The purpose of this study was to investigate the analgesic effects of duloxetine, specifically on postoperative pain, opioid consumption, and related side effects following total hip or knee arthroplasty. In this systematic review and meta-analysis, Medline, Cochrane, EMBASE, Scopus, and Web of Science were searched until November 2022 for studies which compared duloxetine with placebo when added to routine pain management protocols. Individual study risk of bias assessment was conducted based on Cochrane risk of bias tool 2. Random effect model meta-analysis was done on mean differences, to evaluate the outcomes. Nine randomized clinical trials (RCT) were included in the final analysis, totaling 806 patients. Duloxetine reduced opioid consumption (oral morphine milligram equivalents) on postoperative days (POD) two (mean difference (MD): - 14.35, p = 0.02), POD three (MD: -13.6, p < 0.001), POD seven (MD: -7.81, p < 0.001), and POD 14 (MD: -12.72, p < 0.001). Duloxetine decreased pain with activity on POD one, three, seven, 14, 90 (All p < 0.05), and pain at rest on POD two, three, seven, 14, and 90 (All p < 0.05). There was no significant difference in the prevalence of the side effects, except for increased risk of somnolence/drowsiness (risk ratio: 1.87, p = 0.007). Current evidence shows low to moderate opioid sparing effects of perioperative duloxetine and a statistically but not clinically significant reduction in pain scores. Patients treated with duloxetine had an increased risk for somnolence and drowsiness.