Cost-benefit analysis of autologous blood donation in obstetrics

To determine whether there are predictors of peripartum transfusion, other than placenta previa, that identify a population of pregnant women whose risk of transfusion is high enough to justify antepartum autologous blood donation.

Using an established perinatal data base, we studied 14,267 consecutive term deliveries without placenta previa. Univariate and multivariate analyses were performed to assess ten predictors of peripartum transfusion that might reasonably be detected in the antepartum period. Costs were calculated for a hypothetical autologous blood donation program to prevent transfusion-related infection.

Red-cell transfusion was used in 150 deliveries (1.1%). A total of 424 units was transfused (2.9 per 100 deliveries). Four risk factors were significantly (P less than .05) predictive of peripartum red-cell transfusion: preeclampsia (adjusted odds ratio 3.69), multiple gestation (2.82), elective cesarean (1.71), and nulliparity (1.51). Controlling for these, there was no association between transfusion and previous postpartum hemorrhage, previous cesarean with trial of labor, prior abortions, induction of labor, or ethnic group. A hypothetical antepartum blood donation program restricted to patients with three or more risk factors would cost $32,800-130,700 per case to prevent transfusion-related hepatitis and $26,000,000-78,000,000 per case to prevent human immunodeficiency virus infection.

In obstetric patients without placenta previa, the need for peripartum red-cell transfusion cannot be predicted with sufficient accuracy to justify the costs of antepartum autologous blood donation.