Existing literature suggests that there is institutional variation in transfusion practices that may be related to the use of a hemoglobin-based """"""""transfusion trigger"""""""" at some institutions. Given the known risks and costs of transfusions, the decision to transfuse should be based on a careful consideration of the risks and benefits of the procedure. However, little is known about the benefit (or lack of benefit) of transfusions in different clinical settings, and this may be one reason for the continued use of a hemoglobin-based """"""""transfusion trigger"""""""" by some surgeons and/or institutions. Therefore, the Division of General Internal Medicine, Robert Wood Johnson Medical School proposes to perform a multi-center retrospective study to 1) describe the independent risk factors for perioperative transfusions in patients with hip fracture.
The aim of this analysis is to determine whether the decision to transfuse is driven primarily by patient- and procedure-related variables (e.g. age, co-morbid disease, blood loss) or is significantly influenced by institutional practice styles independent of patient and procedure characteristics, and 2) describe the effect of transfusions on the operative death rate in patients with hip fracture by comparison with similar patients not receiving transfusions. The study population will include 10,000 hip fracture patients who had surgical repair of the fracture between 1983 and 1992. The population will be drawn from hospitals located in one of four geographical locations: Philadelphia, PA; New Brunswick, NJ; San Antonio, Texas and Richmond, VA. Patient information will be obtained through a retrospective review of medical records. In the first part of the study, we will develop a logistic regression model to describe the independent predictors of transfusion. In the second part of the study, we will first describe the univariate relationship between transfusion status and post-operative mortality stratified by hemoglobin level. We will then use logistic regression control for the effects of important confounding variables. This analysis will allow us to describe the independent effect of transfusion on the risk of post-operative death at various hemoglobin levels. Given that the risk of blood transfusions is significant and the expense related to maintaining the blood supply is enormous, new research is needed to define more sophisticated clinical guidelines for transfusion that may improve patients' outcomes while conserving the precious supply of blood. The proposed research, in conjunction with the investigators earlier and ongoing studies of anemia and surgery, will help to provide a cohesive picture of transfusion practices. This information will serve as an important step in defining and communicating more sophisticated transfusion guidelines.
