INTRODUCTION — Distal symmetric polyneuropathy (DSPN), often considered synonymous with the term "diabetic neuropathy," is the most common neurologic complication of diabetes and a major cause of morbidity. Diabetic neuropathy leads to gradual loss of integrity of the longest nerve fibers, with symptoms beginning distally and symmetrically in the toes and feet. In addition to neurologic disability related to sensory loss and risk of foot ulcers and amputations, approximately 15 to 20 percent of patients have painful symptoms that can further limit function and decrease quality of life.
Patients with diabetic neuropathy should be treated with a systematic, stepwise approach that includes glycemic control and control of the metabolic syndrome, education and counseling on foot care and safety measures, and symptomatic treatment of pain, when present.
Management of DSPN is reviewed here. The evaluation, diagnosis, and management of other forms of neuropathy in patients with diabetes, including autonomic neuropathy, are reviewed in detail separately. (See "Epidemiology and classification of diabetic neuropathy" and "Screening for diabetic polyneuropathy" and "Diabetic autonomic neuropathy".)
PREVENTIVE CARE — Established, symptomatic diabetic neuropathy is generally not reversible, and management aims to slow further progression and prevent complications, including diabetic foot ulcers, arthropathy, and falls.
Glycemic control — The role and importance of glucose control for slowing the progression of neuropathy and other microvascular complications of diabetes vary by type of diabetes.
In patients with type 1 diabetes, high-quality evidence supports the recommendation to aim for stable and optimal glycemic control in the management of patients with diabetic neuropathy [1]. Long-term follow-up of the landmark Diabetes Control and Complications Trial (DCCT) demonstrated that more intensive glucose control ameliorated the onset of neuropathy as well as progression of surrogate electrophysiologic markers of neuropathy (figure 1) [2,3]. In a meta-analysis of DCCT and one additional trial in a total of 1228 patients with type 1 diabetes, enhanced glucose control lowered the annualized risk of neuropathy (absolute risk difference -1.8 percent, 95% CI -2.6 to -1.1), the incidence of clinical neuropathy after five years (8 versus 17 percent; relative risk [RR] 0.46, 95% CI 0.33-0.63), and the progression of surrogate measures of neuropathy, including nerve conduction velocity and vibration perception thresholds [4]. (See "Glycemic control and vascular complications in type 1 diabetes mellitus".)
In patients with type 2 diabetes, glucose control has a more modest effect on the course of neuropathy. A meta-analysis of four trials in 6669 patients with type 2 diabetes found a nonsignificant trend toward a small reduction in the annualized risk of neuropathy with enhanced glucose control (-0.58 percent, 95% CI -1.4 to 0.01) [4]. The lack of a larger independent effect of glucose control may reflect a difference in the pathophysiology of neuropathy in type 2 diabetes, which may relate to components of the metabolic syndrome collectively [5]. As in all patients with type 2 diabetes, treatment should focus on multiple patient-centered targets to normalize lipids, blood pressure, weight, and glucose, along with lifestyle modifications to achieve a healthy diet and regular exercise [1]. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus", section on 'Multifactorial risk factor reduction'.)
Risk factor modification — Lifestyle interventions are endorsed as an essential practice to prevent onset and progression of neuropathy, particularly in individuals with prediabetes and type 2 diabetes [1,6]. Goals include achieving a normal bodyweight and attaining individualized glycemic, blood pressure, and lipid goals along with 150 minutes of moderate-to-vigorous aerobic activity and two to three sessions of resistance training weekly [7]. In a systematic review of 20 clinical trials, exercise was associated with improvements in function, nerve conduction velocities, and glycemic control in patients with diabetic neuropathy [8]. (See "Nutritional considerations in type 1 diabetes mellitus" and "Nutritional considerations in type 2 diabetes mellitus" and "Exercise guidance in adults with diabetes mellitus".)
Available data suggest that surgical treatment of type 2 diabetes (ie, bariatric surgery) has the potential to reduce neuropathy incidence along with other microvascular complications [9,10]. In a retrospective case-control study of over 15,000 obese adults with type 2 diabetes, bariatric surgery was associated with a lower rate of neuropathy at five years (7 versus 21 percent) [10]. Rates of retinopathy and nephropathy were also lower in the surgical patients. The outcomes of bariatric surgery are reviewed in more detail separately. (See "Outcomes of bariatric surgery".)
Vascular risk factor treatment is also endorsed to slow or prevent the progression of diabetic neuropathy, including improved control of blood pressure, lipids, avoidance of cigarette smoking, and avoidance of excess alcohol consumption [1]. (See "Overview of general medical care in nonpregnant adults with diabetes mellitus".)
Foot care — Peripheral neuropathy is one of the most important risk factors for ulcers and amputations in patients with diabetes. Foot care in patients with neuropathy is essential to help lower risk of complications.
On a daily basis, patients should inspect their feet for the presence of dry or cracking skin, fissures, plantar callus formation, and signs of early infection between the toes and around the toenails. Regular clinical foot examinations to detect peripheral neuropathy are also an essential component of preventive care in all patients with diabetes. (See "Evaluation of the diabetic foot" and "Management of diabetic foot ulcers".)
Safety and falls — Patients with diabetic neuropathy are at increased risk for gait instability and falls due to progressive loss of proprioception, foot pain, orthostatic hypotension due to autonomic dysfunction, age-related functional impairments, and medication side effects [1]. Although not well studied in this specific patient population, multifaceted interventions including home-based exercise, physical and occupational therapy, and home safety evaluation have been shown to reduce fall risk in older adults and should be considered for patients with gait and balance abnormalities on neurologic examination (algorithm 1). (See "Falls: Prevention in community-dwelling older persons", section on 'Preventing falls'.)
Additional safety precautions in patients with diabetic neuropathy may include handlebars in the bathroom and shower, use of nightlights to improve visibility, and testing water temperature with an elbow or uninvolved body part before entering a bath or shower to avoid scalding injury.
B12 deficiency — Metformin reduces intestinal absorption of B12, and the prevalence of borderline or low B12 levels in metformin-treated patients approaches 20 percent at five years [11]. Patients on metformin with worsening neuropathy symptoms should be screened for B12 deficiency, particularly if examination findings are atypical or suddenly worsening. (See "Metformin in the treatment of adults with type 2 diabetes mellitus", section on 'Vitamin B12 deficiency'.)
PAIN MANAGEMENT — Approximately 15 to 20 percent of patients with diabetic neuropathy have pain in the feet, often described as burning or stabbing, as a symptom of small myelinated fiber involvement (table 1) [12,13]. While pain may be self-limited and spontaneously resolve within a year of onset in up to half of patients [14], others have persistent symptoms and disability related to pain. Symptomatic therapies for neuropathic pain are an important component of care in such patients. Pain medications are not useful for nonpainful symptoms of neuropathy, such as numbness.
Initial pharmacotherapy
Choice of agent — First-line pharmacotherapy options for painful diabetic neuropathy include several antidepressants (eg, duloxetine, venlafaxine, amitriptyline and other tricyclic drugs) and the gabapentinoid antiseizure medications (pregabalin, gabapentin) [15-17]. Capsaicin can also be used but is generally poorly tolerated.
The available evidence suggests that all are better than placebo in patients with diabetic neuropathy and that comparative efficacy is similar, although few high-quality comparative trials have been done [18-21]. Side effects, dosing frequency, cost, and regulatory approval status may vary. In the United States, three of these medications are approved by the US Food and Drug Administration (FDA) for the indication of diabetic neuropathy (duloxetine, pregabalin, and the capsaicin patch [8%]).
Selection of a specific agent should therefore be individualized based on comorbidities (table 2), drug interactions, side effect profiles, and cost. In practice, comorbidities and concurrent medications often favor one class or another (ie, an antidepressant or a gabapentinoid) in an individual patient, and then the narrower choice within a class is influenced by the patient's age, preferences with regard to dosing frequency, side effects, and cost/formulary considerations.
●When comorbidities favor an antidepressant – Among the first-line options, a serotonin-norepinephrine reuptake inhibitor (SNRI; duloxetine or venlafaxine) is a common choice for patients of any age who are not already taking an antidepressant, as these two drugs are generally well tolerated and taken once daily. Some clinicians prefer duloxetine over venlafaxine based on FDA approval status, but most consider them interchangeable when confronted with cost/formulary differences. For younger patients without a cardiac history, a tricyclic antidepressant (TCA) is considered an equally reasonable alternative to an SNRI and may be preferred in a patient with difficulty sleeping, for example, as the drowsy side effects of the TCAs may alleviate insomnia.
●When comorbidities favor a gabapentinoid – When a gabapentinoid is favored based on a relevant comorbidity (eg, restless legs syndrome), the choice between pregabalin and gabapentin is also individualized. The two drugs are similar in their side effect profiles and requirement for renal dosing; differences include FDA approval status for diabetic neuropathy, dosing frequency, and sometimes cost.
Administration and efficacy
Duloxetine — Duloxetine, an SNRI, is a first-line treatment option for neuropathic pain in patients with diabetic neuropathy [1,17,22]. It also has efficacy for the treatment of depression, anxiety, and fibromyalgia. The pain modulatory effect is thought to be mediated primarily by synaptic norepinephrine and effects on central perception of pain [23].
The recommended starting dose of duloxetine in patients with diabetic neuropathy is 20 to 30 mg daily [1]. The dose can be gradually titrated upwards based on response and tolerability to a maximum of 60 to 120 mg daily. Pain improvement may be seen as early as the first week of treatment [24-26].
Patients are encouraged to take the drug on a full stomach to reduce risk of nausea. Duloxetine should not be taken with other serotonin or norepinephrine uptake inhibitors (including tricyclic drugs) but can be combined with pregabalin or gabapentin.
Common side effects of duloxetine include nausea, somnolence, dizziness, decreased appetite, constipation, diaphoresis, and sexual dysfunction. Duloxetine and other SNRIs can exacerbate restless legs syndrome, which is common in patients with peripheral neuropathy and can further disrupt sleep. (See "Clinical features and diagnosis of restless legs syndrome and periodic limb movement disorder in adults" and "Serotonin-norepinephrine reuptake inhibitors: Pharmacology, administration, and side effects", section on 'Duloxetine'.)
A 2014 systematic review identified eight randomized trials of duloxetine for painful diabetic neuropathy in a total of 2728 patients [27]. For duloxetine 60 mg daily, a 50 percent or greater reduction in pain at 12 weeks was more likely with duloxetine than placebo (45 versus 26 percent; relative risk [RR] 1.73, 95% CI 1.44-2.08). In trials that included an 11-point Likert score scale to measure pain, duloxetine reduced the mean pain score by approximately 1 point more than placebo by 12 weeks. Across trials for multiple painful conditions, drug discontinuation due to adverse effects was twice as likely for duloxetine compared with placebo (RR 1.96, 95% CI 1.6-2.37). The 120 mg daily dose was similarly effective but not as well tolerated as 60 mg daily.
Venlafaxine — Venlafaxine is an SNRI with a similar spectrum of clinical activity as duloxetine. Although not as well studied as duloxetine in patients with diabetic neuropathy, the available data suggest that venlafaxine is a reasonable alternative to duloxetine for this purpose [1].
The recommended starting dose of venlafaxine in patients with diabetic neuropathy is 37.5 mg daily, titrated gradually upwards to a usual effective dose range of 75 to 225 mg daily [1]. In the largest randomized trial, pain scores declined gradually from week 1 to 6, reaching a maximum reduction at week 6 [28].
Common side effects of venlafaxine include nausea, somnolence, dizziness, dyspepsia, insomnia, diaphoresis, and sexual dysfunction. Increased blood pressure and heart rate occur more often with venlafaxine than placebo. Like duloxetine, venlafaxine can exacerbate restless legs syndrome and should not be taken with other serotonin or norepinephrine uptake inhibitors (including tricyclic drugs). (See "Clinical features and diagnosis of restless legs syndrome and periodic limb movement disorder in adults" and "Serotonin-norepinephrine reuptake inhibitors: Pharmacology, administration, and side effects", section on 'Venlafaxine'.)
A 2015 systematic review identified six randomized trials of venlafaxine versus placebo or an active comparator for neuropathic pain in a total of 460 patients, most having painful diabetic neuropathy [29]. All the included trials showed some benefit for venlafaxine, but all had limitations that could cause an overestimation of effect size. In the largest trial, extended-release (ER) venlafaxine was evaluated in 244 patients with painful diabetic neuropathy [28]. For venlafaxine ER at higher doses (150 to 225 mg daily), a 50 percent or greater reduction in pain at 12 weeks was more likely with venlafaxine than placebo (56 versus 34 percent). The effects of the 75 mg daily dose were smaller and not statistically significant.
Tricyclic drugs — Several TCA drugs (amitriptyline, desipramine, nortriptyline) are effective for neuropathic pain in patients with diabetic neuropathy and can be used cautiously for this purpose, with attention to their higher risk of serious side effects [1,17]. Tricyclics are thought to impact the central perception of pain via effects on synaptic monoamines. Patients should be screened for heart disease before use. (See "Tricyclic and tetracyclic drugs: Pharmacology, administration, and side effects", section on 'Baseline testing and monitoring for safety'.)
The starting dose of amitriptyline or nortriptyline is 10 to 25 mg/day with a gradual titration up to 100 mg/day if painful symptoms persist. Desipramine is begun at 25 mg/day, and the dose can gradually be increased to a maximum of 200 mg/day over a few weeks. Both amitriptyline and desipramine should be given at bedtime because of their sedating properties. The therapeutic effect of tricyclic drugs for pain usually occurs sooner (within six weeks) and at lower doses than is typical when these drugs are given for the treatment of depression.
Common side effects of TCAs include dry mouth and somnolence. Urinary retention may occur, especially those with enlarged prostates. We frequently substitute nortriptyline for amitriptyline if anticholinergic side effects are a problem, and some experts use nortriptyline as first line because it has fewer anticholinergic side effects than amitriptyline. Importantly, TCAs are contraindicated in patients with cardiac disease (either conduction system or ischemic). Similar to the SNRIs, they should not be given in combination with other serotonergic drugs.
Supporting data for tricyclic drugs include results of several randomized controlled trials in patients with painful diabetic neuropathy [19,30-33]. In part because these are older drugs, the number and size of the trials are generally small, and amitriptyline is the most studied agent. Examples of comparative trials illustrating the effects and tolerability of tricyclics in the diabetes population include the following:
●In two randomized blinded crossover trials in 84 patients with painful diabetic neuropathy, amitriptyline and desipramine were equally effective and superior to fluoxetine or placebo [31]. The benefit of the tricyclic drugs was noted within two weeks and continued to increase at six weeks. Desipramine had somewhat fewer side effects than amitriptyline, particularly dry mouth. The average effective dose, titrated over six weeks to achieve control of symptoms, was 111 mg/day for desipramine and 105 mg/day for amitriptyline. There was no correlation between relief of pain, dose, or plasma drug concentrations, suggesting that the clinical response and tolerability of side effects are the best guides to dose titration.
●A randomized, blinded crossover trial in 58 patients with painful diabetic neuropathy that compared amitriptyline (10 to 50 mg daily) and duloxetine (20 to 60 mg daily) given at bedtime found an improvement in pain with both medications compared with pretreatment baseline [19]. A good outcome, defined as a median pain score reduction of >50 percent, was reported at a similar rate for amitriptyline and duloxetine (55 versus 59 percent). Dry mouth was more frequent with amitriptyline compared with duloxetine (55 versus 24 percent), while constipation was nonsignificantly more frequent with duloxetine (37 versus 17 percent).
Pregabalin — Pregabalin is an effective first-line alternative to the antidepressant medications reviewed above in patients with painful diabetic neuropathy [1,17]. It is an alpha-2-delta ligand antiseizure medication that is structurally related to gabapentin. The pain modulatory effect is thought to relate to presynaptic inhibition of the release of excitatory neurotransmitters including glutamate, substance P, and calcitonin gene-related peptide (CGRP) [34,35].
A typical starting dose of pregabalin for diabetic neuropathy is 75 to 150 mg per day. The immediate-release formulation is given in two or three divided doses, and a once-daily extended-release preparation is also available (165 mg of extended-release is equivalent to 150 mg of immediate-release). Most trials have used a starting dose of 150 mg per day [36]. The dose can be titrated in increments of 75 mg per day every three to seven days. A total daily dose of 300 mg is the maximum dose approved by the FDA for diabetes-associated neuropathic pain, although some patients may require 450 mg per day. The FDA has approved doses of pregabalin of up to 600 mg per day for other indications. Higher doses are generally less well tolerated and may have limited additional efficacy.
Common adverse effects include dizziness, somnolence, peripheral edema, and weight gain. The incidence of clinically meaningful weight gain (defined as a ≥7 percent weight increase from baseline to endpoint) was approximately 2 to 4 percent in trials of pregabalin in patients with diabetic neuropathy, without effects on glucose control [36].
A 2019 systematic review identified eight randomized trials of pregabalin for painful diabetic neuropathy in a total of 2320 patients [37]. For pregabalin 300 mg per day, a 50 percent or greater reduction in pain was more likely with pregabalin than placebo (31 versus 24 percent; RR 1.3, 95% CI 1.2-1.5). A higher dose (600 mg per day) was studied in two trials and was also more effective than placebo (41 versus 28 percent). There were no differences in the rate of discontinuation for adverse effects at any dose level compared with placebo. In a separate meta-analysis, the median time to a sustained 1-point improvement on an 11-point pain score for pregabalin (at 150, 300, and 600 mg) and placebo was 13, 5, 4, and 60 days, respectively [36].
Gabapentin — Gabapentin has a similar spectrum of clinical activity and side effects as pregabalin and is a reasonable, sometimes lower-cost alternative to pregabalin when an antiseizure drug is favored for first-line therapy [1].
Typical starting doses for gabapentin are 100 to 300 mg one to three times daily; the drug can be titrated slowly up to a usual effective dose of 900 to 3600 mg daily in three divided doses [1]. The major side effects of gabapentin are somnolence, dizziness, and ataxia.
In a systematic review that identified six trials in a total of 1277 patients with painful diabetic neuropathy, the proportion of patients achieving at least a 50 percent pain intensity reduction was higher with gabapentin (dosed at ≥1200 mg daily) compared with placebo (38 versus 21 percent; RR 1.9, 95% CI 1.5-2.3) [38]. Individual trials were small but showed consistent effects.
Inadequate response to initial therapy — None of the first-line choices are universally effective or tolerated, even with appropriate titration. The time to peak response varies across medications, but, in general, a two- to three-month trial is typically required to titrate to effect and gauge response to initial therapy [17].
For patients who do not improve on the first medication with appropriate titration, we suggest either adding a second first-line drug from a different medication class or switching to an alternative medication among the first-line options discussed above [1]. A common combination strategy is pregabalin with duloxetine, although other combinations are reasonable as well depending on individual patient factors.
Results from multiple trials suggest that the treatment of neuropathic pain with combinations of drugs from different medication classes is modestly more effective than monotherapy. In a multicenter crossover trial of 130 patients with painful diabetic neuropathy that assessed the effect of multiple strategies for adding a second agent to initial monotherapy, combination therapy provided a greater reduction in pain than monotherapy [39]. Patients were given initial monotherapy with amitriptyline, pregabalin, or duloxetine. Those who did not respond to initial monotherapy at six weeks were given a second agent from a different pharmacologic class. Combination strategies consisting of pregabalin added to amitriptyline, amitriptyline added to pregabalin, and pregabalin added to duloxetine were all similarly effective in providing pain relief when assessed at 16-week follow-up. In another single-center trial of 47 patients, the combination of nortriptyline with gabapentin was also found to be more effective than either agent alone for reducing the mean intensity of daily pain during week 4 of treatment at the maximum tolerated daily dose (mean, nortriptyline 50 mg and gabapentin 2180 mg in combination) [40].
For patients who do not tolerate any of these medications or who prefer nonpharmacologic therapies, we discuss capsaicin cream, lidocaine patch, alpha-lipoic acid (ALA), transcutaneous electrical nerve stimulation (TENS), and spinal cord stimulation. (See 'Topical therapies or neuromodulation' below and 'Possibly effective' below.)
Topical therapies or neuromodulation — For patients who do not respond to or tolerate first-line medications for neuropathic pain, topical therapies and neuromodulatory approaches with some evidence of benefit in diabetic neuropathy include the following:
●Capsaicin – Capsaicin cream (0.075% applied topically four times daily) can be used on its own or added to first-line medications for patients with a partial response or intolerance of higher doses. It is also available in gel, liquid, and lotion formulations for topical application. Local burning and skin irritation can occur, but this becomes less of a problem with continued use. Nevertheless, many patients are unable to tolerate the local burning pain, which is exacerbated by contact with warm water and hot weather [22].
The high-concentration capsaicin (8%) patch is FDA approved to treat painful diabetic neuropathy at the feet [41]. Patches are administered by a health care professional as a single 30-minute application, and patients are monitored for up to two hours following treatment. The treatment may be repeated at three months for patients with partial or temporary relief after the initial application. Progressive improvements may be achieved with subsequent applications [42]. However, patients may report localized erythema, pain, or post-application sensory loss with use. Patches must be handled carefully, and the application site is usually pretreated with a local anesthetic such as topical lidocaine.
Capsaicin is a naturally occurring component of many hot peppers and causes analgesia through local depletion of substance P. In randomized trials in patients with painful diabetic neuropathy, capsaicin has been associated with modest but statistically significant improvement in pain compared with placebo [22,43-46].
●Lidocaine patches – Limited evidence suggests that application of lidocaine patches (5%) can improve pain in patients with painful diabetic neuropathy [47]. Patches may remain in place for no more than 12 hours in any 24-hour period.
●Electrical nerve stimulation – Although data are limited, a 2010 statement from the American Academy of Neurology (AAN) assessing the use of TENS for pain in neurologic disorders concluded that TENS is probably effective for reducing pain from diabetic neuropathy [48], based upon the following evidence.
•One trial assigned 31 patients with chronic painful diabetic neuropathy to either TENS or sham treatment to the legs for 30 minutes daily for four weeks [49]. Symptomatic improvement (of at least one grade on a unique 0 to 5 scale) occurred in 15 of 18 patients (83 percent) with TENS treatment, compared with 5 of 13 patients (38 percent) who received sham treatment (odds ratio 6, 95% CI 1.1-33.4) [49].
•Another trial evaluated 19 patients with mild to moderate symptomatic diabetic polyneuropathy [50]. Compared with sham treatment, active treatment with TENS led to a statistically significant reduction in total symptom score at 6 and 12 weeks. In addition, TENS therapy was associated with a statistically significant but modest improvement in pain on the visual analog scale at six weeks.
●Spinal cord stimulation – Small studies have reported symptom improvement with electrical pulsed stimulation of percutaneously implanted epidural electrodes corresponding to the distribution of pain [51-53]. In an open-label trial of 216 patients with refractory painful diabetic neuropathy treated with medical therapy, those assigned to high-frequency spinal cord stimulation were likelier to report at least 50 percent pain reduction at three months than those assigned to medical therapy alone [54]. A systematic review of three trials and five observational studies found moderate-quality evidence to support the use of spinal cord stimulation for pain relief in diabetic neuropathy of the lower extremities [55]. Future studies with effective sham controls are required to understand the therapeutic benefit of and help identify optimal patients for spinal cord stimulation.
●Acupuncture – Pilot studies indicate that acupuncture may be an effective nonpharmacologic option for painful neuropathy when delivered individually [56-58] or in a group setting [59]. A meta-analysis of 25 randomized trials in China concluded that acupuncture improves global symptoms of painful diabetic neuropathy compared with B vitamins (RR 1.3 to 1.6) or no treatment (RR 1.6), with no adverse events reported [60].
Other therapies
Possibly effective
●Alpha-lipoic acid (ALA) – For patients who are refractory to or intolerant of first-line pharmacotherapies and interested in a nutritional supplement approach, we suggest a trial of oral ALA (600 mg daily). ALA is an antioxidant with the potential to diminish oxidative stress, improve the underlying pathophysiology of neuropathy, and reduce pain. (See "Pathogenesis of diabetic polyneuropathy".)
Several prospective, placebo-controlled trials have examined ALA (either intravenous or oral) in patients with painful diabetic neuropathy [61-64]. In an initial trial, daily intravenous ALA for three weeks was associated with reduced pain, paresthesia, and numbness compared with placebo infusions [62]. In a second trial, oral ALA (600, 1200, or 1800 mg daily) versus placebo was studied in 181 patients with diabetes and symptomatic distal symmetric polyneuropathy (DSPN) [64]. All three doses of oral ALA treatment were associated with a reduction in the neuropathy total symptom score (a summation of stabbing pain, burning pain, paresthesia, and asleep numbness) compared with placebo. The benefit of ALA did not differ by dose. A ≥50 percent reduction in neuropathic symptoms was observed in 50 to 62 percent of patients treated with ALA versus 26 percent with placebo. Doses higher than 600 mg daily were limited by increasing adverse events (nausea, vomiting, and vertigo) without increasing efficacy. Confidence in the findings is limited by the short duration of this trial. There are no long-term studies that assess the effect of ALA on progression of neuropathy.
●Valproic acid – Valproate (500 to 1200 mg daily) was effective for reducing pain in diabetic neuropathy in two small placebo-controlled trials from a single center [65,66]. Larger confirmatory studies have not been performed, and the gabapentinoids are generally safer and require less monitoring. Valproate should not be used in patients with liver disease. It has well-recognized teratogenic effects and should be avoided in females of childbearing age.
●Carbamazepine – Carbamazepine has a potential role in patients who have failed standard options based on its efficacy in the treatment of trigeminal neuralgia as well as historical reports in patients with diabetic neuropathy [67,68]. Contemporary data are lacking, however.
Not recommended
●Opioids – We avoid the use of opioids for the treatment of painful diabetic neuropathy because of the lack of evidence regarding long-term effectiveness and because of the potential for tolerance, addiction, and overdose. We acknowledge that tapentadol has an FDA indication for diabetic neuropathy and that the 2017 American Diabetes Association (ADA) guidelines state that tramadol or tapentadol may be used as the "last-line" therapy prior to referral to a pain clinic [1]. However, a more recent systematic review on generalized treatment of pain suggests that the use of chronic opioid therapy has limited benefit and carries clear medical risk [69].
This newer statement is supported by an earlier 2009 systematic review of opioids for chronic noncancer pain, which found a paucity of evidence regarding long-term effectiveness and risks of such treatment, including the potential for opioid abuse, addiction, and overdose [70]. Similarly, a 2013 systematic review noted that the available randomized controlled trials of opioids for neuropathic pain did not clearly address the issues of abuse and addiction [71]. In a cohort study of over 9900 patients prescribed long-term opioid therapy for nonmalignant pain, the use of higher-dose regimens was associated with an increased risk of opioid overdose [72]. Because of these issues, we do not use opioids for treatment of painful diabetic neuropathy. A referral to a pain clinic is suggested for patients whose pain is unresponsive to conventional therapies.
●Topiramate – A systematic review that analyzed data from three randomized trials concluded that topiramate is not effective for painful diabetic polyneuropathy [73].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Neuropathy" and "Society guideline links: Neuropathic pain".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or email these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Nerve damage caused by diabetes (The Basics)" and "Patient education: Neuropathic pain (The Basics)")
●Beyond the Basics topic (see "Patient education: Diabetic neuropathy (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Prevention of complications – Established, symptomatic diabetic neuropathy is generally not reversible, and management aims to slow further progression and prevent complications, including diabetic foot ulcers, arthropathy, and falls. (See 'Preventive care' above.)
●Initial symptomatic therapies – Effective pharmacotherapy options for patients with painful diabetic neuropathy include the following (see 'Initial pharmacotherapy' above):
•Serotonin-norepinephrine reuptake inhibitors (duloxetine, venlafaxine)
•Tricyclic antidepressants (amitriptyline, desipramine, nortriptyline)
•Gabapentinoid antiseizure medications (pregabalin, gabapentin)
Each of these classes of agent have been shown to be more effective than placebo in randomized trials, and limited comparative data suggest that efficacy is similar across agents.
Selection of a specific agent should be individualized. In practice, comorbidities and concurrent medications often favor one class or another (ie, an antidepressant or a gabapentinoid antiseizure medication) (table 2). The narrower choice within a class is influenced by the patient's age, preferences with regard to dosing frequency, side effects, and cost/formulary considerations. (See 'Choice of agent' above and 'Administration and efficacy' above.)
●Alternative therapeutic options – None of the first-line choices are universally effective or tolerated, even with appropriate titration. A two- to three-month trial is generally required to titrate to effect and assess response to initial therapy. (See 'Inadequate response to initial therapy' above.)
•For most patients who do not respond to or tolerate the initial drug, we try an alternative first-line agent or use a combination of drugs from different classes. (See 'Inadequate response to initial therapy' above.)
•For patients who do not tolerate any of the first-line medications and who prefer topical or nonpharmacologic therapies, we discuss capsaicin cream, lidocaine patch, alpha-lipoic acid, transcutaneous electrical nerve stimulation, and spinal cord stimulation. (See 'Topical therapies or neuromodulation' above and 'Possibly effective' above.)
•We avoid use of opioids for the treatment of painful diabetic neuropathy because of the lack of evidence regarding long-term effectiveness; the potential for tolerance, addiction, and overdose; and the availability of multiple safer alternative therapies. (See 'Not recommended' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David K McCulloch, MD, who contributed to earlier versions of this topic review.
