The efficacy of postoperative autotransfusion in patients undergoing cardiac operations.

The efficacy of postoperative autotransfusion in lowering the requirement for banked-blood transfusion was studied in two groups, each having 168 patients, who underwent cardiac operations between April, 1979, and May, 1980. A Sorenson autotransfusion system was available for use in the autotransfusion group, whereas the control group received routine closed mediastinal drainage. Of the autotransfusion group, 81% met the criterion for autotransfusion (mediastinal losses of 450 ml or more during 4 hours), but only 61% of the autotransfusion group actually received autologous blood (mean autotransfusion volume, 399 +/- 25 ml). The patients receiving autologous blood required significantly less banked blood than their matched controls (447 +/- 60 ml and 744 +/- 83 ml, respectively; p less than 0.001). In the subgroup of patients with large mediastinal losses (more than 1,250 ml), this difference was even greater (autotransfusion, 642 ml compared with control, 1,145 ml; p less than 0.01). Postoperative autotransfusion is a simple, safe, and cost-effective method to reduce dependence on banked blood, especially when mediastinal losses are large. Obtaining maximum benefit requires familiarity of staff with the system and use of a consistent protocol.