Background: Transfusion of cellular blood products during left ventricular assist device (LVAD) implantation has been associated with HLA allosensitization, resulting in the need for a negative prospective cross-match and prolonged transplant waiting times. In order to prevent this risk, we developed a protocol to avoid transfusion of cellular blood products.
Methods: The protocol included preoperative patient stabilization, perioperative recombinant erythropoietin and blood conservation strategies, and postoperative monitoring of mixed venous oxygen saturation (SVO2) to assure adequate peripheral oxygen delivery. Panel reactive antibody (PRA) was measured in all patients pre and post LVAD placement to assess HLA sensitization.
Results: Seven consecutive patients underwent LVAD implantation without transfusion of blood or platelets, one of whom expired perioperatively. Mean hematocrit was 35.2% preoperatively, and 21.8% postoperatively, reaching a nadir of 20.2%. Postoperative SVO2 was >60% in all patients. In the six survivors, mean hematocrit reach 24.3%, 27.3%, and 33.0% by postoperative day seven, fourteen, and thirty, respectively. PRA in three patients was 0% preoperatively and remained 0% until transplantation after 33, 34, and 50 days of support. In two patients, preoperative PRA was 7% and 17%, dropped to 3% and 0% after thirty days, then progressively rose to 96% and 100% after 60 and 90 days, respectively. In one other patient, preoperative PRA was 0%, remained at 0% after thirty days, then rose to 96% by 60 days.
Conclusions: Avoiding transfusion of cellular blood products in LVAD recipients is safe and well tolerated, but does not universally protect from HLA allosensitization. Other factors may also produce sensitization, such as immunogenic components of the LVAD, soluble antigen in fresh frozen plasma, or latent sensitization which is not initially evident in critically ill and possibly anergic patients.