In an attempt to replace the oncotic and protein coating capabilities of serum albumin in the perfusate, we established a priming protocol that used autologous blood as part of the perfusate solution. Prior to March 1, 1999, our standard priming protocol was 1650 ml of crystalloid with 250 ml of 5% serum albumin and 5,000 units of heparin. After removing albumin from our prime, our standard protocol was altered to include 40 ml of the patient's autologous blood in 1,800 ml of crystalloid and 10,000 units of heparin. To determine the intraoperative effects of using albumin/crystalloid primes (Group A), autologous blood/crystalloid primes (Group B) and crystalloid primes (Group C), a total of 178 patients were sequentially evaluated. Intraoperative parameters evaluated were total protein (TP), colloid osmotic pressure (COP), platelets (Plts) and fluid requirements during cardiopulmonary bypass (CPB). During an overlapping 12-month period of time, 1,092 consecutive cardiac surgical cases using CPB (584 albumin prime; 508 autologous blood prime) were evaluated for clinical outcomes in terms of mortality and length of hospitalization. In addition, over a period of 15 months, 1,458 patients in both the autologous blood/crystalloid group and the crystalloid only group were evaluated for the incidence of high-pressure excursions (HPE) after going on bypass. Comparative reviews of TP, COP and Plts demonstrated no significant difference 10 min after the start of bypass between Groups A and B. However, in Group C, there was a statistically significant increase in the intraoperative fluid requirements during CPB, compared to both of the other groups. There was no significant difference in the incidence of HPE, with an occurrence of 1.04% in the crystalloid only group and 1.11% in the autologous blood/crystalloid group. Autologous blood perfusates were identical to albumin perfusates in their platelet protection and reduction of fluid shifts during the intraoperative period.