Total knee arthroplasty (TKA) is the most cost-effective, and potent method for the treatment of end-stage knee osteoarthritis. Acute normovolemic haemodilution (ANH) can effectively replace the need for allogeneic transfusions due to the high amount of bleeding during TKA. However, more studies are needed to prove the efficacy and safety of ANH and to clarify its indications in the field of knee replacement. Medical records from June 1, 2019 to June 1, 2021 were searched and grouped according to inclusion and exclusion criteria. PART I: 58 patients with ANH during TKA were selected as the ANH group (n = 58), and 58 patients with allogeneic transfusion were chosen as the control group (n = 58). PART II: Patients with anaemia were divided into the ANH group (n = 18) and the control group (n = 12). PART I: The postoperative inflammatory index and serum albumin in the ANH group were significantly lower than those in the control group. No significant difference was observed in the theoretical loss of red blood cells, postoperative renal function, liver function, cardiac function and biochemical ion index between the two groups. The effective rate of ANH in the normal haemoglobin group was significantly lower than that in the anaemia group. PART II: In patients with anaemia, the theoretical loss of red blood cells in patients with ANH was less than that in the control group. The postoperative inflammation, renal function, liver function and cardiac function in the ANH group were better than those in the control group, and no significant difference was noted in biochemical ions and nutritional status indicators. This paper shows that ANH not only can replace allogeneic transfusion in TKA, especially in patients with anaemia, but also has lower inflammatory indicators than allogeneic transfusion. From a security perspective, the body’s tolerance to ANH is within the body’s compensation range.