Data sources Medline, Embase, CENTRAL, Cumulative Index to Nursing and Allied Health Literature and Scopus databases were searched (until July 31 2019). They searched the WHO International Clinical Trials Registry Platform and ClinicalTrials.gov databases for completed and ongoing trials, as well as ProQuest, Google Scholar (first 100 hits) and the OpenGrey database for those unpublished. In addition, abstracts of annual meetings from the American Association of Endodontists, the International Federation of Endodontic Associations and the European Society of Endodontics were searched. A manual search of references in related papers and textbooks was carried out. Any missing information was filled in by contacting the authors.Study selection All randomised clinical trials (RCT) of nonsurgical endodontic therapy with pharmaceutical treatments provided postoperatively to alleviate pain on adult subjects were eligible. Two reviewers selected the studies and extracted the data and the results were examined by a third reviewer. Any differences were settled with the help of a fourth reviewer.Data extraction and synthesis Two authors independently evaluated and retrieved data from chosen trials found through searches. Another reviewer looked at the trial selection and data extraction. They also assessed the risk of bias in each of the research they chose.The mean and standard deviation of a 0-100 (mm) visual analogue pain scale were extracted or derived from study text, graphs and tables The postoperative medications were compared using network meta-analysis in a Bayesian model. Mean differences (MDs) and 95% credible intervals (CrIs) of post-operative pain were estimated at 6-8 hours and 12, 24 and 48 hours. Two authors utilised the CINeMA web application to assess the confidence in the main results by accounting for six domains: within- and across-study bias; indirectness; imprecision; heterogeneity; and incoherence. Each domain was assigned no concerns, major concerns, or minor concerns. Each result would be assigned a confidence level of high, moderate, low, or very low. To establish an agreement, a third reviewer was consulted.Results From the initial search of 969 records, 11 RCTs matched the inclusion criteria. Five studies were rated as having a low risk of bias, five as having a moderate risk and one as having a high risk. Moderate and major heterogeneity and some inconsistencies were detected, though not statistically significant.Oral medications were classified into placebo, non-steroidal anti-inflammatory drugs (NSAIDs), opioids, acetaminophen, NSAIDs + acetaminophen, corticosteroids, NSAIDs + benzodiazepines and NSAIDs + opioids groups.With moderate confidence, NSAIDs + acetaminophen were shown to be more efficacious than placebo at 6-8 hours following nonsurgical root canal therapy (MD = 22; 95% CrI = [-38, -7.2]). With very low confidence, NSAIDs had a greater effect than placebo after 12 and 24 hours (MD = -28; 95% CrI = [-49, -7] and MD = -15; 95% CrI = [-27, -2.3], respectively). After 6 12 and 24 hours, other medications were no more effective than placebo in pain reduction. After 48 hours, no treatment was more effective than placebo. Subgroup analysis found that other medications, such as corticosteroids and acetaminophen, were no more beneficial than placebo for patients who could not use NSAIDs. Six trials provided safety data and concluded that the therapies were safe and posed no significant risks.Conclusions Very low to moderate-quality evidence suggests that postoperative administration of NSAIDs and acetaminophen, or NSAIDs alone, reduced discomfort after non-surgical root canal therapy in patients with irreversible pulpitis or pulpal necrosis. Trials have shown, however, that using these drugs does not result in major complications. We cannot, however, be confident that they are risk-free.