Predicting blood loss and transfusion requirement in patients undergoing surgery for musculoskeletal tumors

Abstract

Few studies have systematically identified factors associated with blood loss in musculoskeletal tumor surgery. We aimed to identify risk factors for requiring large-volume transfusion in musculoskeletal tumor surgery and created an interactive model to predict red blood cell transfusion requirements based on patient characteristics. These data will facilitate planning in hospital blood banks and aid identification of specific groups for future interventions targeted at reducing blood utilization. Only one similar study has been published and there are minimal data surrounding interventions designed to minimize blood loss in musculoskeletal tumor surgery. We retrospectively analyzed a database containing 1322 consecutive surgeries, performed at a quaternary referral center in Melbourne, Australia. Using logistic regression analysis and a negative truncated binomial logistic regression model, we developed prediction models for transfusion requirement. The following factors were associated with large-volume transfusion: malignant tumors, bone tumors, sacral and pelvic tumors, high American Society of Anesthesiologists (ASA) score, and tumor size of more than 5 cm. High ASA score was also strongly associated with 30-day mortality. Preoperative planning in high-risk patients is critical to ensure adequate blood product supply, minimize wastage, and optimize the patient's general health before surgery. These patients would be ideal targets for future randomized studies aimed at reducing blood utilization.