OBJECTIVE: To study transfusion requirements in total knee anthroplasty (TKA) in order to estimate needs and consider possible ways to improve principled management. PATIENTS AND METHODS: Retrospective study of all TKAs performed over a period of 3 months in 2000. We analyzed age, medication, hypertension, participation in a predeposit autologous blood program (PABP), drainage bleeding, transfusion requirements and hemoglobin (Hb) levels. RESULTS: One hundred ten TKAs were performed. Ten were excluded from analysis (2 TKA reviews, 5 patients with coronary artery disease and 3 with chronic bronchial diseases). We reserved a mean 2.46 +/- 0.5 units of packed red cells (range 2-3), and 30% of the patients were transfused with a mean 0.62 +/- 0.99 units (range 0-4). In 15 cases (50%), transfusions were not clearly indicated or were related only to the surgical procedure. All 5 PABP patients were transfused, but not with homologous blood, even through 4 had Hb levels over 9 g/dL. Only Hb level was an independent risk factor for transfusion. Transfusions were given to 80% of patients with Hb levels under 12 g/dL, to 18.8% of those with hemoglobin levels over 14 g/dL, and to 35.7% of those with Hb levels between 12 g/dL and 14 g/dL. CONCLUSIONS: Preoperative Hb level was the single predictor of transfusion in our series of patients. In some cases the transfusion trigger was too liberal. Alternatives to homologous transfusion were hardly used, with PABP applied in only 5% of the cases. These findings encourage us to continue trying to improve the PABP program and transfusion criteria, to introduce alternatives and to improve baseline Hb levels.
OBJECTIVE: To study transfusion requirements in total knee anthroplasty (TKA) in order to estimate needs and consider possible ways to improve principled management. PATIENTS AND METHODS: Retrospective study of all TKAs performed over a period of 3 months in 2000. We analyzed age, medication, hypertension, participation in a predeposit autologous blood program (PABP), drainage bleeding, transfusion requirements and hemoglobin (Hb) levels. RESULTS: One hundred ten TKAs were performed. Ten were excluded from analysis (2 TKA reviews, 5 patients with coronary artery disease and 3 with chronic bronchial diseases). We reserved a mean 2.46 +/- 0.5 units of packed red cells (range 2-3), and 30% of the patients were transfused with a mean 0.62 +/- 0.99 units (range 0-4). In 15 cases (50%), transfusions were not clearly indicated or were related only to the surgical procedure. All 5 PABPpatients were transfused, but not with homologous blood, even through 4 had Hb levels over 9 g/dL. Only Hb level was an independent risk factor for transfusion. Transfusions were given to 80% of patients with Hb levels under 12 g/dL, to 18.8% of those with hemoglobin levels over 14 g/dL, and to 35.7% of those with Hb levels between 12 g/dL and 14 g/dL. CONCLUSIONS: Preoperative Hb level was the single predictor of transfusion in our series of patients. In some cases the transfusion trigger was too liberal. Alternatives to homologous transfusion were hardly used, with PABP applied in only 5% of the cases. These findings encourage us to continue trying to improve the PABP program and transfusion criteria, to introduce alternatives and to improve baseline Hb levels.