Growth and Distribution of Bacteria in Contaminated Whole Blood and Derived Blood Components

Abstract

Introduction: Bacterial contamination of blood products presumably occurs mainly during blood collection, starting from low initial concentrations of 10-100 colony-forming units (CFUs) per bag. As little is known about bacterial growth behavior and distribution in stored whole blood (WB) and WB-derived blood products, this study aims to provide data on this subject.

Methods: WB units were inoculated with transfusion-relevant bacterial species (Acinetobacter baumannii, Bacillus cereus, Escherichia coli, Klebsiella pneumoniae, Listeria monocytogenes, Pseudomonas fluorescens, Serratia marcescens, Staphylococcus aureus, Staphylococcus epidermidis, Streptococcus dysgalactiae, Streptococcus pyogenes, Yersinia enterocolitica; n = 12 for each species), stored for 22-24 h at room temperature, and then centrifuged for separation into plasma, red blood cells (RBCs), and buffy coats (BCs). The latter were pooled with 3 random donor BCs and one unit of PAS-E each to yield plasma-reduced platelet concentrates (PCs). Samples for bacterial colony counting were collected after WB storage and immediately after blood component production. Sterility testing in PCs (n = 12 for each species) was performed by bacterial culture after 7 days of storage.

Results: Bacterial growth in WB varied remarkably between donations and species. Streptococcus species produced the highest titers in WB, whereas Staphylococcus aureus, Staphylococcus epidermidis, Escherichia coli, and Pseudomonas fluorescens did not multiply. Centrifugation resulted in preferential accumulation of bacteria in BCs, with titers of up to 3.5 × 10³ CFU/mL in BCs and up to ≤0.9 × 10³ CFU/mL in BC-derived PCs. Overall, 72/144 PCs (50%) tested positive for bacteria after storage. Sterility test results were species-dependent, ranging from 12 of 12 PCs tested positive for Streptococcus pyogenes to 1 of 12 PCs positive for Escherichia coli. Bacterial contamination of RBC and plasma units was much less common and was associated with higher initial bacterial counts in the parent WB units.

Conclusions: Bacterial growth in WB is species-dependent and varies greatly between donations. Preferential accumulation of bacteria in BCs during manufacturing is a critical determinant of the contamination risk of BC-derived pooled PCs.

Keywords: Bacterial contamination; Buffy coat platelets; Whole blood.