Number needed to treat and cost of recombinant human erythropoietin to avoid one transfusion-related adverse event in critically ill patients

Objective: To calculate the absolute risk reduction of transfusion-related adverse events, the number of patients needed to treat, and cost to avoid one transfusion-related adverse event by using erythropoietin in critically ill patients Design: Number needed to treat with sensitivity analysis. Setting: Teaching hospital. Patients: Hypothetical cohort of critically ill patients who were candidates to receive erythropoietin. Interventions: Using vs. not using erythropoietin to reduce the need for packed red blood cell transfusions. Measurements and Main Results: We used published estimates of known transfusion risks: transfusion-related acute lung injury, transfusion-related errors, hepatitis B and C, human immunodeficiency virus, human T-cell lymphotropic virus, and bacterial contamination, stratified by severity. Based on the estimated risk and frequency of transfusions with and without erythropoietin, we calculated the absolute risk reduction of transfusion-related adverse events, the number needed to treat, and cost to avoid one transfusion-related adverse event by using erythropoietin. The estimated incidence of transfusion-related adverse event was 318 permillion units transfused for all transfusion-related adverse events, 58 per million for serious transfusion-related adverse events, and 21 per million for likely fatal transfusion-related adverse events. The routine use of erythropoietin resulted in an absolute risk reduction of 191 per million for all transfusion-related adverse events, 35 per million for serious transfusion-related adverse events, and 12 per million for likely fatal transfusion-related adverse events. The number needed to treat was 5,246 to avoid one transfusion-related adverse event, 28,785 to avoid a serious transfusion-related adverse event, and 81,000 for a likely fatal transfusion-related adverse event. The total cost was $4,700,000 to avoid one transfusion-related adverse event, $25,600,000 to avoid one serious transfusion-related adverse event, and $71,800,000 to avoid a likely fatal transfusion-related adverse event. The magnitude of these results withstood extensive sensitivity analysis. Conclusions: From the perspective of avoidance of adverse events, erythropoietin does not appear to be an efficient use of limited resources for routine use in critically ill patients.