Peter Jones, Stuart R Dalziel, Rain Lamdin, Jennifer L Miles‐Chan, Christopher Frampton

Acute soft tissue injuries are common and costly. The best drug treatment for such injuries is not certain, although non-steroidal anti-inflammatory drugs (NSAIDs) are often recommended. To assess the effects (benefits and harms) of NSAIDs compared with other oral analgesics for treating acute soft tissue injuries. We searched the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (12 September 2014), the Cochrane Central Register of Controlled Trials (The Cochrane Library, 2014 Issue 8), MEDLINE (1966 to September 2014), EMBASE (1980 to September 2014), CINAHL (1937 to November 2012), AMED (1985 to November 2012), International Pharmaceutical Abstracts (1970 to November 2012), PEDro (1929 to November 2012), and SPORTDiscus (1985 to November 2012), plus internet search engines, trial registries and other databases. We also searched reference lists of relevant articles and contacted authors of retrieved studies and pharmaceutical companies to obtain relevant unpublished data. We included randomised or quasi-randomised controlled trials involving people with acute soft tissue injury (sprain, strain or contusion of a joint, ligament, tendon or muscle occurring up to 48 hours prior to inclusion in the study) and comparing oral NSAID versus paracetamol (acetaminophen), opioid, paracetamol plus opioid, or complementary and alternative medicine. The outcomes were pain, swelling, function, adverse effects and early re-injury. At least two review authors independently assessed studies for eligibility, extracted data and assessed risk of bias. We assessed the quality of the evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology. We included 16 trials, with a total of 2144 participants. Two studies included children only. The other 14 studies included predominantly young adults, of whom over 60% were male. Seven studies recruited people with ankle sprains only. Most studies were at low or unclear risk of bias; however, two were at high risk of selection bias, three were at high risk of bias from lack of blinding, one was at high risk of bias due to incomplete outcome data, and four were at high risk of selective outcome reporting bias. The evidence was usually either low quality or very low quality, reflecting study limitations, indirectness such from as suboptimal dosing of single comparators, imprecision, or one or more of these. Thus we are either uncertain or very uncertain of the estimates.Nine studies, involving 991 participants, compared NSAIDs with paracetamol. While tending to favour paracetamol, there was a lack of clinically important differences between the two groups in pain at less than 24 hours (377 participants, 4 studies; moderate-quality evidence), at days 1 to 3 (431 participants, 4 studies; low quality), and at day 7 or over (467 participants, 4 studies; low quality). A similar lack of difference between the two groups applied to swelling at day 3 (86 participants, 1 study; very low quality) and at day 7 or over (77 participants, 1 study; low quality). There was little difference between the two groups in return to function at day 7 or over (316 participants, 3 studies; very low quality): based on an assumed recovery of function of 804 per 1000 participants in the paracetamol group, 8 fewer per 1000 recovered in the NSAID group (95% confidence interval (CI) 80 fewer to 73 more). There was low-quality evidence of a lower risk of gastrointestinal adverse events in the paracetamol group: based on an assumed risk of gastrointestinal adverse events of 16 per 1000 participants in the paracetamol group, 13 more participants per 1000 had a gastrointestinal adverse event in the NSAID group (95% CI 0 to 35 more).Four studies, involving 958 participants, compared NSAIDs with opioids. Since a study of a selective COX-2 inhibitor NSAID (valdecoxib) that was subsequently withdrawn from the market dominates the evidence for this comparison (706 participants included in the analyses for pain, function and gastrointestinal adverse events), the applicability of these results is in doubt and we give only a brief summary. There was low quality evidence for a lack of clinically important differences between the two groups regarding pain at less than 24 hours, at days 4 to 6, and at day 7. Evidence from single studies showed a similar lack of difference between the two groups for swelling at day 3 (68 participants) and day 10 (84 participants). Return to function at day 7 or over favoured the NSAID group (low-quality), and there were fewer gastrointestinal adverse events in the selective COX-2 inhibitor NSAID group (very low quality).Four studies, involving 240 participants, compared NSAIDs with the combination of paracetamol and an opioid. The applicability of findings from these studies is partly in question because the dextropropoxyphene combination analgesic agents used are no longer in general use. While the point estimates favoured NSAID, the very low-quality evidence did not show a difference between the two interventions in the numbers with little or no pain at day 1 (51 participants, 1 study), day 3 (149 participants, 2 studies), or day 7 (138 participants, 2 studies). Very low-quality evidence showed a similar lack of difference between the two groups applied to swelling at day 3 (reported in two studies) and at day 7 (reported in two studies), in return to function at day 7 (89 participants, 1 study), and in gastrointestinal adverse events (141 participants, 3 studies).No studies compared NSAIDs with complementary and alternative medicines, and no study reported re-injury rates. There is generally low- or very low-quality but consistent evidence of no clinically important difference in analgesic efficacy between NSAIDs and other oral analgesics. There is low-quality evidence of more gastrointestinal adverse effects with non-selective NSAID compared with paracetamol. There is low- or very low-quality evidence of better function and fewer adverse events with NSAIDs compared with opioid-containing analgesics; however, one study dominated this evidence using a now unavailable COX-2 selective NSAID and is of uncertain applicability. Further research is required to determine whether there is any difference in return to function or adverse effects between both non-selective and COX-2 selective NSAIDs versus paracetamol.