Skip navigation

Topical capsaicin NNTs

Bandolier seems to get a lot of the "is X effective" or "does X really work" type of question these days. When we were asked about topical capsaicin we were relieved to find a well done systematic review which allowed the calculation of NNTs [1].

Capsaicin

This is an alkaloid from chillies that first entered European knowledge after Columbus' second voyage to the new world in 1494. It has long been a feature of pharmacopoeias.

Recent interest concerns the use of topical capsaicin as an analgesic for conditions where pain may not be responsive to classical analgesics. There is evidence that capsaicin can deplete substance P in local sensory nerve terminals. Substance P is thought to be associated with initiation and transmission of painful stimuli, as well as a number of diseases - arthritis, psoriasis and inflammatory bowel disease.

This hypthesis gives topical capsaicin some degree of logic - remove the neurotransmitter, so remove the pain.

Does it work?

Zhang and Li Wan Po searched the literature for capsaicin papers using a sensitive strategy. They sought reports of clinical investigations. Only information from randomised, double-blind and placebo-controlled studies were used for quantitative analysis by clinical condition.

Results for the 13 trials that fulfilled these criteria and where there were extractable data are shown in the figure as a L'Abbé plot. Each symbol represents the proportion of patients in each trial reaching some clinical end-point for benefit, and the number next to the symbol the number of patients treated with topical capsaicin. Capsaicin results are plotted against placebo results. Points lying between the line of equality and the capsaicin axis are trials showing benefit. This plot is a simple representation of how similar or dissimilar trial results were found to be.

Diabetic neuropathy

Four trials reported the use of capsaicin 0.075% cream applied four times daily for 4 - 8 weeks in diabetic neuropathy. A total of 144 patients were treated with capsaicin and 165 with placebo cream. The end point was a physician global assessment of pain relief. Clinical improvement was pain completely gone, much better or better (and not no change, worse or much worse).

105/144 (73%) patients responded with capsaicin compared with 81/165 (49%) patients given placebo. The odds ratio favouring capsaicin was 2.7 (95%CI 1.7 - 4.3) and number-needed-to-treat (NNT) was 4.2 (2.9 - 7.5).

For every four patients treated with topical capsaicin, one would have had the pain of diabetic neuropathy relived who would not have had they been treated with placebo.

For comparison, oral anticonvulsant therapy for diabetic neuropathy in 66 treated patients in two trials yielded an NNT of 2.5 (1.8 - 4.0) [2].

Osteoarthritis

Three trials reported the use of capsaicin cream (0.025% in two, and 0.075% in one) four times daily for four weeks in osteoarthritis. The end point was articular tenderness or physicians global assessment of pain relief.

87/192 (45%) patients responded with capsaicin compared with 30/190 (16%) with placebo. The odds ratio favouring capsaicin was 4.4 (2.8 - 6.9) and NNT was 3.4 (2.6 - 4.8).

For every three patients treated with topical capsaicin, one would have had pain of osteoarthritis relieved who would not have been relieved had they been treated with placebo.

Postherpetic neuralgia

Only a single trial that fulfilled the inclusion criteria was available. Use of a 0.075% cream three or four times daily for six weeks resulted in pain relief in 4/16 patients with capsaicin compared with 1/16 patients with placebo.

Postmastectomy pain

A single trial of 0.075% cream four times daily for six weeks resulted in pain relief in 5/13 patients with capsaicin compared with 1/10 patients with placebo.

Psoriasis

Four trials reported on topical capsaicin 0.025% four times daily for 6 - 8 weeks in psoriasis. Psoriasis was rated by the degree of itching, scaling and erythema. The end point was a much better or better rating of overall appearance.

78/115 (68%) patients responded with capsaicin compared with 55/130 (42%) with placebo. The odds ratio favouring capsaicin was 2.8 (1.7 - 4.6) and NNT was 3.9 (2.7 - 7.4).

For every four patients treated with topical capsaicin, one would have had symptoms of psoriasis relieved who would not have been relieved had they been treated with placebo.

Cautious comments

The authors of the review suggested that blinding of trials of capsaicin might be difficult because of its irritant effects when applied to the skin. There may also be suggestions from some of the reports that skin irritation wears off with time, while analgesic effects may improve with time.

How much the placebo effect may influence the results is uncertain. These are difficult clinical conditions, and patients used the creams for up to eight weeks. Variability in the results in placebo groups can be seen from the figure to be from 0% to 50% of patients on placebo getting benefit. All the trials showed benefits over placebo.

The numbers of patients treated in these studies was not large, but that is not unusual in these difficult clinical conditions [2]. The review did not include results of adverse effects, which is a shame, since any treatment choice should balance the probability of benefit and the risk of harm.

References:

  1. WY Zhang, A Li Wan Po. The effectiveness of topically applied capsaicin. European Journal of Clinical Pharmacology 1994 46:517-22.
  2. H McQuay, D Carroll, A Jadad, P Wiffen, A Moore. Anticonvulsant drugs for management of pain: a systematic review. British Medical Journal 1995 311: 1047-52.




previous or next story in this issue