Intraoperative autotransfusion in aortic surgery

Abstract

Purpose: Intraoperative autotransfusion is frequently used in aortic surgery, despite the paucity of data regarding its safety and efficacy. This study was designed to compare whole blood autotransfusion with homologous transfusion for the replacement of blood lost during abdominal aortic procedures.

Methods: Whole blood autotransfusion was evaluated in 200 patients undergoing aortic reconstructive procedures during a 3-year period. Collection and reinfusion of unwashed filtered shed blood was undertaken in 100 patients, and clinical, laboratory, and economic parameters were compared with those in a group of 100 patients undergoing aortic operation with homologous banked blood replacement.

Results: The two groups were comparable with respect to demography, the type of procedure, baseline laboratory profile, and the frequency of coexistent medical illnesses. The amount of blood salvaged and reinfused averaged 1729 +/- 68 ml in the autotransfusion group. Patients undergoing autotransfusion received a mean of 0.6 +/- 0.1 units of banked blood during operation, compared with 3.4 +/- 0.1 units in the homologous group (p < 0.001). Operative morbidity and mortality rates were comparable between the groups, as were length of hospital stay and total hospital costs. Coagulopathy, renal insufficiency, abnormalities of oxygen exchange, and electrolyte disorders were infrequent. Autotransfusion offered significant advantages over homologous blood replacement with respect to improved preservation of circulating platelets (201 +/- 9 vs 157 +/- 6 x 10(3)mm3, p < 0.001) and coagulation factors (242 +/- 11 vs 196 +/- 14 mg fibrinogen/dl, p < 0.01) and limitation of exposure to homologous blood (34% vs 92%, p < 0.001). There was a significant cost advantage with the use of autotransfusion, with an average savings of $288 in hospital expenses associated with blood products and infusion equipment. Patients undergoing autotransfusion demonstrated aberrations in fibrin degradation products (33 +/- 4.4 vs 9.6 +/- 3.2 micrograms/ml, p < 0.001) and free plasma hemoglobin (29 +/- 9.1 vs 9.4 +/- 0.5 mg/dl, p < 0.05), but these laboratory abnormalities did not acquire clinical significance.

Conclusion: These data suggest that autotransfusion of unwashed, filtered blood is a safe and efficacious alternative to homologous blood replacement in patients undergoing major aortic reconstructive procedures.