ABSTRACT
STUDY OBJECTIVE: To compare, by model simulation, acute normovolemic hemodilution (ANH) and preoperative autologous blood donation (PABD) to predict their efficacy in current clinical practice. To discuss their similarities and offer guidelines based on expected operative blood loss. DESIGN: Model simulation using data obtained from total hip arthroplasty procedures. SETTING: University medical center. PATIENTS: 91 patients who participated in PABD undergoing single, primary, total hip replacement surgeries from January to December 1997. INTERVENTIONS: A nonanemic (Hb baseline 14 g/dL), average-sized patient (estimated blood volume 5,000 mL) who donated two units by either PABD or ANH was assumed for model simulation. The Hb-final, as a function of 250-mL blood loss increments after retransfusion of two autologous units, was calculated for each technique and compared to a control (nondonor) to predict the effectiveness of PABD and ANH in conserving red cell mass. Data from hip arthroplasties over a 12-month period were used to determine the parameters for the model. Results were subjected to regression analysis and tested for parallelism of slopes, with p < 0.05 accepted to indicate a statistical difference. MAIN RESULTS: The difference in Hb-final between PABD and ANH was not statistically different over a wide range of blood loss. When compared to control, there was no difference in Hb-final measurements in the estimated blood loss (EBL) range of less than 1,000 mL. ANH and PABD provide some benefit when EBL is greater than 1,500 mL in nonanemic, average-sized patients. As blood loss increases, the benefit of autologous collection becomes more clinically evident. CONCLUSION: PABD exemplifies a "chronic" form of ANH in current clinical practice and offers little advantage over ANH as a blood conservation technique for high-blood-loss operations. When surgical blood loss is predictably mild to moderate (range 250 to 1,000 mL), neither blood conservation technique is necessary.
