Red blood cell (RBC) concentrate (RCC) transfusion, a life-saving intervention, can cause in rare cases side effects. These adverse events may be related to the patient's medical condition and/or to the properties of the transfused RCC. The quality of RCCs depends on the manufacturing process, the storage conditions and donor characteristics. ln this context, different observational studies have been conducted evaluating the risk of death of the transfused patient in relation to the RCC storage duration or to the donor's and recipient's sex. 1n both cases, no clear association could be established.
To understand what happens during transfusion at the cellular and molecular level as well as to assess potential effects of RBC characteristics on the patient and vice versa, we developed, as a first approach, a simple transfusion model. ln vivo conditions were simulated such as temperature and plasma composition with nutrient sufficiency and cell waste limitation aiming at maintaining RBCs in a physiological-close state. RBCs from RCCs (representing the donor) at different storage days, were incubated in plasma (representing the recipient) either at a physiologie-close haematocrit (HCT) (30 %) or at a low HCT (5 %), at 37 °C for up to 48 h. Haematological parameters, metabolism and cell morphology and deformability were monitored. The low HCT allowed an increased availability of glucose boosting the energetic metabolism and limited lactate accumulation avoiding thus medium acidification which would lead to a metabolic slow down due to a decreased activity of pH-sensitive glycolytic key enzymes. Consequently, the 5 % model - showing a better RBC preservation - was retained to model in vitro transfusion. An artificial "universal" plasma (HPLM) was tested as well at 5 % HCT. lts universality allows evaluating RBC properties in a "recipient"-independent transfusion context. RBCs haemolysed more in HPLM than in plasma which may be due to the absence of plasma lipids and proteins. However, haemolysis remained quite low which makes HPLM suitable for this model.
The 5 % HCT model was used to evaluate the effect of donor and recipient sex on RBCs (in plasma) and to evaluate the sex-dependent RBC properties independently of the "recipient" sex (in HPLM). Despite a rather similar energy metabolism of RBCs, RBCs from men haemolysed more than those from women did in plasma. A reason may be that female-derived RBCs better stored during cold storage than male ones thanks to increased antioxidant defense capacities. ln addition, female plasma showed a protective effect against haemolysis of RBCS from both sexes which may be due to a sex-dependent plasma composition of hormones and various bioactive lipids. Incubation in HPLM confirmed the dependence of haemolysis difference on the plasma donor sex.
ln summary, the developed in vitro transfusion model is suitable for studying RBC behavior and effects of storage in a transfusion context. The discoveries in sex-related experiments are in agreement with clinical data and would require further investigations.