Processing methods and storage duration impact extracellular vesicle counts in red blood cell units.

Extracellular vesicles (EVs) are active components of red blood cell (RBC) concentrates and may be associated with beneficial and adverse effects of transfusion. Elucidating controllable factors associated with EV release in RBC products is thus important to better manage the quality and properties of RBC units. Erythrocyte-derived EVs (EEVs) and platelet-derived EVs (PEVs) were counted in 1226 RBC units (administered to 280 patients) using a standardized cytometry-based method. EV size and CD47 and annexin V expression were also measured. The effects of donor characteristics, processing methods, and storage duration on EV counts were analyzed by using standard comparison tests, and analysis of covariance was used to determine factors independently associated with EV counts. PEV as well as EEV counts were higher in whole-blood-filtered RBC units compared with RBC-filtered units; PEV counts were associated with filter type (higher with filters associated with higher residual platelets), and CD47 expression was higher on EEVs in RBC units stored longer. Multivariate analysis showed that EEV counts were strongly associated with filter type (P < .0001), preparation, and storage time (+25.4 EEV/µL per day [P = .01] and +42.4 EEV/µL per day [P < .0001], respectively). The only independent factor associated with PEV counts was the residual platelet count in the unit (+67.1 PEV/µL; P < .0001). Overall, processing methods have an impact on EV counts and characteristics, leading to large variations in EV quantities transfused into patients. RBC unit processing methods might be standardized to control the EV content of RBC units if any impacts on patient outcomes can be confirmed. The IMIB (Impact of Microparticles in Blood) study is ancillary to the French ABLE (Age of Transfused Blood in Critically Ill Adults) trial (ISRCTN44878718).

Genotypic study of Citrobacter koseri, an emergent platelet contaminant since 2012 in France.

BACKGROUND: Transfusion-transmitted bacterial infection is a rare occurrence but the most feared complication in transfusion practices. Between 2012 and 2017, five cases of platelet concentrates (PCs) contaminated with the bacterial pathogen Citrobacter koseri (PC-Ck) have been reported in France, with two leading to the death of the recipients. We tested the possibilities of the emergence of a PC-specific clone of C. koseri (Ck) and of specific bacterial genes associated with PC contamination. STUDY DESIGN AND METHODS: The phylogenetic network, based on a homemade Ck core genome scheme, inferred from the genomes of 20 worldwide Ck isolates unrelated to PC contamination taken as controls (U-Ck) and the genomes of the five PC-Ck, explored the clonal relationship between the genomes and evaluated the distribution of PC-Ck throughout the species. Along with this core genome multilocus sequence typing approach, a Ck pan genome has been used to seek genes specific to PC-Ck isolates. RESULTS: Our genomic approach suggested that the population of C. koseri is nonclonal, although it also identified a cluster containing three PC-Ck and eight U-Ck. Indeed, the PC-Ck did not share any specific genes. CONCLUSION: The elevated incidence of PCs contaminated by C. koseri in France between 2012 and 2017 was not due to the dissemination of a clone. The determinants of the recent outbreaks of PC contamination with C. koseri are still unknown.

In vitro quality of amotosalen-UVA pathogen-inactivated mini-pool plasma prepared from whole blood stored overnight.

BACKGROUND AND OBJECTIVES: Small batch-pooled (mini-pool) whole blood (WB)-derived plasma could be an alternative cost-effective source of therapeutic plasma (TP), but carries an increased risk of transfusion-transmitted infection due to exposure of the recipient to several donors. This risk can be mitigated by inactivation of pathogens susceptible to the amotosalen-UVA (AUVA)-treatment. We evaluated the conservation of coagulation factors in AUVA-plasma prepared from WB stored overnight under routine operating conditions, to determine its therapeutic efficacy. Thrombin generation (TG) by the AUVA-plasma was used to provide an integrated measure of the hemostatic capacity. MATERIALS AND METHODS: WB-donations (~450 ml) stored overnight were processed to prepare five leucocyte-depleted plasma mini-pools (1300 ml), which were divided into two parts and treated with AUVA. Each mini-pool yielded six AUVA-plasma units (200 ml) which were frozen (-25°C) within 19 h of WB-collection. Their hemostatic quality was evaluated before and after treatment for up to 12 months of storage. RESULTS: Immediately after AUVA-treatment, the regulatory criteria for FVIII activity and fibrinogen content were met. As compared to untreated plasma there was a reduction in fibrinogen (14%), FV (9%), FVII (25%) and FVIII (32%). However, TG was similar in treated and untreated plasma at all-time-points. CONCLUSIONS: Frozen WB-derived AUVA-plasma prepared from mini-pools within 19 h of WB-collection met the quality standards required for TP and retained hemostatic capacity for up to 12 months. This product could provide a cost-effective convenient substitute for apheresis plasma.

How to quantify microparticles in RBCs? A validated flow cytometry method allows the detection of an increase in microparticles during storage.

BACKGROUND: The procoagulant and proinflammatory microparticles (MPs) released during storage of packed red blood cells (pRBCs) can potentially modify transfusion benefits. A robust method to quantify MPs in pRBCs is needed to evaluate their impact in clinical trials. STUDY DESIGN AND METHODS: The objective was to validate the preanalytic conditions required to prepare pRBC supernatant as well as a method to quantify and evaluate MP variations over 42 days of pRBC storage.A flow cytometry method with size-calibrated beads was developed and fully validated. Quantification of MPs in pRBCs (n = 109) was assessed during short-term (7 days) and long-term (42 days) storage at 4°C, during short-term storage (8 hours) at room temperature, and after 2 years frozen. RESULTS: Repeatability, reproducibility, and linearity of the quantification method were validated, and variations during conservation are presented. There was high variability in RBC (erythrocyte) MP (ERMP) and platelet MP (PMP) levels between RBC units, depending on the filter used for leukocyte reduction. During the 42 days of storage at 4°C, significant increases in ERMPs and PMPs occurred (from 58 to 138 ERMPs/µL from Day 2 to Day 42; p = 0.0002; and from 326 to 771 PMPs/µL from Day 2 to Day 42; p = 0.00026). CONCLUSION: We use a robust method to confirm that ERMPs and PMPs are present to various degrees in pRBCs and that storage for 42 days significantly increases their generation. This method is robust enough to allow MP quantification in pRBCs and is adapted to evaluate the clinical impact of transfused MPs in prospective clinical trials.

Plasma transfusion in liver transplantation: a randomized, double-blind, multicenter clinical comparison of three virally secured plasmas.

BACKGROUND: The clinical equivalence of plasma treated to reduce pathogen transmission and untreated plasma has not been extensively studied. A clinical trial was conducted in liver transplant recipients to compare the efficacy of three plasmas. STUDY DESIGN AND METHODS: A randomized, equivalence, blinded trial was performed in four French liver transplantation centers. The three studied (fresh-frozen) plasmas were quarantine (Q-FFP), methylene blue (MB-FFP), and solvent/detergent (S/D-FFP) plasmas. The primary outcome was the volume of plasma transfused during transplantation. Secondary outcomes included intraoperative blood loss, hemostasis variables corrections, and adverse events. RESULTS: One-hundred patients were randomly assigned in the MB-FFP, 96 in the S/D-FFP, and 97 in the Q-FFP groups, respectively. The median volumes of plasma transfused were 2254, 1905, and 1798 mL with MB-FFP, S/D-FFP, and Q-FFP, respectively. The three plasmas were not equivalent. MB-FFP was not equivalent to the two other plasmas, but S/D-FFP and Q-FFP were equivalent. The median numbers of transfused plasma units were 10, 10, and 8 units with MB-FFP, S/D-FFP, and Q-FFP, respectively. Adjustment on bleeding risk factors diminished the difference between groups: the excess plasma volume transfused with MB-FFP compared to Q-FFP was reduced from 24% to 14%. Blood loss and coagulation factors corrections were not significantly different between the three arms. CONCLUSION: Compared to both Q-FFP and S/D-FFP, use of MB-FFP was associated with a moderate increase in volume transfused, partly explained by a difference in unit volume and bleeding risk factors. Q-FFP was associated with fewer units transfused than either S/D-FFP or MB-FFP.