[Impact of acute normovolemic hemodilution on primary hemostasis]

Acute normovolemic hemodilution (ANH) is performed with the intention to reduce the requirement for allogeneic blood transfusions. After preoperative withdrawal of whole blood, corresponding amounts of crystalloids and/or colloids are infused to maintain normovolemia. The main benefit of ANH is the availability of whole blood containing red blood cells, clotting factors and platelets for reinfusion after removal during the dilution process. Until retransfusion whole blood components are stored at the patient's bedside in the operating theatre.

It was the aim of the present investigation to analyze potential changes in ex vivo induced platelet aggregation in stored blood components.

After obtaining approval 15 patients undergoing complex cardiac surgery were enrolled into this prospective observational study. Acute normovolemic hemodilution (ANH) was routinely performed in this collective based on institutional standards. Besides analyses of pH and plasma concentrations of ionized calcium and hemoglobin, hematological analyses included aggregometric measurements using multiple electrode aggregometry (MEA, Multiplate®, Roche, Grenzach, Germany). Ex vivo platelet aggregation was induced using arachidonic acid (ASPI test), as well as thrombin receptor activating peptide (TRAP test) and adenosine diphosphate (ADP test). Laboratory analyses were performed before beginning ANH (baseline), as well as immediately (T1), 30 min (T2), 60 min (T3), 90 min (T4), 120 min (T5), 150 min (T6) and 180 min (T7) after beginning of storage. The areas under the aggregation curves (AUC) in the MEA were defined as primary (ASPI test) and secondary endpoints (ADP test, TRAP test).

As compared to baseline, arachidonic acid induced platelet aggregation was significantly reduced at T1 [77 U (68/94 U) vs. 53 U (25/86 U), p = 0.003] and each consecutive measuring point. As compared to T1 (begin of storage), arachidonic acid induced platelet aggregation was significantly reduced at T4 [26 U (14/54 U); p = 0.002], T5 [30 U (21/36 U); p = 0.007], T6 [25 U (17/40 U); p = 0.004] and T7 [28 U (17/39 U); p < 0.001]. The extent of ex vivo induced platelet aggregation in the TRAP test and ADP test remained unchanged during the study period. The pH as well as the concentrations of ionized calcium and hemoglobin remained unchanged in the blood component during storage.

The results of the present study indicate that disturbances of platelet aggregation may occur during storage of whole blood components prepared for the purpose of ANH. Further investigations are needed to analyze whether the observed phenomena are of hemostatic relevance.