BACKGROUNDPostoperative analgesic effects of systemic glucocorticoids given as an adjunct to treatment are largely undetermined in alloplastic procedures.OBJECTIVESTo investigate the beneficial and harmful effects of peri-operative systemic glucocorticoid treatment for pain after total hip arthroplasty (THA) or total knee arthroplasty (TKA).DESIGNA systematic review of randomised clinical trials (RCTs) with meta-analyses, trial sequential analyses and GRADE. Primary outcome was 24 h intravenous (i.v.) morphine (or equivalent) consumption with a predefined minimal important difference (MID) of 5 mg. Secondary outcomes included pain at rest and during mobilisation (MID, VAS 10 mm), adverse and serious adverse events (SAEs).DATA SOURCESWe searched EMBASE, Cochrane CENTRAL, PubMed and Google Scholar up to October 2021.ELIGIBILITY CRITERIARCTs investigating peri-operative systemic glucocorticoid versus placebo or no intervention, for analgesic pain management of patients at least 18 years undergoing planned THA or TKA, irrespective of publication date and language.RESULTSWe included 32 RCTs with 3521 patients. Nine trials were at a low risk of bias. Meta-analyses showed evidence of a reduction in 24 h cumulative morphine consumption with glucocorticoids by 5.0 mg (95% CI 2.2 to 7.7; P = 0.0004). Pain at rest was reduced at 6 h by 7.8 mm (95% CI 5.5 to 10.2; P < 0.00001), and at 24 h by 6.3 mm (95% CI 3.8 to 8.8; P < 0.00001). Pain during mobilisation was reduced at 6 h by 9.8 mm (95% CI 6.9 to 12.8; P < 0.00001), and at 24 h by 9.0 mm (95% CI 5.5 to 12.4, P < 0.00001). Incidence of adverse events was generally lower in the glucocorticoid treatment group. SAEs were rarely reported. The GRADE rated quality of evidence was low to very low.CONCLUSIONPeri-operative systemic glucocorticoid treatment reduced postoperative morphine consumption to an individually relevant level following hip and knee arthroplasty. Pain levels were reduced but were below the predefined MID. The quality of evidence was generally low.