic pain across subjects, independent of the level of ongoing pain [42]. In a preliminary analysis of an ongoing fMRI study, we have looked at the BOLD response of subjects with painless DPN, painful DPN and healthy volunteers in response to heat pain [43]. We demonstrated that nociceptive heat stimulation applied to both foot and thigh regions elicited significant increase in neuronal activation within the pain matrix in all the groups except in painless DPN during foot stimulation. This is not unexpected because these latter subjects were insensate at this stimulation site. Subsequent group level comparisons revealed that subjects with painful DPN displayed significantly greater neuronal activation within the pain matrix, particularly in the PFC and ACC during foot pain stimulation compared with painless DPN [43]. This pattern of increased activation persisted in painful DPN when compared with healthy volunteers, with increased activation mainly in the PFC and ACC [43]. Finally, a greater understanding of these central processes afforded by imaging is likely to lead to the development of robust, objective radiological correlates to the subjective pain experience and the targeting of specific pain networks/pathways to develop more effective and novel treatments with less side effects.

Treatment of painful DPN The current approach to management of painful DPN centres around achieving and maintaining near-normal glycaemia (HbA1c) as an initial step. However, many patients with diabetes, particularly those with type 2 diabetes, find this difficult. The assessment and Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr

13

Advances in Epidemiology, Pathogenesis and Management of DPN

pharmacological treatment of painful DPN has been reviewed recently, and the reader is advised to read this consensus report [20]. On the basis of the clinical trial evidence for the various pharmacological agents (efficacy and safety) for painful DPN, the Toronto Consensus Panel on Diabetic Neuropathy recommended that a tricyclic antidepressant (TCA), a serotonin–norepinephrine reuptake inhibitor (SNRI) or an a-2-d agonist should be considered for first line treatments. On the basis of trial data, duloxetine would be the preferred SNRI and pregabalin would be the preferred a-2-d agonist. If pain is inadequately controlled, depending upon contraindications, these first line agents can be combined, although this is not backed by trial evidence [20]. If pain is still inadequately controlled, opioids such as tramadol and oxycodone might be added in a combination treatment [20]. Initial selection of first line treatment will be influenced by the assessment of contraindications, consideration of co-morbidities and cost; for example, in diabetic patients with a history of heart disease, elderly patients on other concomitant medications such as diuretics and antihypertensives, patients with co-morbid orthostatic hypotension and others, TCA have relative contraindications. In patients with liver disease, duloxetine should not be prescribed, and in those with oedema, pregabalin or gabapentin should be avoided [20].

Pathogenetic treatments Apart from glycaemic control, there has been little advance in the development compounds that can halt the neuropathic process. Although several disease-modifying agents are under investigation, only the antioxidant a-lipoic acid is supported by a meta-analysis and is marketed in certain countries [44].

Conclusions Diabetic peripheral neuropathy is common, affecting up to 50% of patients with diabetes. In addition, it accounts for considerable morbidity, mortality and reduced quality of life. Glycaemic control is the central component of treatment, but it is difficult to achieve for many patients. Because cardiovascular risk factors play a major role in diabetes and the pathogenesis of DPN, they should be controlled as well. Painful DPN is difficult to treat. On the basis of the trial evidence, first line therapies include a TCA, the SNRI duloxetine and the anticonvulsant pregabalin. Combination therapy might be useful for those with more severe pain, but there is paucity of studies and further research is required. Studies are also required on direct head-tohead comparative trials and long-term efficacy of drugs, as most trials have lasted less than 6 months. Imaging evidence that the CNS is involved in DPN should open new avenues of investigation. Key target areas generating or modulating pain in painful DPN including peripheral small fibres with modulation at the level of the spinal cord, the thalamus and the other pain matrix areas in the brain require further studies to develop more effective treatments. The association of painful DPN with autonomic neuropathy also merits further investigation. Finally, as available therapies for pain and nerve impairment are less than satisfactory, rational therapies that address the underlying pathogenesis need to be developed.

Conflict of Interest Both S. Tesfaye and D. Selvarajah have received honoraria from Eli Lilly and Company. S. Tesfaye has also received honoraria from Pfizer.

REFERENCES 1. Tesfaye S. Recent advances in the management of diabetic symmetrical polyneuropathy. J Diabetes Invest 2010; 2: 33–42. 2. http://www.idf.org/diabetesatlas/ diabetes-and-impaired-glucose-tolerance 3. Gordois A, Scuffham P, Shearer A, Oglesby A, Tobian JA. The health care costs of diabetic peripheral neuropathy in the US. Diabetes Care 2003; 26(6): 1790–1795. 4. Tesfaye S, Boulton AJ, Dyck PJ, et al. Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity and treatments. Diabetes Care 2010; 33(10): 2285–2293. 5. Davies M, Brophy S, Williams R, Taylor A. The prevalence, severity and impact of painful diabetic peripheral neuropathy in type 2 diabetes. Diabetes Care 2006; 29: 1518–1522. 6. Quattrini C, Tesfaye S. Understanding the impact of painful diabetic neuropathy. Copyright © 2012 John Wiley & Sons, Ltd.

7.

8. 9.

10.

11.

Diabetes Metab Res Rev 2003; 19(suppl 1): S2–S8. Gore M, Brandenburg N, Dukes E, et al. Pain severity in diabetic peripheral neuropathy is associated with patient functioning, symptom levels of anxiety and depression, and sleep. J Pain Symptoms manage 2005; 30: 374–385. Boulton AJM, Kirsner RS, Vileikyte L. Neuropathic diabetic foot ulcers. N Engl J Med 2004; 351: 48–55. Tesfaye S, Stephens LK, Stephenson JM, et al. Prevalence of diabetic peripheral neuropathy and its relation to glycaemic control and potential risk factors: the EURODIAB IDDM Complications Study. Diabetologia 1996; 39(11): 1377–1384. Tesfaye S, Chaturvedi N, Eaton SEM, Witte D, Ward JD, Fuller J. Vascular risk factors and diabetic neuropathy. New Engl J Med 2005; 352: 341–350. Ziegler D, Rathmann W, Dickhaus T, Meisinger C, Mielck A, KORA Study

Group. Prevalence of polyneuropathy in pre-diabetes and diabetes is associated with abdominal obesity and macroangiopathy: the MONICA/KORA Augsburg Surveys S2 and S3. Diabetes Care 2008; 31(3): 464–469. 12. Vincent AM, Hayes JM, McLean LL, Vivekanandan-Giri A, Pennathur S, Feldman EL. Dyslipidemia-induced neuropathy in mice: the role of oxLDL/LOX-1. Diabetes 2009; 58(10): 2376–85. 13. Wiggin TD, Sullivan KA, Pop-Busui R, Amato A, Sima AA, Feldman EL. Elevated triglycerides correlate with progression of diabetic neuropathy. Diabetes 2009; 58(7): 1634–40. 14. Forsblom CM, Sane T, Groop PH, et al. Risk factors for mortality in Type II (non-insulin-dependent) diabetes: evidence of a role for neuropathy and a protective effect of HLA-DR4. Diabetologia 1998; 41(11): 1253–1262. Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr

14 15. Coppini DV, Bowtell PA, Weng C, Young PJ, Sonksen PH. Showing neuropathy is related to increased mortality in diabetic patients – a survival analysis using an accelerated failure time model. J Clin Epidemiol. 2000; 53(5): 519–523. 16. Cameron NA, Eaton SE, Cotter MA, Tesfaye S. Vascular factors and metabolic interactions in the pathogenesis of diabetic neuropathy. Diabetologia 2001; 44(11): 1973–1988. 17. Malik RA, Tesfaye S, Thompson SD, et al. Endoneurial localisation of microvascular damage in human diabetic neuropathy. Diabetologia 1993; 36(5): 454–459. 18. Newrick PG, Wilson AJ, Jakubowski J, Boulton AJ, Ward JD. Sural nerve oxygen tension in diabetes. Br Med J (J Clin Res Ed) 1986; 293(6554): 1053–1054. 19. Tesfaye S, Harris N, Jakubowski JJ, et al. Impaired blood flow and arterio-venous shunting in human diabetic neuropathy: a novel technique of nerve photography and fluoroscein angiography. Diabetologia 1993; 36(12): 1266–1274. 20. Tesfaye S, Vileikyte L, Rayman G, et al. Painful diabetic peripheral neuropathy: consensus recommendations on diagnosis, assessment and management. Diabetes Metab Res Rev 2011; 27 629–638. 21. Oyibo S, Prasad YD, Jackson NJ, et al. The relationship between blood glucose excursions and painful diabetic peripheral neuropathy: a pilot study. Diabet Med 2002; 19: 870–873. 22. Eaton SE, Harris ND, Ibrahim S, et al. Differnces insural nerve haemodynamics in painful and painless neuropathy. Diabetologia 2003; 46: 934–939. 23. Quattrini C, Harris ND, Malik RA, Tesfaye S. Impaired skin microvascular reactivity in painful diabetic neuropathy. Diabetes Care 2007; 30: 655–659. 24. Sorensen L, Molyneaux L, Yue DK. The relationship among pain, sensory loss,

Copyright © 2012 John Wiley & Sons, Ltd.

S. Tesfaye and D. Selvarajah

25.

26.

27.

28.

29.

30.

31.

32.

33.

and small nerve fibers in diabetes. Diabetes Care 2006; 29: 883–887. Selvarajah D, Wilkinson ID, Gandhi R, Griffiths PD, Tesfaye S. Microvascular n perfusion abnormalities of the thalamus in painful but not painless diabetic polyneuropathy: a clue to the pathogenesis of pain in type 1 diabetes. Diabetes Care 2011; 34(3): 718–720. Gandhi R, Marques JLB, Selvarajah D, et al. Painful diabetic neuropathy is associated with greater autonomic dysfunction than painless diabetic neuropathy. Diabetes Care 2010; 33: 1585–1590. Manschot SM, Brands AM, van der Grond J, et al. Brain magnetic resonance imaging correlates of impaired cognition in patients with type 2 diabetes. Diabetes 2006; 55(4): 1106–1113. Selvarajah D, Wilkinson ID, Davies J, Gandhi R, Tesfaye S. Central nervous system involvement in diabetic neuropathy. Curr Diab Rep 2011; 11(4): 310–322. Tesfaye S, Watt J, Benbow SJ, Pang KA, Miles J, MacFarlane IA. Electrical spinal-cord stimulation for painful diabetic peripheral neuropathy. Lancet 1996; 348(9043): 1698–1701. Eaton SE, Harris ND, Rajbhandari SM, et al. Spinal-cord involvement in diabetic peripheral neuropathy. Lancet 2001; 358(9275): 35–36. Selvarajah D, Wilkinson ID, Emery CJ, et al. Early involvement of the spinal cord in diabetic peripheral neuropathy. Diabetes Care 2006; 29: 2664–2669. Wilson P, Kitchener PD, et al. Cutaneous receptive field organization in the ventral posterior nucleus of the thalamus in the common marmoset. J Neurophysiol 1999; 82: 1865–1875. McCormick DA, Bal T. Sensory gating mechanisms of the thalamus. Curr Opin Neurobiol 1994; 4: 550–556.

34. Selvarajah D, Wilkinson ID, Emery CJ, et al. Thalamic neuronal dysfunction and chronic sensorimotor distal symmetrical polyneuropathy in patients with type 1 diabetes mellitus. Diabetologia 2008; 51: 2088–2092. 35. Fischer TZ, Tan AM, Waxman SG. Thalamic neuron hyperexcitability and enlarged receptive fields in the STZ model of diabetic pain. Brain Res 2009; 1268(1): 154–161. 36. Fischer TZ, Waxman SG. Neuropathic pain in diabetes—evidence for a central mechanism. Nat Rev Neurol. 2010; 6(8): 462–466. 37. Tracey I. Can neuroimaging studies identify pain endophenotypes in humans? Nat Rev Neurol. 2011; 7(3): 173–181. 38. Bushnell MC, Duncan GH, Hofbauer RK, Ha B, Chen JI, Carrier B. Pain perception: is there a role for primary somatosensory cortex? Proc Natl Acad Sci U S A 1999; 96(14): 7705–7709. 39. Melzack R. From the gate to the neuromatrix. Pain 1999; Suppl 6: S121–S126. 40. Tracey I, Mantyh P. The cerebral signature for pain perception and its modulation. Neuron 2007; 55(3): 377–391. 41. Wilkinson ID, Gandhi RA, Hunter MD, et al. Functional MRI and response to pain in diabetic neuropathy. Diabet Med 2007; 24(suppl 1): P187. 42. Schweinhardt P, Glynn C, Brooks J, et al. An fMRI study of cerebral processing of brush-evoked allodynia in neuropathic pain patients. NeuroImage 2006; 32: 256–265. 43. Selvarajah D, Davies J, Sankar A, et al. Diabetologia 2011; Suppl 1: S81. 44. Ziegler D, Nowak H, Kempler P, Vargha P, Low PA. Treatment of symptomatic diabetic polyneuropathy with alphalipoic acid: a meta-analysis. Diabetic Med 2004; 21: 114–121.

Diabetes Metab Res Rev 2012; 28(Suppl 1): 8–14. DOI: 10.1002/dmrr