INTRAOPERATIVE AUTOLOGOUS TRANSFUSION DURING ELECTIVE INFRARENAL AORTIC RECONSTRUCTION - A DECISION-ANALYSIS MODEL

Purpose: The use of intraoperative autologous transfusion devices expa nded during the last decade as a result of the increased awareness of transfusion-associated complications. This study was designed to deter mine whether routine use of an intraoperative autologous transfusion d evice (Haemonetics Cell Saver [CS]) during elective infrarenal aortic reconstructions is cost-effective ($50,000/QALYs threshold). Methods: A decision analysis tree was constructed to model all of the complicat ions that are associated with red blood cell replacement during aortic reconstructions for both abdominal aortic aneurysm (AAA) and aortoili ac occlusive disease (AIOD). It was assumed that a unit of CS return ( CSR; 250 ml/unit) equaled a unit of packed red blood cells (PRBCs) and that all CS transfusions were necessary. Transfusion requirements (AA A: PRBC = 2.8 +/- 3.2 units, CSR = 3.7 +/- 3.2 units; AIOD:PRBC = 3.1 +/- 3.0 units, CSR = 2.1 +/- 1.7 units) were determined from retrospec tive review of all elective aortic reconstructions (AAA, N = 63; AIOD, N = 75) from Jan. 1991 to June 1995 in which the CS was used (82.1% o f all reconstructions). Risk of allogenic transfusion-related complica tions (transfusion reaction, hepatitis B, hepatitis C, human immunodef iciency virus, human T-cell lymphotropic virus types I and II) and the ir associated treatment costs (expressed in dollars and quality-adjust ed life years (QALYs) were obtained from the medical literature, insti tutional audit, and a consensus of physicians. Results: Routine use of the CS during elective infrarenal aortic reconstructions was not cost -effective in our practice. Use during reconstructions for AAA repairs cost $263.75 but added only 0.00218 QALY's, for a rate of $120,794/QA LY. Use during reconstructions for AIOD was even more costly at $356.6 8 and provided even less benefit at 0.00062 QALYs, for a rate of $578, 275/QALY. The sensitivity analyses determined that the routine use of the CS would be cost-effective in our practice only for AAA repairs if the incidence of hepatitis C were tenfold greater than the baseline a ssumption. The model determined that CS was cost-effective if the CSR exceed 5 units during reconstructions for AAA and 6 units during recon structions for AIOD. Conclusions: The routine use of the CS during ele ctive infrarenal aortic reconstructions is not cost-effective. The use of the device should be reserved for a select group of aortic reconst ructions, including those in which cost-effective salvage volumes are anticipated. Alternatively, the CS should be used as a reservoir and a ctivated as a salvage device if significant bleeding is encountered.