Search tips
Search criteria 


Logo of bmjThis ArticleThe BMJ
BMJ. 2007 July 14; 335(7610): 57–58.
PMCID: PMC1914521

Treating painful diabetic polyneuropathy

Edward B Jude, consultant physician and honorary senior lecturer1 and Nicolaas Schaper, professor2

As consensus is lacking, protocols need to be devised and implemented locally

The incidence of diabetes is expected to double over the next two decades, which will result in more people with complications of diabetes.1 Diabetic polyneuropathy is one of the most common, with a prevalence of around 30-50%. It can have a major impact on patients' quality of life, and treatment is usually needed for many years.2 3

Community based studies report the prevalence of painful diabetic polyneuropathy as around 16-26%,4 and in one study 80% of the patients had moderate or severe pain.5 Quality of life is reduced in patients with painful diabetic polyneuropathy, with restriction in daily and social activities, and the condition is associated with depression, sleep disturbances, and anxiety.3 6 7 In this week's BMJ Wong and colleagues report a systematic review of the effects of drug treatment in painful diabetic polyneuropathy.8

Many types of drugs have been studied for relief of pain in diabetic polyneuropathy,9 as little evidence exists that classic analgesics such as paracetamol or non-steroidal anti-inflammatory drugs are effective. Surveys show that there is substantial scope for improvement in clinical care. In one UK population based study almost 40% of people with painful neuropathy reported that they had never received any treatment, almost a third had been prescribed drugs with no known efficacy in neuropathic pain, and only a minority had been treated with tricyclic antidepressants or anticonvulsants.4 Similar findings of inadequate treatment was found in two more recent European and American studies.6 10

Wong and colleagues' review identified the randomised placebo controlled trials of topically applied and orally administered drugs.8 Clinical success was defined as a 50% reduction in pain; withdrawal due to adverse events was a secondary outcome. The review found that tricyclic antidepressants were most effective in reducing pain by 50% (odds ratio 31.73; 95% confidence interval 3.68 to 189.89), followed by traditional anticonvulsants (lamotrigine, sodium valproate, carbamazepine; odds ratio 7.59; 2.16 to 26.58), and the newer generation anticonvulsants (gabapentin, oxacarbazepine, pregabalin; odds ratio 3.25; 2.27 to 4.66). The newer generation anticonvulsants were most likely to cause withdrawals due to adverse events, followed by tricyclic antidepressants, and the traditional anticonvulsants (odds ratios 2.98, 2.32, 1.51 respectively).

This review can help clinicians make evidence based choices in the management of painful diabetic polyneuropathy, and it also highlights some of the problems in the treatment regimens often prescribed for this condition. The efficacy of pharmacological treatment is limited: with of one of the most efficacious drugs in the meta-analysis, amitriptyline, three patients need to be treated to achieve a 50% reduction in pain score in one patient; in the other two patients the drug will have no effect or a limited effect. Adverse effects were not infrequent, and in the analysis of the effects of tramadol the odds ratio for withdrawal due to adverse events was higher than the odds ratio for 50% pain relief. It is difficult to compare the results of different trials in the review since the duration of treatment was highly variable (between 2 to 16 weeks) and could have influenced the number of withdrawals. Unfortunately all the trials in the review were relatively short; many included only a few patients), and for some drugs the confidence intervals were large. Clearly, robust studies with sufficient size and duration, preferably of at least one year, are needed.

Treatment of painful diabetic polyneuropathy is a clinical challenge and requires a treatment plan that should include psychosocial factors, glucose control, and, if necessary, pharmacological treatment. Listening to the patient and explaining the cause of the pain can help to reduce anxiety. Patients' beliefs and perceptions of the pain and its cause, coping strategies, mood changes, disturbed sleep, and anxiety all need to be addressed.

Several studies have looked at the impact of analgesic treatment on quality of life, sleep patterns, anxiety, and depression and most of them have shown an improvement in quality of life which might be directly related to improvement in pain score.11 Therefore, treating anxiety or depression first might also reduce the need for analgesics. A few observational studies have indicated that large variations in blood glucose can exacerbate the pain, so glycaemic control should be optimised.12 13

If, despite these measures, the pain persists and is so severe that pharmacological treatment is indicated tricyclic antidepressants seem the best first step, as suggested by Wong and colleagues' review. But contraindications include cardiac conduction disturbances and glaucoma, and side effects are not infrequent. The dilemma in treating painful diabetic polyneuropathy is what should be done with the many people who do not respond sufficiently to tricyclic antidepressants or in whom tricyclic antidepressants are contraindicated. Wong and colleagues suggest that the next step should be to use one of the older generation anticonvulsants. However, this advice is based on clustering a group of anticonvulsants with very different modes of action; if they are analysed separately, the number of studies for each drug is relatively small and the evidence for efficacy does not seem convincing. Therefore, we and others suggest pregabalin, duloxetine, or gabapentin as second line agents, as these drugs do not seem to differ much in efficacy or in the frequency and severity of side effects.9 Only one or two of these drugs should be included in the local treatment protocol, so that clinicians can develop sufficient experience with them. If this treatment fails, combination therapy or tramadol (or another opioid) can be considered.

Studies that evaluate the efficacy of the combination of two or more drugs to relieve pain are few. Only one study combined gabapentin and morphine in people with neuropathic pain; it showed that the combination gave better pain relief, required lower doses of both drugs, and fewer side effects than if the drugs were used singly.14

Although our knowledge of the treatment of painful diabetic polyneuropathy is growing, there is still no consensus on the most effective protocol. Therefore doctors treating people with diabetes must develop their own local screening and treatment protocols. They should include a treatment algorithm, indications and contraindications for drug treatment, potential adverse effects to look out for, and a simple technique to monitor the effectiveness of treatment.


Competing interests: EJ has received funding for research, speaker fees and support for education from Pfizer and speaker fees from Boehringer Ingelheim.

Provenance and peer review: Commissioned, not externally peer reviewed.


1. Wild S, Roglic G, Green A, Sicree R, King H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004;27:1047-53. [PubMed]
2. Jude E, Boulton AJM. End stage complications of diabetic neuropathy. Diabetes Rev 1999;7:395-410.
3. Vileikyte L, Rubin RR, Leventhal H. Psychological aspects of diabetic neuropathic foot complications: an overview. Misconceptions, assessment of patient beliefs and behaviour, illness perception at the individual level. Diabetes Metab Res Rev 2004;20:S13-8. [PubMed]
4. Daousi C, MacFarlane IA, Woodward A, Nurmikko TJ, Bundred PE, Benbow SJ. Chronic painful peripheral neuropathy in an urban community: a controlled comparison of people with and without diabetes. Diabet Med 2004;21:976-82. [PubMed]
5. Davies M, Brophy S, Williams R, Taylor A. The prevalence, severity, and impact of painful diabetic peripheral neuropathy in type 2 diabetes. Diabetes Care 2006;29:1518-22. [PubMed]
6. Gore M, Brandenburg NA, Hoffman DL, Tai KS, Stacey B. Burden of illness in painful diabetic peripheral neuropathy: the patients' perspectives. J Pain 2006;7:892-900. [PubMed]
7. Benbow SJ, Wallymahmed ME, MacFarlane IA. Diabetic peripheral neuropathy and quality of life. QJM 1998;91:733-7. [PubMed]
8. Wong M, Chung JWY, Wong TKS. Effects of treatments for symptoms of painful diabetic neuropathy: systematic review. BMJ 2007;335:87-90. [PMC free article] [PubMed]
9. Jensen TS, Backonja MM, Hernández Jiménez S, Tesfaye S, Valensi P, Ziegler D. New perspectives on the management of diabetic peripheral neuropathic pain. Diabetes Vasc Dis Res 2006;3:108-19. [PubMed]
10. Toelle T, Xu X, Sadosky AB. Painful diabetic neuropathy: a cross-sectional survey of health state impairment and treatment patterns. J Diabetes Complicat 2006;20:26-33. [PubMed]
11. McCarberg B, Billington R. Consequences of neuropathic pain: quality of life issues and associated costs. Am J Manag Care 2006;12:S263-8. [PubMed]
12. Boulton AJM, Drury J, Clarke, Ward JD. Continous subcutaneous insulin infusion in the management of painful diabetic neuropathy. Diabetes Care 1982;5:386-90. [PubMed]
13. Oyibo S, Prasad YD, Jackson NJ, Jude EB, Boulton AJ. The relationship between blood glucose excursions and painful diabetic peripheral neuropathy. Diabet Med 2002;19:870-3. [PubMed]
14. Gilron I, Bailey JM, Tu D, Holden RR, Weaver DF, Houlden RL. Morphine, gabapentin, or their combination for neuropathic pain. New Engl J Med 2005;352:1324-34. [PubMed]

Articles from The BMJ are provided here courtesy of BMJ Publishing Group