In patients with liver failure complicated by acute kidney injury, renal replacement therapy (RRT) is often required to improve the internal environment. The use of anticoagulants for RRT in patients with liver failure remains controversial. We searched the PubMed, Embase, Cochrane Library, and Web of Science databases for studies. The methodological quality of the included studies was assessed using the Methodological Index for Nonrandomized Studies. A meta-analysis was performed using R software (version 3.5.1) and Review Manager (version 5.3.5). During RRT, 348 patients from 9 studies received regional citrate anticoagulation (RCA), and 127 patients from 5 studies received heparin anticoagulation (including heparin and LMWH). Among patients who received RCA, the incidence of citrate accumulation, metabolic acidosis, and metabolic alkalosis were 5.3% (95% confidence interval [CI]: 0%-25.3%), 26.4% (95% CI: 0-76.9), and 1.8% (95% CI: 0-6.8), respectively. The potassium, phosphorus, total bilirubin (TBIL), and creatinine levels were lower, whereas the serum pH, bicarbonate, base excess levels, and total calcium/ionized calcium ratio were higher after treatment than before treatment. Among patients who received heparin anticoagulation, the TBIL levels were lower, whereas the activated partial thromboplastin clotting time and D-dimer levels were higher after treatment than before treatment. The mortality rates in the RCA and heparin anticoagulation groups were 58.9% (95% CI: 39.2-77.3) and 47.4% (95% CI: 31.1-63.7), respectively. No statistical difference in mortality was observed between the 2 groups. For patients with liver failure, the administration of RCA or heparin for anticoagulation during RRT under strict monitoring may be safe and effective.