Neuromuscular Complications of Diabetes Mellitus

Abstract

Purpose of Review:

Diabetes mellitus has become a modern global epidemic, with steadily increasing prevalence rates related to lifestyle such that 27% of individuals aged 65 years or older have diabetes mellitus, 95% of whom have type 2. This article reviews the effects of diabetes mellitus on the neuromuscular system.

Recent Findings:

Diabetes mellitus leads to diverse forms of peripheral neuropathy as the major neuromuscular complication. Both focal and diffuse types of neuropathy can develop, with the most common form being diabetic sensorimotor polyneuropathy. Small fibers are damaged early in the development of diabetic sensorimotor polyneuropathy and are not assessed by nerve conduction studies. Small fiber damage occurs even in the prediabetes stage. No disease-modifying therapy for diabetic sensorimotor polyneuropathy is available at this time, but this complication can be limited in patients who have type 1 diabetes mellitus with strict glycemic control; the same outcome is not clearly observed in patients who have type 2 diabetes mellitus. Recently, the evidence base for symptomatic treatments of painful diabetic sensorimotor polyneuropathy underwent systematic review. Effective evidence-based treatments include some anticonvulsants (eg, pregabalin, gabapentin), antidepressants (eg, amitriptyline, duloxetine), opioids (eg, morphine sulfate, oxycodone), capsaicin cream, and transcutaneous electrical nerve stimulation.

Summary:

This article reviews the increasing prevalence of diabetes mellitus and diabetic sensorimotor polyneuropathy and discusses recent consensus opinion on the objective confirmation needed for the diagnosis in the clinical research setting. The evidence from clinical trials shows that intensive glycemic control reduces prevalence of diabetic sensorimotor polyneuropathy in patients with type 1 diabetes mellitus, but variable outcomes are observed in patients with type 2 diabetes mellitus. Finally, despite the lack of disease-modifying treatment, effective evidence-based therapy can control painful symptoms of diabetic sensorimotor polyneuropathy.

INTRODUCTION

The effects of diabetes mellitus on the neuromuscular system are widespread, but the common clinical manifestations, including peripheral sensorimotor polyneuropathy, small fiber neuropathy, mononeuropathy, lumbosacral radiculoplexus neuropathy, and autonomic neuropathy, involve primarily the peripheral nervous system. Although research studies have demonstrated that diabetes mellitus affects skeletal muscle metabolism and function, a presentation of clinical myopathy is not typical of patients with diabetes mellitus. In fact, the presentation of clinical myopathy in a diabetic patient should alert the treating physician to an alternate diagnosis, such as statin-induced myopathy, polymyositis, muscular dystrophy, or inclusion body myositis, among others. The most common form of neuropathy in diabetes mellitus is diabetic sensorimotor polyneuropathy (DSP), and, in fact, DSP is the most common form of polyneuropathy in the world today. A diagnosis of DSP should be considered in anyone presenting to the neurologist with polyneuropathy, and appropriate investigation to exclude diabetes mellitus or prediabetes, including a 2-hour glucose tolerance test, should be done. DSP leads to insensate feet, foot ulceration, gangrene and amputation, and sensory ataxia.1 Patients may present with isolated small fiber involvement and neuropathic pain early in the course of diabetic neuropathy or even in the prediabetic state.2,3 Involvement of autonomic nerve fibers occurs commonly as part of DSP, producing erectile dysfunction, cardiac dysfunction, orthostatic hypotension, bladder dysfunction, gastrointestinal dysmotility, and disorders of vision and sweating. At times, autonomic neuropathy can present without major evidence of DSP. This article discusses the diabetic peripheral neuropathies in the most detail because that is what most neurologists will observe in their daily practice.

SCOPE OF THE PROBLEM

As a result of the epidemic of diabetes mellitus in the developed world and its increasing prevalence in developing countries, DSP now presents a global burden as the most common form of polyneuropathy. Currently, 8.3% of the population has diabetes mellitus, and, in those aged 65 years and older, the prevalence is 26.9%.4 Furthermore, 35% of those aged 20 years or older have prediabetes.4 It is estimated that 1 in 3 people in the United States will have diabetes mellitus by 2050 if the current increase in prevalence continues. In adults with diabetes mellitus, 95% have type 2 diabetes mellitus.4 About 50% of patients with diabetes mellitus develop DSP during the course of their illness, with significant morbidity and mortality.5,6,7,8,9 Furthermore, no disease-modifying treatment is available for DSP. Strict glycemic control,10,11,12 maintained indefinitely, and attention to potentially modifiable risk factors (such as hypertension, hyperlipidemia, smoking, and high body mass index [BMI]) may help prevent the progression of DSP,7,13 but so far no effective therapy that leads to regression of this disorder exists. Even pancreatic transplant in patients with type 1 diabetes mellitus does not reverse DSP, although there is some evidence of limited nerve regeneration using novel measurement techniques.14,15 Other forms of diabetic neuropathy—particularly the focal neuropathies such as lumbosacral radiculoplexus neuropathy—recover spontaneously, although the course may be prolonged. Alternatively, patients with entrapment neuropathies such as carpal tunnel syndrome may benefit from surgical intervention. The autonomic neuropathies, particularly cardiac autonomic neuropathy, pose a threat to life and function; for example, cardiac autonomic neuropathy increases the risk of mortality.16 A simple classification scheme of diabetic neuropathy is shown in Table 1-1.

Table 1-1.

Types of Diabetic Neuropathy in Order of Decreasing Frequency

DIAGNOSIS

Screening

The diagnosis of diabetic neuropathies can be challenging because of the diverse manifestations. Also, sophisticated diagnostic techniques continue to evolve.17,18 Simple screening instruments, such as the monofilament examination, can identify DSP in the endocrinologist’s or primary care physician’s office,19 and the test results have prognostic implications for the development of DSP.20 These simple tests of sensation at the toes identify subjects at high risk for having or developing polyneuropathy and are easily and quickly performed. Consequently, such tests can form part of the routine screening for DSP as recommended in various professional guidelines.21,22 A simplified scoring system, such as the Toronto Clinical Neuropathy Score (a summed score of abnormal symptoms [pain, tingling, numbness, ataxia, weakness, upper limb symptoms], knee and ankle reflex scores, and sensory examination tests [light touch, pinprick, temperature, vibration and position sensations] considered to be due to DSP), can identify patients with a high probability of having this condition.23,24 However, neurologists need to perform a full evaluation for peripheral neuropathy.25

Clinical Evaluation

A complete neurologic history and examination are required, with attention to historical factors based on the type and duration of diabetes mellitus, the level of glycemic control, the presence of other complications such as retinopathy or nephropathy, and the presence of hyperlipidemia and hypertension. Furthermore, to exclude non–diabetes-related causes of peripheral polyneuropathy, a thorough patient and family history of nutritional deficiencies, neoplastic processes, paraproteinemias, and kidney disease, and a patient history of exposures to toxins such as alcohol and neurotoxic drugs (eg, amiodarone, vinca alkaloids, diphenylhydantoin) needs to be ascertained. This approach will help exclude other diagnoses in the differential of DSP because diabetic neuropathy is a diagnosis of exclusion (Case 1-1).21 Furthermore, a history concerning autonomic function should be ascertained. Questions about erectile function, possible orthostatic hypotension, arrhythmia, bowel or bladder disturbances, sweating irregularities, and gastrointestinal dysmotility problems need to be part of the symptom screen. In addition to a detailed neurologic examination (with particular attention to the patient’s peripheral sensory system, deep tendon reflexes, and strength), the morphology of the feet needs to be considered and the presence of ulcerations or amputations noted. The peripheral pulses should be assessed. If there is any concern about possible syncopal or presyncopal events, supine and standing blood pressure and heart rate should be measured. The pupillary reactions to light and accommodation should be noted. The state of the skin in the lower extremities needs to be assessed for the presence or absence of hair and the presence of trophic changes (such as thin, shiny, and discolored skin) that may signal the presence of polyneuropathy. The range of spinal motion should be evaluated, as well as gait and posture.

Case 1–1

A 58-year-old mechanic presented with a 2-year history of unsteadiness and numbness and tingling in his feet. His sensory symptoms progressed to above the ankles. He had two falls in the past 6 months. He denied any pain, weakness, muscle cramping, or other neurologic symptoms other than erectile dysfunction. He did not know of any medical illness in his history and rarely sought medical attention. He drank four beers daily, had a 30-pack/year history of smoking cigarettes, and led a sedentary lifestyle.

Examination showed a body mass index of 33, blood pressure 150/90 mm Hg supine and standing, and heart rate 85 beats/min supine and 90 beats/min standing. Ankle reflexes were absent, and he had loss of sensations of pinprick, temperature, vibration, and light touch to the ankles; he demonstrated a broad-based gait and was unable to tandem walk. The rest of the examination was normal.

Laboratory testing showed a normal vitamin B₁₂ level and normal results for serum immunoelectrophoresis and fasting blood sugar; an abnormal 2-hour glucose tolerance test; and a hemoglobin A_1c level of 7.5% (normal <6%).

The sural sensory nerve action potential amplitude was 4 µV (normal ≥ 6 µV) and the sural nerve conduction velocity was 38 m/s (normal ≥ 40 m/s). The peroneal compound muscle action potential amplitude was 1.5 mV (normal ≥ 5 mV), and the peroneal motor nerve conduction velocity was 39 m/s (normal ≥ 40 m/s).

The patient was diagnosed with diabetic sensorimotor polyneuropathy (DSP) and referred to an endocrinologist for treatment of hyperglycemia and advice on lifestyle modification, nutrition, exercise, and assessment of other potential risk factors. Given his history of regular ethanol intake and unsteadiness, a brain MRI was done to assess for the presence of midline (vermian) cerebellar degeneration, which was observed.

Comment. The neuropathy in this patient led to the diagnosis of type 2 diabetes mellitus. This patient had the classical presentation of DSP with the confirmatory axonal changes on nerve conduction studies. Given his degree of ataxia, concurrent disorders needed to be investigated and excluded, as DSP is a diagnosis of exclusion.

Laboratory Tests

After a clinical evaluation, laboratory testing should include at a minimum assessment of vitamin B₁₂ level, serum immunoelectrophoresis (to exclude monoclonal gammopathy), and hemoglobin A_1c.26 The hemoglobin A_1c will give an estimate of the average glycemic control for the past 3 months. Other tests directed to specific functions such as antinuclear antibody, rheumatoid factor, cryoglobulins, erythrocyte sedimentation rate, liver function tests, and kidney function, might need to be considered in particular circumstances. Genetic testing for possible hereditary neuropathies depends on the presentation of both the patient and family histories.26

If there is a concern about underlying neoplasm, then systemic imaging studies, such as abdominal ultrasound and CT scan of the chest, need to be considered. If there is a clinical concern for spinal stenosis or lumbar radiculopathies, then MRI of the spine needs to be considered.

Nerve conduction studies. Electrodiagnostic evaluation is necessary to diagnose and stage DSP reliably. DSP is the prototype of distal axonal neuropathy, and the evidence base suggests abnormality of nerve conduction studies as highly confirmatory for the diagnosis.25 The diagnosis of DSP for research purposes requires confirmation with objective tests such as nerve conduction studies. This recommendation has been endorsed in recent expert opinion.27 At a minimum, upper and lower limb nerve conduction studies, including motor and sensory nerves, need to be completed. The typical screening involves peroneal and tibial motor, sural sensory, and median motor and sensory nerve conduction studies. The addition of other nerve conduction studies (eg, ulnar, radial, or femoral) may be necessary depending on the patient’s clinical symptoms. EMG studies may help to determine whether the neuropathy is axonal and to exclude other disorders such as radiculopathy. The presence of DSP is predicted by a combination of peroneal nerve conduction velocity and sural sensory nerve action potential amplitude, whereas the future prediction of DSP is supported by tibial F-wave latencies, peroneal conduction velocity, and the sum of lower limb conduction velocities (sural, peroneal, and tibial).28 In patients with type 2 diabetes mellitus, demyelinating changes are unexpected and lead to alternative diagnoses such as chronic inflammatory demyelinating polyneuropathy.29 However, in patients with poorly controlled type 1 diabetes mellitus, with mean hemoglobin A_1c values of 9.5%, unexpected degrees of conduction velocity slowing suggestive of demyelination are observed, and these changes are indicative of poor control even in those with long-standing diabetes mellitus.30

Other objective tests (small fiber tests). In the absence of abnormal nerve conduction study testing, the diagnosis of DSP is less definite.25 In addition, nerve conduction studies are typically normal in isolated small fiber neuropathy, which can be the early manifestation of DSP. Other objective tests of peripheral nerve function may help provide confirmation of the diagnosis of a type of diabetic neuropathy, although most are still under investigation for usefulness (Table 1-2). One option is to perform quantitative sensory threshold testing, such as thermal threshold or vibration perception threshold testing.31 Assessment of vibration perception thresholds can be abnormal in patients with normal nerve conduction studies despite the fact that both are large fiber test modalities. This abnormality of large fiber sensory testing can be related to abnormality within the CNS affecting large fiber sensory tracts and is not specific for peripheral neuropathy. In such cases, additional investigations may be necessary. Testing of small fiber function, including cold detection and heat-as-pain thresholds, can show abnormality in the presence of normal nerve conduction studies.32 Other novel small fiber tests that may become useful are cutaneous axon-mediated flare laser Doppler imaging studies,33 corneal confocal microscopy,34,35 and intraepidermal nerve fiber density.26,36 These studies appear to be complementary in the diagnosis of DSP so that a single noninvasive small fiber test is insensitive for the diagnosis of DSP, whereas a combination of tests gives better diagnostic results.37 Intraepidermal nerve fiber density is invasive, expensive, and not done routinely. Only testing of thermal thresholds is readily available for clinical testing, and other noninvasive small fiber tests are still used mainly for research purposes.

Table 1-2.

Tests for Diabetic Sensorimotor Polyneuropathy

Autonomic system tests. Testing of the autonomic nervous system can be complex and is not done routinely. However, simple tests in the neuromuscular laboratory can assess autonomic activity in patients with diabetes mellitus.21 These include postural changes in blood pressure and heart rate, RR-variation, and sympathetic skin responses. A normal beat-to-beat heart rate variation with expiration and inspiration of more than 15 beats/min occurs. A reduction in normal RR-variation to fewer than 10 beats/min is abnormal, indicating vagal dysfunction, and provides early evidence of cardiac autonomic neuropathy that can be observed despite a lack of clinical abnormalities.21,38 The absence of the sympathetic skin response reflects the presence of DSP.39

Surrogate Markers for Diabetic Sensorimotor Polyneuropathy

Reliable and valid surrogate markers for DSP would be helpful in clinical trials to assess response to therapy and perhaps in the clinic to allow earlier diagnosis and interventions. DSP is indolent, and novel ways to measure disease activity might improve care delivery to affected patients. These methods generally measure small nerve fiber function and structure, as noted above. Serial punch biopsies of the skin for intraepidermal nerve density and morphology might be a useful surrogate for DSP,36,40 although this procedure is not yet standard practice and is invasive and expensive.26 Noninvasive measures, such as corneal confocal microscopy, might become useful.34,35,41 None of these measures is as robust as nerve conduction studies currently, and even the older method of quantitative sensory threshold testing is not as reliable as nerve conduction studies.32,42,43 Nonetheless, it is important to develop ways to measure small nerve fiber function and structure as involvement of small nerve fibers occurs earliest in DSP, and some patients have minimal or no demonstrable large fiber abnormalities. Finally, nerve biopsy sampling of large nerves such as the sural nerve is not indicated in the routine diagnosis of DSP.

The diagnosis of mononeuropathies is straightforward, involving symptoms and signs within the territory of a single peripheral nerve, such as the median, ulnar, radial, or peroneal nerves. For cranial mononeuropathies, the differential diagnosis is crucial. For example, in diabetes mellitus, a pupillary-sparing third nerve palsy is typically observed. Involvement of the pupil should lead to a search for an aneurysm of the posterior communicating artery compressing the nerve, or some other compressive or intrinsic third nerve lesion. In this case, appropriate diagnostic imaging tests, such as MRI, magnetic resonance angiography, or conventional angiography, should be done. It is probable that noncompressive mononeuropathies are microvascular in nature and produce nerve ischemia. In the case of cranial mononeuropathies, spontaneous recovery is the rule.

Diabetic Lumbosacral Radiculoplexus Neuropathy

Diabetic lumbosacral radiculoplexus neuropathy presents with acute and severe proximal lower limb pain, followed by muscle weakness and atrophy. It is most commonly observed in patients with well-controlled type 2 diabetes mellitus. The clinical presentation often involves the distribution of the femoral nerve, which led to the former names of “diabetic amyotrophy” or “femoral neuropathy.” However, the involvement is more widespread in most cases, leading to the newer nomenclature of diabetic lumbosacral radiculoplexus neuropathy. The symptoms can spread to the contralateral limb, and weight loss can be observed, as can upper limb changes. Although the course can be prolonged for months with major disability and limitations in ambulation, spontaneous recovery is observed in most cases. Microscopic vasculitis producing nerve ischemia may be the underlying pathophysiology of the disorder.44 Perivascular epineural inflammation and infiltration of adjacent endoneurium have been observed, and epineurial microvasculitis without vessel damage has been described in others. Some pathologists question the significance of such microvasculitis. Evidence of ischemic infarcts is demonstrable in some patients but not in others.44 It is important to exclude other diagnoses, such as intraspinal processes with polyradicular involvement or compressive causes of plexopathy, and appropriate investigations, such as MRI scanning of the lumbosacral spine and pelvis, may need to be done.

Thoracoabdominal radiculopathy is thought to be a focal radiculopathy analogous to diabetic lumbosacral radiculoplexopathy affecting nerve roots in the thoracic or abdominal dermatomes. The patients present with acute radicular pain in the thoracoabdominal region, and investigations such as spinal MRI do not show a structural cause. It is important to exclude intrathoracic or intra-abdominal causes of the pain with appropriate investigations, and spontaneous resolution is the typical outcome.

Diabetic cachexia is a rare disorder manifested by severe diffuse pain and paresthesia, sensorimotor polyneuropathy, and weight loss in diabetic patients that can be precipitated by rapid glycemic control with insulin therapy. Symptomatic recovery is associated with recovery of nerve function, indicating that the pathogenesis is likely due to nerve hypoxia but not irreversible ischemic destruction.45

TREATMENT

For many mononeuropathies, spontaneous resolution is the expected course. In the case of carpal tunnel syndrome, nerve conduction studies cannot distinguish reliably between the presence of entrapment of the median nerve at the carpal tunnel and DSP.46 Similar changes in the median nerve conduction tests are observed in patients who have diabetes mellitus with and without the clinical features of carpal tunnel syndrome, and the severity of the changes is related to the severity of DSP but not to the clinical features of carpal tunnel syndrome. For this reason, diagnosis and management should be based on the clinical presentation. Conservative measures may fail to relieve the compression, and surgical release should be considered for both carpal tunnel syndrome and cubital tunnel syndrome, although the outcomes may not be as good as in nondiabetic patients. In diabetic lumbosacral radiculoplexus neuropathy, spontaneous resolution is expected. No current clinical trial evidence supports the use of immune therapies in diabetic lumbosacral radiculoplexus neuropathy, so IV immunoglobulin, plasma exchange, or immunosuppressant medications are not recommended.47,48 Physiotherapy, analgesia, and supportive therapy are recommended for this condition.

In small fiber neuropathy—likely to be the earliest manifestation of DSP as well as the form observed in prediabetes—lifestyle modification may improve nerve function and symptoms, although the evidence is variable.49,50 The mainstay of treatment in patients with DSP remains strict glycemic control maintained indefinitely. In patients with type 1 diabetes mellitus, good evidence has shown that intensive glycemic control minimizes the incidence of neuropathy,10,51 and there appears to be a metabolic memory effect in that the beneficial effect of a 5-year period of intensive diabetes control persists for more than 10 years after the end of that treatment.11,52 In patients with type 2 diabetes mellitus, the beneficial outcomes are less apparent, as strict glycemic control can be associated with negative effects such as increased mortality and no definite change in DSP.53,54,55 However, the findings from a recent Cochrane review support enhanced glucose control in both type 1 and type 2 diabetes mellitus to improve nerve function: in type 1 diabetes mellitus to prevent the development of clinical neuropathy and in type 2 diabetes mellitus because of a possible tendency to reduce the incidence of clinical neuropathy (Case 1-2).56 Attention to other modifiable risk factors, such as hyperlipidemia, hypertension, and high BMI, may ameliorate DSP, particularly in patients with type 1 diabetes mellitus.13 Finally, diet and exercise counseling in patients with prediabetes (impaired glucose tolerance) may help reduce symptoms and lead to peripheral nerve reinnervation.49 A hemoglobin A_1c level under 7% in type 1 diabetes mellitus would minimize progression of DSP, and a level of less than 7.5% might be beneficial in type 2 diabetes mellitus, but these types of management decisions need to be formulated by the primary care physician and endocrinologist providing diabetes care to these patients.51,55 Surveillance and attention to foot care are needed to prevent foot ulceration and amputation.

No effective disease-modifying therapies are available for DSP. Many drugs that appeared promising in phase 2 trials have failed to demonstrate efficacy in phase 3 trials, or results from different phase 3 trials have been contradictory.5,57 Various interventions are believed by some researchers to have disease-modifying activity in DSP, but insufficient evidence exists for any intervention at this stage. In fact, the same interventions can have variable evidence, with positive studies in some regions of the world and negative studies in other regions of the world.

Although no specific disease-modifying therapy for DSP exists, several interventions for neuropathic pain in DSP have been shown to be effective. A recent evidence-based practice parameter developed by the American Academy of Neurology, the American Association of Neuromuscular and Electrodiagnostic Medicine, and the American Academy of Physical Medicine and Rehabilitation showed benefits from some anticonvulsants (eg, pregabalin, gabapentin), antidepressants (eg, amitriptyline, duloxetine), and opioids (eg, morphine sulphate, oxycodone).58 See Table 1-3 for a complete listing and doses used in the supporting clinical trials. Similar results have been published by other groups.22 Other interventions with good evidence include long-acting capsaicin and transcutaneous electrical nerve stimulation. High-quality studies are either lacking or negative in many alternative treatments of DSP, such as acupuncture or Reiki therapy.58,59 However, the placebo effect is very high in patients who have painful DSP, accounting for up to 62% of the response,60 so that any safe and inexpensive treatment modality should not be dismissed out of hand if the patient reports benefit from that treatment. There are several concerns with the existing evidence base for the treatment of painful DSP.58 This is a chronic disorder, and many of the interventions have been studied only for short intervals of 4 to 12 weeks; therefore, the long-term efficacy of these interventions remains unknown. Furthermore, high-level comparative studies of different pharmacologic therapies are still required. In addition, it is rare that painful symptoms are completely eliminated. Rather, a 30% reduction in pain is considered a good outcome as this degree of pain relief relates to patients reporting that they feel much improved, and those with a 50% reduction in pain state that they are very much improved.61

Table 1-3.

Symptomatic Treatment of Painful Diabetic Neuropathy^a

For those with autonomic neuropathy, exclusion of organ-based disease is required, and often patients need to be referred to the appropriate specialist in that field (eg, gastroenterologist, urologist, cardiologist) for investigation and management.

One other entity that neurologists should keep in mind is insulin neuritis, also known as treatment-induced diabetic neuropathy. This is the development of neuropathic symptoms (often painful peripheral paresthesia) starting soon after the initiation of insulin therapy and thought to be related to a rapid normalization of nerve glucose levels. Autonomic neuropathy accompanies insulin neuritis with cardiovascular, gastrointestinal, genitourinary, and sudomotor symptoms.62 Spontaneous recovery is observed after about 18 months of glycemic control. This entity differs from diabetic cachexia in that weight loss is not a feature of insulin neuritis.

Case 1–2

A 68-year-old man presented with an 18-month history of burning pain along the soles of his feet that he rated as 8/10.The pain was worse when he was trying to rest and also interfered with his sleep. He had erectile dysfunction for 1 year but was free of other symptoms. Diabetes mellitus was diagnosed 3 years ago, and he started a diet and metformin. He had hypertension and hyperlipidemia. He was a nondrinker and nonsmoker without a history of toxic exposures or familial neuropathy.

Examination showed a blood pressure of 160/90 mm Hg supine and 155/85 mm Hg standing, with a heart rate of 90 beats/min supine and 100 beats/min standing. His body mass index (BMI) was 35. The peripheral pulses and skin in the lower limbs were normal. He had blunting of sensations of pinprick, temperature, vibration, and light touch distally in the toes. His reflexes and the rest of the neurologic examination were normal.

He had normal nerve conduction studies and elevated cold-detection thresholds. Laboratory tests showed a hemoglobin A_1c level of 8.5% and normal vitamin B₁₂ and serum immunoelectrophoresis.

The patient was diagnosed with small fiber diabetic polyneuropathy. He was advised to increase exercise, reduce caloric intake (reduce BMI), and aim for a hemoglobin A_1c level less than 7%. Control of hypertension and hyperlipidemia were recommended. Pregabalin 75 mg at bedtime was initiated, with the recommendation for a slow upward titration to a maximum dose of 300 mg twice daily. After 2 weeks, his pain was 3/10, and he slept better and was happy with this outcome.

Comment. In this case, early diabetic sensorimotor polyneuropathy presented as a small fiber neuropathy. The presence of normal ankle reflexes and nerve conduction studies do not exclude this diagnosis, and a small fiber test showed neuropathy. Symptomatic treatment reduced the patient’s pain to a manageable level but did not completely eliminate the pain.

SUMMARY

DSP remains the most common neurologic complication of diabetes mellitus. Optimal glycemic control and attention to hypertension, hyperlipidemia, BMI, and foot care are fundamental in those with type 1 diabetes mellitus. The benefits of strict glycemic control are not as evident in those with type 2 diabetes mellitus, although lifestyle modification may be of benefit. No disease-modifying intervention has proven effective in reversing DSP or even in slowing the progression, beyond pancreatic transplantation or intensive glycemic control in type 1 diabetes mellitus. In prediabetes, attention to lifestyle modification may modify polyneuropathy. Attention should be paid to foot care in order to avoid the complications of foot ulceration and amputation. Symptom control for painful symptoms of DSP is available, and good evidence supports treatment with several anticonvulsants, antidepressants, and opioids.

KEY POINTS

• Diabetes mellitus produces clinical peripheral neuropathies as the major neuromuscular complication; symptoms suggestive of myopathies or myoneural junction disorders should raise other diagnostic concerns.

• An epidemic of type 2 diabetes mellitus related to lifestyle predicts that one-third of adults will have this disorder, and one-half of these will develop diabetic sensorimotor polyneuropathy during their lifetime. No effective disease-modifying agent is available, and only strict glycemic control may help prevent progression of neuropathy in patients with type 1 diabetes mellitus.

• Regular screening for diabetic neuropathy is recommended in guidelines from professional bodies such as the American Diabetes Association and the Canadian Diabetes Association.

• Diabetic neuropathy is a diagnosis of exclusion.

• Nerve conduction studies remain essential in the diagnosis and staging of diabetic neuropathy for clinical research.

• The diagnosis of diabetic sensorimotor polyneuropathy for research purposes requires confirmation with objective tests such as nerve conduction studies. This recommendation has been endorsed in recent expert opinion.

• Demyelinating changes in the nerve conduction studies should lead to suspicion of poor glycemic control in patients with type 1 diabetes mellitus and of the presence of chronic inflammatory demyelinating polyneuropathy in patients with type 2 diabetes mellitus.

• Nerve conduction studies are normal in small fiber neuropathy, an early stage of diabetic sensorimotor polyneuropathy.

• If nerve conduction studies are normal in suspected diabetic neuropathy, then specific small fiber function tests should be considered in order to confirm peripheral neuropathy.

• Simple tests of autonomic function should be considered in patients with possible diabetic sensorimotor polyneuropathy.

• Surrogate measures for diabetic sensorimotor polyneuropathy include small fiber tests such as intraepidermal nerve fiber density, corneal confocal microscopy, quantitative thermal threshold testing, and cutaneous axon-mediated flare laser Doppler imaging studies.

• Intensive glycemic control helps minimize diabetic sensorimotor polyneuropathy in type 1 diabetes mellitus, and improved glycemic control is the only option also for patients with type 2 diabetes mellitus. No other disease-modifying treatment is available. Control of hypertension, hyperlipidemia, and elevated body mass index may be helpful.

• Control of painful symptoms can be achieved with pregabalin, duloxetine, gabapentin, venlafaxine, tricyclic antidepressants, topical capsaicin, and transcutaneous electrical nerve stimulation. Many other symptomatic treatments are ineffective or lack evidence.

• A high placebo effect is present in patients with painful diabetic sensorimotor polyneuropathy, so inexpensive and safe treatments may be considered if they provide potential benefit, although evidence for their efficacy is lacking.

• Painful symptoms in diabetic sensorimotor polyneuropathy are rarely eliminated by any treatment modality, but pain relief of at least 30% is rated as much improved and of 50% as very much improved by patients.