Skip navigation
Acute Pain | Chronic Pain | General

Topically applied non-steroidal anti-inflammatory drugs in acute pain

Clinical bottom line:

Topical NSAIDs are effective in the relief of pain caused by soft tissue injury, sprains, strains and trauma. Ketoprofen, felbinac, ibuprofen and piroxicam have proven efficacy. The NNT was 3.9 (3.4 to 4.3) was generated for at least 50% pain relief after one week of topical NSAID use compared with placebo.

Benzydamine and indomethacin did not appear to be effective. No significant local or systemic adverse effects are associated with topical NSAID use.

Topical NSAIDs in acute pain

Oral NSAIDs are associated with increased risk of adverse effects such as gastrointestinal problems. The question arises therefore whether topical NSAIDs, which are widely available without prescription, are more suitable for a number of conditions. However, it has also been argued that topical NSAIDs have no action other than as rubefacients.

Systematic review

Moore RA, Carroll D, Wiffen PJ, Tramer M, McQuay HJ A systematic review of topically-applied non-steroidal anti-inflammatory drugs. British Medical Journal 1998; 316:333-8.

Inclusion criteria were randomised controlled trials of NSAIDs with pain outcomes; acute pain conditions (strains, sprains, sports injuries); full journal publication and unpublished drug company trials. Trials in vaginitis, oral and buccal conditions, thrombophlebitis and experimental pain were excluded.

Dichotomous data on pain outcomes were extracted from trials for analysis. At least 50% pain relief was regarded as a clinically relevant outcome. From the dichotomous data, information on the proportion, and then the number of patients who achieved at least 50% pain relief were calculated. A number-needed-to-treat for at least 50% pain relief and the relative risk and benefit of the treatment were then calculated


Included trials were in recent soft tissue injury, sprains, strains and trauma.

Twenty-seven of 37 trials with dichotomous outcomes showed significant improvement with topical NSAIDs compared with placebo. This was also true for the three trials with no dichotomous outcomes.

For all 37 comparisons the NNT was 3.9 (3.4 to 4.3). However, this NNT was increased (i.e. drug less effective) when large trials were analysed separately.

Based on pooled data of a minimum of three trials per drug, ketoprofen, felbinac, ibuprofen and piroxicam were statistically superior to placebo. Indomethacin and benzydamine were not.

Figure: At least 50% pain relief with topical NSAID versus placebo

Table: NNTs and NNHs for comparisons of topical NSAIDs with placebo

  Trials Patients Average number of treated patients Percent success or harm with placebo Percent success or harm with topical NSAID Relative benefit or risk (95% CI) NNT (95% CI)

Combined data/adverse effects            

Combined efficacy data 37 3239 47 39 71 1.7 (1.5 to 1.9) 3.9 (3.4 to 4.4)
Local adverse effects       3 2.6 1.2 (0.8 to 1.7) not calculated
Systemic adverse effects       0.7 0.8 1.0 (0.6 to 1.8) not calculated
Withdrawal due to adverse effects       0.4 0.6 0.8 (0.4 to 1.4) not calculated

Effect of particular topical NSAIDs            

Ketoprofen 9 724 43 36 74 2.0 (1.5 to 2.6) 2.6 (2.3 to 3.2)
Felbinac 3 413 70 32 66 2.0 (1.5 to 2.7) 3.0 (2.4 to 4.1)
Ibuprofen 4 284 36 34 70 1.9 (1.2 to 3.0) 3.5 (2.5 to 5.6)
Piroxicam 4 589 74 39 69 1.6 (1.2 to 2.2) 4.2 (3.1 to 6.1)
Benzydamine 4 245 31 62 84 1.4 (0.9 to 2.0) not calculated
Indomethacin 3 394 66 32 47 1.3 (0.9 to 1.8) not calculated
Response is either proportion of patients with successful outcome or percent of patients with an adverse effect.

Three trials compared topical with oral NSAIDs. In all cases topical NSAIDs were as effective as oral doses.

Adverse effects

There was no significant difference in the frequency of local or systemic adverse effects or drug-related withdrawals.

Related topics