THE EFFECT OF MULTIPLE BLOOD CONSERVATION TECHNIQUES ON DONOR BLOOD EXPOSURES IN ADULT CORONARY AND VALVE SURGERY PERFORMED WITH A MEMBRANE-OXYGENATOR - A MULTIVARIATE-ANALYSIS ON 1310 PATIENTS

The object of the study was to retrospectively evaluate protective and risk factors for receiving donor blood products and red cell transfus ions after coronary and valve surgery performed with a hollow-fiber ox ygenator and with multiple blood-saving techniques. During the period of January 1991 to June 1993, 1310 patients underwent primary coronary and valve surgery using a hollow-fiber oxygenator at our institution; the mean age of this population was 61 +/-10 years; 977 patients were men (74.6%). Of these patients, 73.5% (963/1310) underwent coronary, 21.5% (281/1310) valve, and 5% (66/1310) combined surgery. Two hundred seventy-six (21.1%) needed donor blood product transfusions, while 15 3 (11.7%) patients underwent red cell transfusions. Significant risk f actors for homologous blood product exposure after multivariate logist ic regression analysis were, in order of importance: (1) postoperative blood loss (O.R. = 1.0009 per mL, p = 0.0000); (2) cardiopulmonary by pass (CPB) time (O.R. = 1.008 per min, p 0.0001); (3) age at intervent ion (O.R. = 1.031 per calendar year, p = 0.0026); and (4) reoperation for bleeding (O.R. = 1.71, p = 0.0078). Protective factors were: (1) m ale gender (O.R. = 0.56, p = 0.0000); (2) preoperative withdrawal of a utologous blood (O.R. = 0.66, p = 0.0018); and (3) a preoperative hema tocrit greater than 34% (O.R. = 0.76, p = 0.005). When considering ris k factors only for donor red cell exposure, multivariate regression an alysis identified the following risk factors: (1) reoperations for ble eding (O.R. = 2.04, p = 0.0002); (2) postoperative blood losses (O.R. = 1.0007 per mt, p = 0.0005); (3) CPB time (O.R. = 1.0075 per min, p = 0.0008); and (4) age at intervention (O.R. = 1.03 per calendar year, p = 0.0160). Protective factors were: (1) intraoperative ''high-dose'' aprotinin administration (O.R. = 0.61, p = 0.0024); and (2) preoperat ive donation if autologous blood (O.R. = 0.65, p = 0.0093); and (3) in traoperative withdrawal of autologous blood by phlebotomy (O.R. = 0.67 , p = 0.0114). Applying multiple blood-saving techniques, coronary and valve procedures can now be performed with a low incidence of postope rative donor blood products and red cells exposures; autologous blood predonation and aprotinin administration were highly effective in redu cing postoperative transfusion needs.