LAUSR

BACKGROUND Paracetamol and non-steroidal anti-inflammatory drugs (NSAIDs) are recommended as the basic pain treatment regimen for most surgeries. Glucocorticoids have well-known anti-inflammatory and anti-emetic properties and may also demonstrate analgesic effects. We assessed benefit and harm of adding glucocorticoids to a combination of paracetamol and NSAIDs for postoperative pain management. METHODS We searched Embase, Medline, and CENTRAL for randomised clinical trials investigating the addition of glucocorticoids vs placebo/no intervention to paracetamol and an NSAID in adults undergoing any type of surgery. We assessed three primary outcomes: cumulative opioid consumption at 24 hours postoperatively, serious adverse events, and pain at rest at 24 hours postoperatively. We performed meta-analysis and Trial Sequential Analysis, assessed risk of bias using the Risk of Bias 2 tool, and used the Grading of Recommendations Assessment, Development and Evaluation approach to evaluate the certainty of the evidence. RESULTS We identified 12 relevant trials of which nine trials randomising 804 participants were included in quantitative analysis. When added to paracetamol and NSAIDs, we found no evidence of a difference of glucocorticoids versus placebo/no intervention in cumulative opioid consumption at 24 hours postoperatively (MD -0.28, TSA-adjusted 95% CI -1.90 to 1.33, p = 0.68, moderate certainty of evidence), serious adverse events (RR 0.99, TSA-adjusted 95% CI 0.27 to 3.63, p = 0.93, very low certainty of evidence), or pain on the Numeric Rating Scale at 24 hours postoperatively (MD -0.39, TSA-adjusted 95% CI -0.84 to 0.17, p = 0.10, moderate certainty of evidence). All outcomes were assessed to be at high risk of bias and Trial Sequential Analysis showed that we had insufficient information for most outcomes. CONCLUSION Glucocorticoids added to a baseline therapy of paracetamol and an NSAID likely result in little to no difference in cumulative opioid consumption and pain at rest at 24 hours postoperatively. In addition, the evidence is very uncertain about the effect on serious adverse events. For most outcomes we did not have sufficient information to draw firm conclusions, and the certainty of the evidence varied from moderate to very low. This article is protected by copyright. All rights reserved.