Nine randomized controlled trials (806 patients) comparing the effect of duloxetine against a control in patients undergoing total hip or knee replacement (THA & TKA, respectively) were included in this systematic review and meta-analysis. Outcomes of interest included total opioid consumption, pain at rest and with activity on a Normalized Pain Scale (NPS), and the incidence of adverse events. Opioid consumption on days 2, 3, 7, and 14 favored the duloxetine group. Pain activity favored the duloxetine group at all time points except for 2 days and 6 weeks. Pain at rest favored the duloxetine group at 2, 3, 7, and 14 days with similar results at all other time points. All adverse events were similar among the two groups except for somnolence which favored the control group.