Diabetes mellitus affects over 14 million people in the United States and the number of diabetics is increasing by 5% per year. Diabetic neuropathy (DN) is a common complication of diabetes and occurs in approximately 50% of diabetic patients over time. Clinical trials have proven that hyperglycemia almost certainly conditions the development of DN. Despite this fact, we still do not understand the mechanism(s) underlying DN. Several possible etiologies have been proposed including altered metabolism of polyol, lipids, or amino acids, vascular insufficiency, increased superoxide-induced free radical formation, impaired axonal transport or reduced neurotrophism. Accumulating evidence suggests that these defects are likely interrelated and that their interaction(s) within the diabetic milieu are responsible for the development and progression of DN. In this review we will discuss these theories, their interrelationships and how, collectively, these ideas may begin to explain the etiology of DN.

Neuropathy is the most common symptomatic chronic complication in diabetic patients and accounts for substantial morbidity in the diabetic population. It is predominently a disease of the older diabetic population, and shows a progressive course with limb amputation as the final end-point of the disease. Pathologically the disorder is characterised by progressive degeneration as well as impaired regenerative ability of peripheral nerve fibers, resulting in a progressive loss and dying-back of the longest nerve fibres innervating the distal limbs. These changes are associated with progressive impairment of nerve function leading to impaired sensitivity in the limbs, which sometimes is associated with troublesome pain. Qualitatively similar but much milder functional and structural changes occur during normal aging processes, which potentially could make elderly diabetic patients more susceptible to an additional hyperglycaemic insult. The mechanisms underlying the development of diabetic neuropathy involve hyperglycaemia-induced metabolic abnormalities of peripheral nerve fibres and the supporting nutritive vascular supply. One of the major abnormalities involves activation of the polyol pathway with subsequent impairments in nerve function and vascular supply. Since hyperglycaemia appears to be the major culprit in the development of diabetic neuropathy, good glycaemic control is paramount in the long term treatment of diabetic patients to attenuate the development and/or progression of the disorder. Furthermore, elimination of risk factors such as obesity, smoking and excessive alcohol (ethanol) consumption, as well as patient education, are all important factors in the care of diabetic patients. In symptomatic neuropathic patients, including those with painful neuropathy, symptomatic and palliative measures are often effective. Stepwise addition of antidepressants to simple analgesics has proven to be effective in patients with troublesome pain. During recent years a class of drugs have been developed that inhibits the activation of the polyol pathway in diabetic nerves. These so-called aldose reductase inhibitors hold promise for a targeted treatment regimen in the near future. The aldose reductase inhibitors are already available in several European countries and in Japan.

OBJECTIVE—To evaluate the impact of prior intensive diabetes therapy on neuropathy among former Diabetes Control and Complications Trial (DCCT) participants. RESEARCH DESIGN AND METHODS—At the conclusion of the DCCT, subjects in the intensive group were encouraged to maintain intensive therapy, and subjects in the conventional group were encouraged to begin intensive therapy. Thereafter, we annually assessed neuropathy as part of the Epidemiology of Diabetes Intervention and Complications (EDIC) study. Neuropathy was defined using the Michigan Neuropathy Screening Instrument (MNSI). We recorded potential adverse consequences of neuropathy. RESULTS—At the first EDIC examination, 1,257 subjects participated in the neuropathy assessment. Consistent with DCCT results, the former intensive group showed a lower prevalence of neuropathy than the conventional group based on positive questionnaire (1.8 vs. 4.7%; P = 0.003) or examination (17.8 vs. 28.0%; P < 0.0001) results. Despite similar levels of glycemic control, symptoms and signs of neuropathy remained less prevalent among the former intensive group compared with the conventional group. At the beginning of the EDIC study, prior intensive therapy reduced the odds of having symptoms and signs of neuropathy using MNSI criteria by 64% (P = 0.0044) and 45% (P < 0.0001), respectively, with similar odds reductions observed for both neuropathic symptoms (51%, P < 0.0001) and neuropathic signs (43%, P < 0.0001) across 8 years of EDIC follow-up. CONCLUSIONS—The benefits of 6.5 years of intensive therapy on neuropathy status extended for at least 8 years beyond the end of the DCCT, similar to the findings described for diabetic retinopathy and nephropathy.

OBJECTIVE To evaluate the impact of systematic patient evaluation and patient and provider feedback on the processes and intermediate outcomes of diabetes care in Independent Practice Association model internal medicine practices. RESEARCH DESIGN AND METHODS Nine practices providing care to managed care patients were randomly assigned as intervention or comparison sites. Intervention-site subjects had Annual Diabetes Assessment Program (ADAP) assessments (HbA(1c), blood pressure, lipids, smoking, retinal photos, urine microalbumin, and foot examination) at years 1 and 2. Comparison-site subjects had ADAP assessments at year 2. At Intervention sites, year 1 ADAP results were reviewed with subjects, mailed to providers, and incorporated into electronic medical records with guideline-generated suggestions for treatment and follow-up. Medical records were evaluated for both groups for the year before both the year 1 and year 2 ADAP assessments. Processes and intermediate outcomes were compared using linear and logistic mixed hierarchical models. RESULTS Of 284 eligible subjects, 103 of 173 (60%) at the Intervention sites and 71 of 111 (64%) at the comparison sites participated; 83 of 103 (81%) of the intervention-site subjects returned for follow-up at year 2. Performance of the six recommended assessments improved in intervention-site subjects at year 2 compared with year 1 (5.8 vs. 4.3, P = 0.0001) and compared with comparison-site subjects at year 2 (4.2, P = 0.014). No significant changes were noted in intermediate outcomes. CONCLUSIONS The ADAP significantly improved processes of care but not intermediate outcomes. Additional interventions are needed to improve intermediate outcomes.

To explore whether microangiopathy is associated with disturbed glucose tolerance and peripheral neuropathy, we assessed endoneurial capillary morphology in sural nerve biopsies from men with diabetes, impaired glucose tolerance (IGT), and normal glucose tolerance (NGT). Baseline morphology was related to glucose tolerance and neuropathy at baseline and at follow-up 6 years later. Capillary density (in number per millimeters squared) at baseline was higher in subjects with diabetes (n = 10) compared with those with NGT (n = 5) at follow-up (median [interquartile range]) (86.0 [24.3] vs. 54.9 [17.1]; P = 0.0200) and in those progressing from IGT to diabetes (n = 4) compared with those with persistent IGT (n = 4) (86.7 [25.2] vs. 54.1 [14.6]; P = 0.0433). The capillary luminal area (in micrometers squared) was lower in subjects with NGT progressing to IGT (n = 2) or subjects with IGT progressing to diabetes (n = 3) compared with subjects with constant NGT (n = 6) or constant IGT (n = 4) (11.9 [2.4] vs. 20.8 [7.8]; P = 0.0201). The capillary basement membrane area (in micrometers squared) was increased in patients with peripheral neuropathy (n = 10) compared with those without (n = 7) (114.6 [68.8] vs. 75.3 [28.7]; P = 0.0084). In conclusion, increased capillary density was associated with current or future diabetes, decreased capillary luminal area with future deterioration in glucose tolerance, and increased basement membrane area with peripheral neuropathy.

OBJECTIVE Therapies with metformin, sulfonylureas, or insulin improve glycemic control in the short term but do not prevent progressive islet beta-cell failure or long-term deterioration in glycemia. Our goal was to evaluate, in patients recently diagnosed with type 2 diabetes (<3 years), the long-term efficacy of monotherapy with rosiglitazone on glycemic control and on the progression of pathophysiological abnormalities associated with type 2 diabetes as compared with metformin or glyburide monotherapy. RESEARCH DESIGN AND METHODS A Diabetes Outcome Progression Trial (ADOPT) is a randomized, double-blind, parallel-group study consisting of a screening visit, a 4-week placebo run-in, a 4-year treatment period, and an observational follow-up of approximately 3,600 drug-naïve patients with type 2 diabetes diagnosed within the previous 3 years. After run-in, patients will be randomized to rosiglitazone, glyburide, or metformin titrated to the maximum effective daily doses (8 mg rosiglitazone, 15 mg glyburide, or 2 g metformin). The primary outcome is time to monotherapy failure, defined as the time following titration to the maximal effective or tolerated dose when fasting plasma glucose exceeds 180 mg/dl (10 mmol/l). Secondary outcomes include measures of islet beta-cell function, insulin sensitivity, dyslipidemia, changes in urinary albumin excretion, plasminogen activator inhibitor-1 antigen, fibrinogen, and C-reactive protein. Safety and tolerability will also be evaluated. Patient-reported outcomes and resource utilization data will be collected and analyzed. CONCLUSIONS ADOPT will provide data on the effect of mechanistically differing treatment options on metabolic control, beta-cell function, and markers of macrovascular disease risk in type 2 diabetes.

Alterations in cyclooxygenase (COX) pathway activity have been implicated in the pathogenesis of experimental diabetic neuropathy (EDN). These studies explore the relationships between COX-mediated and acetyl-L-carnitine (ALC)-sensitive defects that contribute to functional, metabolic, and vascular abnormalities of EDN. The effects of nonselective COX inhibition with flurbiprofen were contrasted with selective COX-2 inhibition with meloxicam, administered alone and in combination with ALC in nondiabetic (ND) and streptozotocin-induced diabetic (STZ-D) rats. Flurbiprofen treatment of ND rats replicated many of the biochemical and physiological abnormalities of EDN, i.e., reduced motor nerve conduction velocity (MNCV), total and endoneurial nerve blood flow (NBF), Na,K-ATPase activity, and myo-inositol (MI) and taurine content. In STZ-D rats, however, flurbiprofen paradoxically prevented endoneurial NBF deficits but not MNCV slowing. Coadministration of 50 mg x kg(-1) x day(-1) ALC prevented reductions in MNCV, Na,K-ATPase activity, and endoneurial NBF in flurbiprofen-treated ND and STZ-D rats. In contrast, selective COX-2 inhibition with meloxicam was without effect on MNCV, NBF, or MI content in ND rats and prevented MNCV slowing and NBF deficits in STZ-D rats. Western blot analysis showed unchanged sciatic nerve COX-1 protein but increased COX-2 protein abundance in STZ-D versus ND rats. These results imply 1) a tonic role of the COX-1 pathway in the regulation of nerve osmolytes and Na,K-ATPase activity and the maintenance of NBF in ND animals and 2) activation of the COX-2 pathway as an important mediator of NBF and MNCV deficits in EDN.

OBJECTIVE:The aim of the current study was to determine the independent clinical risk factors for predicting morphological severity of distal diabetic sensorimotor polynecuropathy (DSP) as determined by fiber density (FD) on sural nerve biopsy. RESEARCH DESIGN AND METHODS:A total of 89 patients with both type 1 and type 2 diabetes, ascertained from a large therapeutic randomized clinical trial, were included in this observational cohort study. Morphological severity of DSP was expressed as the myelinated FD in the sural nerve biopsy General linear models were used to assess the relationship between the morphological severity of DSP and various clinical risk factors. RESULTS:Glycated hemoglobin (GHb) was significantly related to FD in univariate and multivariate regression analyses. This relationship was present in models in which GHb was handled either as a continuous variable or as a categorical variable with the highest significance level, with a GHb cutoff level of 9%. After dividing patients into groups with optimal to moderate GHb < or = 9%) and suboptimal (GHb >9%) glycemic control, the difference in FD between the two groups ranged between 3,461 and 2,334 per mm2. FD was also significantly related to duration of diabetes and age of the patient. CONCLUSIONS:The severity of peripheral DSP expressed by morphological criteria was significantly related to glycemic control in type 1 and type 2 diabetic patients. Inconsistent with previously published electrophysiological data demonstrating a correlation between height and conduction velocity, increasing height is not associated with morphological severity. Based on the results of the present study, it might be hypothesized that improving glycemic control will lessen severity of DSP in terms of FD loss in subjects with diabetes.

Aldose reductase inhibitors (ARIs) prevent peripheral nerve dysfunction and morphological abnormalities in diabetic animal models. However, some experimental intervention studies and clinical trials of ARIs on diabetic neuropathy appeared disappointing because of either 1) their inadequate design and, in particular, insufficient correction of the sorbitol pathway activity or 2) the inability to reverse established functional and metabolic deficits of diabetic neuropathy by AR inhibition in general. We evaluated whether diabetes-induced changes in nerve function, metabolism, and antioxidative defense are corrected by the dose of ARI (sorbinil, 65 mg/kg/d in the diet), resulting in complete inhibition of increased sorbitol pathway activity. The groups included control rats and streptozotocin-diabetic rats treated with/without ARI for 2 weeks after 4 weeks of untreated diabetes. ARI treatment corrected diabetes-induced nerve functional changes; that is, decrease in endoneurial nutritive blood flow, motor and sensory nerve conduction velocities, and metabolic abnormalities (i.e., mitochondrial and cytosolic NAD+/NADH redox imbalances and energy deficiency). ARI restored nerve concentrations of two major non-enzymatic antioxidants, reduced glutathione (GSH) and ascorbate, and completely arrested diabetes-induced lipid peroxidation. In conclusion, treatment with adequate doses of ARIs (that is, doses that completely inhibit increased sorbitol pathway activity) is an effective approach for reversal of, at least, early diabetic neuropathy.

The role for nerve blood flow (NBF) vs. other factors in motor nerve conduction (MNC) slowing in short-term diabetes was assessed by evaluating alpha(1)-adrenoceptor antagonist prazosin on NBF, MNC, as well as metabolic imbalances and oxidative stress in the neural tissue. Control and diabetic rats were treated with or without prazosin (5 mg.kg(-1).d(-1) for 3 wk). NBF was measured by hydrogen clearance. Both endoneurial vascular conductance and MNC velocity were decreased in diabetic rats vs. controls, and this decrease was prevented by prazosin. Free NAD(+):NADH ratios in mitochondrial cristae, matrix, and cytosol assessed by metabolite indicator method, as well as phosphocreatine levels and phosphocreatine/creatine ratios, were decreased in diabetic rats, and this reduction was ameliorated by prazosin. Neither diabetes-induced accumulation of two major glycation agents, glucose and fructose, as well as sorbitol and total malondialdehyde plus 4-hydroxyalkenals nor depletion of myo-inositol, GSH, and taurine or decrease in (Na/K)-ATP-ase activity were affected by prazosin. In conclusion, decreased NBF, but not metabolic imbalances or oxidative stress in the neural tissue, is a key mechanism of MNC slowing in short-term diabetes. Further experiments are needed to estimate whether preservation of NBF is sufficient for prevention of nerve dysfunction and morphological abnormalities in long-standing diabetes or whether the aforementioned metabolic imbalances closely associated with impaired neurotropism are of greater importance in advanced than in early diabetic neuropathy.

Experimental diabetic peripheral neuropathy (DPN) is marked by impaired nerve conduction velocity (NCV), reduced nerve blood flow (NBF), and a variety of metabolic abnormalities in peripheral nerve that have been variously ascribed to hyperglycemia, abnormal fatty acid metabolism, ischemic hypoxia, and/or oxidative stress. Some investigators propose that NCV slowing in experimental DPN can be explained entirely on the basis of nerve energy depletion secondary to reduced NBF. This article reports highly selective effects of administration of the antioxidant DL-alpha-lipoic acid (LA) to streptozotocin-injected diabetic rats. LA improved digital sensory but not sciatic-tibial motor NCV, corrected endoneurial nutritive but not composite NBF, increased the mitochondrial oxidative state without correcting nerve energy depletion, and enhanced the accumulation of polyol pathway intermediates without worsening myo-inositol or taurine depletion. These studies implicate oxidative stress as an important pathophysiological factor in experimental DPN. They reveal complex interrelationships among nerve perfusion, energy metabolism, osmolyte content, conduction velocity, and oxidative stress that may reflect the heterogeneous and compartmentalized composition of peripheral nerve.

The relationship between aldose reductase (AR) activity and oxidative stress in tissue sites for diabetic complications remains controversial. We evaluated the ARI sorbinil and three antioxidants, i.e. DL‐α‐lipoic acid (LA), taurine (T) and vitamin E (VE), on indices of oxidative stress and sorbitol pathway intermediates in diabetic peripheral nerve. The groups included control (C) and streptozotocin‐diabetic (D) rats treated with/without: 1.ARI (65 mg/kg, for 2 wks after 4 wks of untreated diabetes); 2.T (1% of the diet); 3.VE (1% of the diet); or 4.LA(100 mg/kg i.p.). T, VE and LA were administered for 6 wks from induction of diabetes. Malondialdehyde plus 4‐hydroxyalkenal levels were increased in D vs C (p < 0.01) and this increase was corrected in D+ARI (p < 0.01 vs D). GSH and ascorbate levels as well as superoxide dismutase activity were decreased in D vs C, and this decrease was completely (GSH) or partially corrected by ARI. Three antioxidants counteracted oxidative stress, but not accumulation of the sorbitol pathway intermediates in the diabetic nerve. In conclusion, increased AR activity contributes to rather than protects from oxidative stress in the diabetic nerve. Oxidative stress is not a cause of sorbitol pathway intermediate accumulation in experimental diabetic neuropathy.

Aims Sorbitol and myo-inositol levels and morphology of sural nerve were compared with nerve function and clinical neuropathy in men with diabetic, impaired (IGT), and normal glucose tolerance.Methods After neurography of sural nerve and determinations of sensory thresholds for vibration, warm and cold on the foot, whole nerve sural nerve biopsy was performed in 10 men with Type 1 diabetes mellitus, 10 with IGT, and 10 with normal glucose tolerance. Polyol levels were assessed by gas-liquid chromatography/mass spectrometry.Results Sural nerve amplitudes were significantly lower and sorbitol levels significantly higher in diabetic patients (median (interquartile range)) (3.7 (3.5) mu V and 643 (412) pmol/mg protein, respectively) both compared with IGT (11.3 (10.6) mu V; P = 0.04 and 286 (83) pmol/mg protein; P = 0.0032, respectively) and normally glucose tolerant (10.0 (11.6); P = 0.0142 and 296 (250) pmol/mg protein; P = 0.0191, respectively) subjects. There were no differences in nerve morphology between the three groups. Nerve myo-inositol levels correlated, however, positively with cluster density (r(s) = 0.56; P = 0.0054). In diabetic and IGT subjects, sural nerve amplitudes (2.6 (3.8) vs. 12.1 (10.6) mu V; P = 0.0246) and myelinated nerve fibre density (MNFD; 4076 (1091) vs. 5219 (668) nerve fibres/mm(2); P = 0.0021) were significantly lower in nine subjects with clinical neuropathy than in 10 without.Conclusions Nerve degeneration (i.e. MNFD) correlated with clinical neuropathy but not with glucose tolerance status whereas nerve myo-inositol levels positively correlated with signs of nerve regeneration (i.e. increased cluster density).

Several lines of evidence support peripheral nerve ischemia as a contributing factor in the etiology of human diabetic neuropathy. We questioned whether diabetic subjects with relatively normal nerve function in the baseline state would be more likely than healthy control subjects to show either improvement of ulnar nerve function with acute intraarterial infusion of nitroprusside (vasodilation) or be more sensitive than control subjects to worsening of nerve function with acute intraarterial infusion of norepinephrine (vasoconstriction). We measured forearm blood flow (FABF) using venous occlusion plethysmography and assessed ulnar nerve function at baseline and during two intrabrachial artery infusions. Six nondiabetic control subjects (mean age, 56 years) and 11 subjects with type 2 diabetes (mean age, 58 years) in good general health participated. Only three type 2 diabetic subjects had peripheral sensory neuropathy, which was mild. Among control subjects, there was no significant change in sensory distal latency, motor distal latency, motor proximal latency, or sensory or motor conduction velocity during norepinephrine infusion. In contrast, among type 2 diabetic subjects, there was a significant increase in sensory (baseline vnorepinephrine, 2.73+/-0.10 v 2.94+/-0.10 milliseconds [MS], P< or =.01) and motor distal latencies (baseline v norepinephrine, 2.90+/-0.06 v 3.18+/-0.1 ms, P< or =.001) and motor proximal latency (baseline v norepinephrine, 7.15+/-0.18 v 7.60+/-0.23 ms, P<.01) and a decrease in sensory conduction velocity (baseline v norepinephrine, 52.1+/-2.0 v 47.7+/-1.6 m/s, P<.01) during norepinephrine infusion. There were no consistent changes in nerve function during nitroprusside infusion in either group. In summary, we found that subjects with type 2 diabetes, but not control subjects, demonstrate a decrement in nerve function with vasoconstriction during intraarterial infusion of norepinephrine, but no consistent change during nitroprusside-induced vasodilation. These findings suggest there may be enhanced sensitivity of nerve function to ischemia in type 2 diabetic subjects with mild or absent clinical neuropathy.

The effect of the adenosine (AD) analog 2-chloroadenosine (C-AD) on glucose-induced inhibition of phosphoinositide synthesis was studied in human retinal pigment epithelial (RPE) cells by monitoring the level of the phosphatidylinositol (PI) synthase substrate, cytidine diphosphate diglyceride (CDP-DG). In high-aldose reductase (AR)-expressing RPE 91 cells, C-AD decreased CDP-DG at 5 mmol/L glucose and reversed the increase by 20 mmol/L glucose. AD deaminase (ADA), which inactivates endogenously released AD, potentiated the hyperglycemia-induced increase in CDP-DG. Theophylline, an AD-A1 and AD-A2 receptor antagonist, caused an increase in CDP-DG at 20 mmol/L glucose. C-AD did not alter CDP-DG in low-AR-expressing RPE 45 cells, but did decrease CDP-DG after cells were conditioned in 300 mmol/L glucose for 1 week (which induces AR). The mechanism by which AD regulates PI synthase in cells with high AR activity is unknown, but it is independent of Gi or Gs proteins, adenylate cyclase and phospholipase C (PLC) activation, myo-inositol (MI) uptake, or MI efflux. Administration of C-AD to streptozotocin-induced diabetic rats prevented the slowing of motor nerve conduction velocity (MNCV). Thus, AD derivatives, which reverse a glucose-induced deficit in phosphoinositide metabolism, might serve as a useful pharmacological tool to intervene in hyperglycemia-induced diabetic complications.