Current issues of ACP Journal Club are published in Annals of Internal Medicine


Topical capsaicin relieved painful diabetic neuropathy

ACP J Club. 1992 Mar-April;116:50. doi:10.7326/ACPJC-1992-116-2-050

Source Citation

The Capsaicin Study Group. Treatment of painful diabetic neuropathy with topical capsaicin. A multicenter, double-blind, vehicle-controlled study. Arch Intern Med. 1991 Nov;151:2225-9.



To determine the efficacy of topical 0.075% capsaicin cream in painful diabetic neuropathy.


Randomized, double-blind, placebo-controlled trial of 8 weeks duration.


12 medical centers in North America.


Participants included men or nonpregnant, nonlactating women aged 18 to 85 years, with fasting glucose < 10 mmol/L or glycosylated hemoglobin ≤ 11%, and with daily painful peripheral neuropathy or radiculopathy, and who had had ≥ 1 abnormal nerve conduction study. 277 patients were randomized; 121 of 138 (88%) patients randomized to the capsaicin group and 131 of 139 (94%) patients randomized to the placebo group completed at least the first evaluation and were included in the efficacy analysis.


After ≥ 7 days' abstinence from topical agents, patients were randomized to capsaicin cream or placebo and instructed to apply it to affected areas 4 times daily. Burning on cream application was treated by aspirin, acetaminophen, or ibuprofen; severe, persistent burning was treated by discontinuation of the study cream.

Main outcome measures

Pain was assessed at 2, 4, 6, and 8 weeks by a study physician using a 6-point scale (physician's global evaluation) for degree of improvement or worsening. At each visit patients also completed 100-mm, visual-analog scales (VAS) for evaluating pain intensity and degree of relief.

Main results

38 capsaicin-treated patients (28%) failed to complete the study compared with 20 controls (14%) (P < 0.02). After 8 weeks, more capsaicin-treated than placebo-treated patients were improved among those who completed ≥ 2 weeks of the treatment (69.5% vs 53.4%, P = 0.012). {This absolute benefit increase (ABI) of 16.1% means that 6 patients would need to be treated (NNT) with capsaicin for ≥ 2 weeks for 1 additional patient to show improvement, 95% CI 4 to 25; the relative benefit increase (RBI) was 30%, CI 7% to 59%.}* This difference was apparent at all assessment points. More capsaicin-treated patients recorded decreased pain intensity (38.1% vs 27.4%, P = 0.037) and increased pain relief (58.4% vs 45.3%, P = 0.04; {ARI 13.1%, NNT 8 CI 4 to 109; RRI 30%, CI 1.8% to 66%}*. 108 of 138 (78%) patients assigned to capsaicin reported side effects, especially local burning and coughing-sneezing compared with 41 of 139 (29%) patients assigned to placebo (P < 0.001). An intention-to-treat analysis, including all patients and assuming no change in global scores for the 25 patients who dropped out before the first follow-up, showed a nonsignificant difference between the 2 treatment groups for the study-physician assessment (P = 0.06).


Application of 0.075% capsaicin cream for 8 weeks relieved painful diabetic neuropathy or radiculopathy, but caused minor adverse effects in most patients.

Source of funding: GalenPharma.

Address for article reprint: Dr. P.D. Donofrio, Department of Neurology, Bowman Gray School of Medicine, Wake Forest University, Medical Center Boulevard, Winston-Salem, NC 27157-1078, USA.

*Numbers calculated from data in article.


Capsaicin cream has been effective in treating post-herpetic neuralgia. This is the first major study involving its use in diabetic neuropathy. Capsaicin application resulted in pain relief, as noted by the study physician, in 71% of patients after 8 weeks, whereas a similar effect was noted in 51% when only the vehicle was applied. A 40% improvement from baseline pain intensity occurred with capsaicin. Failure to eradicate pain completely may be due to associated neuropathic involvement in deep connective tissue, muscles, and bone sites inaccessible to topical drugs.

Because burning discomfort occurred more often with capsaicin (87 vs 23 cases), there may have been a breakdown in the ability to maintain a double-blind design. A patient with burning was 3 to 4 times more likely to be in the capsaicin group. It is possible that the degree of unblinding caused by this side effect created the small differences noted in the percentage of responding patients and in the subjective measures of pain intensity. The use of new analgesics by patients in the capsaicin-treated group (10%) and in the vehicle-treated group (16%) may have also affected the results.

Capsaicin clearly offers some relief for patients with painful neuropathy. Further trials to confirm the results of this study and trials using capsaicin in combination with other drugs with activity against neuropathic pain (carbamazepine, phenytoin, amitriptyline) will be of interest.

Stanley L. Wiener, MD
University of IllinoisChicago, Illinois, USA