These reports are available in PDF only (final report [3.9 MB]; disposition of comments [143 kB]). People using assistive technology may not be able to fully access information in these files. For additional assistance, please contact us.
Purpose of the Review
To assess benefits and harms of interventions for preventing complications and treating symptoms of diabetic peripheral neuropathy.
- Intensive glycemic control is not more effective than standard control for preventing foot ulcers but prevents amputations. Intensive glycemic control has higher rates of hypoglycemia than standard treatment.
- Home monitoring of foot skin temperature, therapeutic footwear, and integrated foot care are effective for preventing foot ulcers.
- Pregabalin, oxcarbazepine, duloxetine, venlafaxine, tricyclic antidepressants, tramadol, tapentadol, botulinum toxin, alpha-lipoic acid, and spinal cord stimulation are effective for reducing pain. However, oral drugs and spinal cord stimulation have substantial adverse effects.
- Longer term pharmacologic studies and studies of nonpharmacologic approaches for pain management, perceived risk of falling, and falls are needed.
To assess benefits and harms of interventions for preventing diabetic peripheral neuropathy (DPN) complications and treatment of DPN symptoms.
We searched PubMed® and the Cochrane Database of Systematic Reviews for systematic reviews from January 1, 2011, to October 12, 2015. For questions for which we did not identify high-quality relevant systematic reviews, we searched for primary studies using PubMed®, Embase®, and the Cochrane Central Register of Controlled Trials (CENTRAL) from inception to May 24, 2016. We searched ClinicalTrials.gov for pharmacologic treatment of DPN symptoms.
For the prevention of DPN complications, we included a systematic review of primary randomized controlled trials and nonrandomized studies with a concurrent comparison group. For the treatment of DPN symptoms, we included a systematic review of primary parallel or crossover randomized controlled trials that were blinded for interventions where blinding was possible from the published literature and ClinicalTrials.gov. Two reviewers evaluated studies for eligibility, serially abstracted data using standardized forms, independently evaluated the risk of bias of the reviews and studies, and graded the strength of evidence (SOE) for critical outcomes (foot ulcers, amputations, falls, pain, and quality of life).
We included 62 studies (30 studies from an existing systematic review and 32 newly identified studies reported in 37 articles) for prevention of DPN complications and 131 studies (58 studies from an existing systematic review, 48 newly identified additional studies reported in 50 articles, and 25 studies from ClinicalTrials.gov) for treatment of DPN symptoms. For prevention of DPN complications, although intensive glycemic control (as defined by each individual study) does not prevent foot ulcers more than standard control for type 2 diabetes, it prevents lower extremity amputations (moderate SOE). Intensive glycemic control had higher rates of hypoglycemia than standard treatment. For nonpharmacologic treatment options, specific types of therapeutic footwear (moderate SOE), integrated foot care (low SOE), home monitoring of foot skin temperature (moderate SOE), and specific types of surgical interventions (low SOE) are effective for lowering incidence and/or recurrence of foot ulcers. There is insufficient evidence to evaluate whether physical therapy, exercise, or balance training reduces falls. For treatment of DPN pain symptoms, the serotonin-noradrenaline reuptake inhibitors duloxetine and venlafaxine (moderate SOE), the anticonvulsants pregabalin and oxcarbazepine (low SOE), the drug classes of tricyclic antidepressants (low SOE) and atypical opioids (tramadol and tapentadol) (low SOE), and the injectable neurotoxin botulinum toxin (low SOE) are more effective than placebo for reducing pain in short-term followup. For harms, all effective oral drugs had more than 9 percent dropouts due to adverse effects. For nonpharmacologic treatments, alpha-lipoic acid is more effective than placebo (low SOE) and spinal cord stimulation is more effective than usual care for pain (low SOE), but spinal cord stimulation had risks of serious complications. We were unable to draw conclusions about quality of life for any of the treatments due to incomplete reporting (insufficient SOE).
For prevention of complications, intensive glycemic control is more effective than standard control for prevention of amputation, and home monitoring of foot skin temperature, therapeutic footwear, and integrated interventions are effective for preventing incidence and/or recurrence of foot ulcers. For reducing pain, the only class with moderate strength of evidence was serotonin-noradrenaline reuptake inhibitors; pregabalin and oxcarbazepine, atypical opioids, botulinum toxin, alpha-lipoic acid and spinal cord stimulation are more effective than placebo but with low SOE. However, studies were generally short term with unclear risk of bias, we could not draw conclusions for quality of life, all oral drugs had significant side effects, opioids have significant long-term risks including abuse, and spinal cord stimulation has risks of serious complications.
Dy SM, Bennett WL, Sharma R, Zhang A, Waldfogel JM, Nesbit SA, Yeh H, Chelladurai Y, Feldman D, Wilson LM, Robinson KA. Preventing Complications and Treating Symptoms of Diabetic Peripheral Neuropathy. Comparative Effectiveness Review No. 187. (Prepared by the Johns Hopkins University Evidence-based Practice Center under Contract No. 290-2015-00006-I.) AHRQ Publication No. 17-EHC005-EF. Rockville, MD: Agency for Healthcare Research and Quality; March 2017. www.effectivehealthcare.ahrq.gov/reports/final.cfm. doi: https://doi.org/10.23970/AHRQEPCCER187.
Waldfogel JM, Nesbit SA, Dy SM, Sharma R, Zhang A, et al. Pharmacotherapy for diabetic peripheral neuropathy pain and quality of life. A systematic review. Published online before print March 24, 2017. Neurology. doi: http://dx.doi.org/10.1212/WNL.0000000000003882.