An evidence-based guide to treating diabetic neuropathy pain

Joshua Johnson, M.D.
General neurologist, peripheral nerve disorders
Providence Newberg Medical Center

Published September 2011

Painful diabetic neuropathy affects 16 percent of people with diabetes, yet it is frequently undiagnosed and undertreated. Treatment is usually determined by personal preferences or by the few drug trials that measure efficacy only in the most recently introduced neuropathic pain medications.

The choice of medication is further complicated by the large number of medications available, the variable quality of efficacy studies and the scarcity of head-to-head medication trials. This past May, three professional associations with an interest in neuropathy published a consensus statement.

Its objective is “to develop a scientifically sound and clinically relevant evidence-based guideline for the treatment of painful diabetic neuropathy.” The recommendations are summarized here.

Pregabalin is the only oral neuropathic pain medication with data of sufficient quality to allow a Level A recommendation for controlling diabetic neuropathic pain.

Pregabalin also improves quality of life and improves sleep. Unfortunately, pregabalin (Lyrica) remains quite expensive and often is authorized by insurance companies only after less expensive medications have failed.

The consensus panel’s secondary recommendations are:

  • Gabapentin (900-3600 mg/day)
  • Valproate (500-1200 mg/day)
  • Venlafaxine (75-225 mg/day)
  • Duloxetine (60-120 mg/day)
  • Amitriptyline (25-100 mg/day)
  • Dextromethorphan (400 mg/day)

Data also support the use of morphine, oxycodone and tramadol for diabetic neuropathic pain. Some efficacy is seen in the combination of duloxetine and venlafaxine or nortriptyline and fluphenazine.

The following oral medications, however, were not recommended to treat painful diabetic neuropathy:

  • Oxcarbazepine
  • Lamotrigine
  • Clonidine
  • Mexiletine
  • Alpha-lipoic acid and other vitamins

The limitations of drug trials

The consensus panel notes some limitations in the published drug trials. The placebo effect ranges from 0 to 50 percent. Most trials last only two to 20 weeks despite the chronicity of diabetic neuropathic pain. Narcotics have the potential for habituation and abuse. Valproate is teratogenic and probably should not be considered as first- or second-line treatment in women of childbearing age.

The study reviews other pharmacologic and non-pharmacologic treatments for painful diabetic neuropathy. Class I studies have supported the use of two topical medications: capsaicin 0.075 percent, applied four times daily, and isosorbide dinitrate spray. The Lidoderm patch was found effective in two studies, but the data quality was poor. Percutaneous electrical stimulation improved pain and sleep in one Class I study.

My own approach is to start with gabapentin or nortriptyline, which have fewer side effects than amitriptyline. If these agents fail or are not tolerated, then there is a chance that insurance will authorize Lyrica. If these agents are ineffective, I consider trying duloxetine, tramadol or carbamazepine (the consensus recommendations do not discuss the latter agent).

Capsaicin is difficult for patients to tolerate, but some patients have success with it, and it can be useful for those who do not tolerate oral medications. I have not used isosorbide dinitrate spray. If a narcotic is required to control the neuropathic pain, I ask the patient to see a pain physician for prescription of the narcotic and overall management of the neuropathic pain.

Clinical articles by Joshua Johnson, M.D.