ABSTRACT: Perioperative anemia frequently occurs in patients undergoing orthopedic surgery. We aimed to evaluate the efficacy of perioperative intravenous iron therapy (IVIT) on transfusion and recovery profiles during orthopedic surgery. We searched PubMed, Embase, Cochrane, and Google Scholar for eligible clinical trials (randomized controlled trials, RCTs; case-control studies, CCSs) in comparing IVIT and no iron therapy, up to September 2018. Primary outcomes were the effects of IVIT on the proportion of patients transfused and units of red blood cells (RBCs) transfused perioperatively. Secondary outcomes were the effects of IVIT on recovery profiles, such as length of hospital stay (LOS), post-operative infection, and mortality. Subgroup analysis was performed based on iron dose (low: ? 300 mg, high: > 400 mg), IVIT period (pre-operative, post-operative, perioperative), and study design. We identified 12 clinical trials (4 RCTs with 616 patients and 8 CCSs with 1,253 patients). IVIT significantly reduced the proportion of patients transfused by 31% (RR, 0.69; P = 0.0002), and units of RBCs transfused by 0.34 units/person (MD, -0.34; P = 0.0007). For subgroup analysis by iron dose, low- or high-dose IVIT significantly reduced the proportion of patients transfused (RR, 0.73, P = 0.005; RR, 0.68, P = 0.008), and RBC units transfused (MD, -0.47, P < 0.0001; MD, -0.28, P = 0.04). For subgroup analysis by period, IVIT administered post-operatively significantly reduced the proportion of patients transfused (post-operative: RR, 0.60, P = 0.002; pre-operative: RR, 0.74, P = 0.06) and RBC units transfused (post-operative: MD, -0.44, P <0.00001; pre-operative: MD, -0.29, P = 0.06). For subgroup analysis by study design, IVIT decreased the proportion of patients transfused and RBC units transfused in the group of CCSs, but IVIT in the group of RCTs did not. IVIT significantly shortened LOS by 1.6 days (P = 0.0006) and reduced post-operative infections by 33% (P = 0.01). IVIT did not change mortality. Perioperative IVIT during orthopedic surgery, especially post-operatively, appears to reduce the proportion of patients transfused and units of RBCs transfused, with shorter LOS and decreased infection rate, but no change in mortality rate. These were only found in CCSs and not in RCTs due to the relatively small number of RCTs with low to high risk of bias.