Sheena Derry, Philip J Wiffen, Eija A Kalso, Rae F Bell, Dominic Aldington, Tudor Phillips, Helen Gaskell, R Andrew Moore

Topical analgesic drugs are used for a variety of painful conditions. Some are acute, typically strains or sprains, tendinopathy, or muscle aches. Others are chronic, typically osteoarthritis of hand or knee, or neuropathic pain. To provide an overview of the analgesic efficacy and associated adverse events of topical analgesics (primarily nonsteroidal anti-inflammatory drugs (NSAIDs), salicylate rubefacients, capsaicin, and lidocaine) applied to intact skin for the treatment of acute and chronic pain in adults. We identified systematic reviews in acute and chronic pain published to February 2017 in the Cochrane Database of Systematic Reviews (the Cochrane Library). The primary outcome was at least 50% pain relief (participant-reported) at an appropriate duration. We extracted the number needed to treat for one additional beneficial outcome (NNT) for efficacy outcomes for each topical analgesic or formulation, and the number needed to treat for one additional harmful outcome (NNH) for adverse events. We also extracted information on withdrawals due to lack of efficacy or adverse events, systemic and local adverse events, and serious adverse events. We required information from at least 200 participants, in at least two studies. We judged that there was potential for publication bias if the addition of four studies of typical size (400 participants) with zero effect increased NNT compared with placebo to 10 (minimal clinical utility). We extracted GRADE assessment in the original papers, and made our own GRADE assessment. Thirteen Cochrane Reviews (206 studies with around 30,700 participants) assessed the efficacy and harms from a range of topical analgesics applied to intact skin in a number of acute and chronic painful conditions. Reviews were overseen by several Review Groups, and concentrated on evidence comparing topical analgesic with topical placebo; comparisons of topical and oral analgesics were rare.For at least 50% pain relief, we considered evidence was moderate or high quality for several therapies, based on the underlying quality of studies and susceptibility to publication bias.In acute musculoskeletal pain (strains and sprains) with assessment at about seven days, therapies were diclofenac Emulgel (78% Emulgel, 20% placebo; 2 studies, 314 participants, NNT 1.8 (95% confidence interval 1.5 to 2.1)), ketoprofen gel (72% ketoprofen, 33% placebo, 5 studies, 348 participants, NNT 2.5 (2.0 to 3.4)), piroxicam gel (70% piroxicam, 47% placebo, 3 studies, 522 participants, NNT 4.4 (3.2 to 6.9)), diclofenac Flector plaster (63% Flector, 41% placebo, 4 studies, 1030 participants, NNT 4.7 (3.7 to 6.5)), and diclofenac other plaster (88% diclofenac plaster, 57% placebo, 3 studies, 474 participants, NNT 3.2 (2.6 to 4.2)).In chronic musculoskeletal pain (mainly hand and knee osteoarthritis) therapies were topical diclofenac preparations for less than six weeks (43% diclofenac, 23% placebo, 5 studies, 732 participants, NNT 5.0 (3.7 to 7.4)), ketoprofen over 6 to 12 weeks (63% ketoprofen, 48% placebo, 4 studies, 2573 participants, NNT 6.9 (5.4 to 9.3)), and topical diclofenac preparations over 6 to 12 weeks (60% diclofenac, 50% placebo, 4 studies, 2343 participants, NNT 9.8 (7.1 to 16)). In postherpetic neuralgia, topical high-concentration capsaicin had moderate-quality evidence of limited efficacy (33% capsaicin, 24% placebo, 2 studies, 571 participants, NNT 11 (6.1 to 62)).We judged evidence of efficacy for other therapies as low or very low quality. Limited evidence of efficacy, potentially subject to publication bias, existed for topical preparations of ibuprofen gels and creams, unspecified diclofenac formulations and diclofenac gel other than Emulgel, indomethacin, and ketoprofen plaster in acute pain conditions, and for salicylate rubefacients for chronic pain conditions. Evidence for other interventions (other topical NSAIDs, topical salicylate in acute pain conditions, low concentration capsaicin, lidocaine, clonidine for neuropathic pain, and herbal remedies for any condition) was very low quality and typically limited to single studies or comparisons with sparse data.We assessed the evidence on withdrawals as moderate or very low quality, because of small numbers of events. In chronic pain conditions lack of efficacy withdrawals were lower with topical diclofenac (6%) than placebo (9%) (11 studies, 3455 participants, number needed to treat to prevent (NNTp) 26, moderate-quality evidence), and topical salicylate (2% vs 7% for placebo) (5 studies, 501 participants, NNTp 21, very low-quality evidence). Adverse event withdrawals were higher with topical capsaicin low-concentration (15%) than placebo (3%) (4 studies, 477 participants, NNH 8, very low-quality evidence), topical salicylate (5% vs 1% for placebo) (7 studies, 735 participants, NNH 26, very low-quality evidence), and topical diclofenac (5% vs 4% for placebo) (12 studies, 3552 participants, NNH 51, very low-quality evidence).In acute pain, systemic or local adverse event rates with topical NSAIDs (4.3%) were no greater than with topical placebo (4.6%) (42 studies, 6740 participants, high quality evidence). In chronic pain local adverse events with topical capsaicin low concentration (63%) were higher than topical placebo (5 studies, 557 participants, number needed to treat for harm (NNH) 2.6), high quality evidence. Moderate-quality evidence indicated more local adverse events than placebo in chronic pain conditions with topical diclofenac (NNH 16) and local pain with topical capsaicin high-concentration (NNH 16). There was moderate-quality evidence of no additional local adverse events with topical ketoprofen over topical placebo in chronic pain. Serious adverse events were rare (very low-quality evidence).GRADE assessments of moderate or low quality in some of the reviews were considered by us to be very low because of small numbers of participants and events. There is good evidence that some formulations of topical diclofenac and ketoprofen are useful in acute pain conditions such as sprains or strains, with low (good) NNT values. There is a strong message that the exact formulation used is critically important in acute conditions, and that might also apply to other pain conditions. In chronic musculoskeletal conditions with assessments over 6 to 12 weeks, topical diclofenac and ketoprofen had limited efficacy in hand and knee osteoarthritis, as did topical high-concentration capsaicin in postherpetic neuralgia. Though NNTs were higher, this still indicates that a small proportion of people had good pain relief.Use of GRADE in Cochrane Reviews with small numbers of participants and events requires attention.