A prospective bonus of platelet storage additive solutions: a reduction in biofilm formation and improved bacterial detection during platelet storage

“The characteristic of scientific progress is our knowing that we did not know.”Gaston Bachelard, French philosopher and poet, 1884-1962 Progress in our understanding of how to successfully store platelets (PLTs), ex vivo, for safe and effective transfusion therapy has been a long and arduous journey—one that continues to this day. It is worth noting a few of the seminal events along this path before exploring some of the new developments, concepts, and challenges involving PLT storage and transfusion safety. Such developments include the role of PLT storage additive solutions (PAS) and their impact on PLT bacterial contamination as well as an understanding of ex vivo biofilm formation in PLT storage bags. These developments are highlighted by Greco and coworkers in this issue1 as well as prior issues2, 3 of TRANSFUSION. This year represents the centennial year of our first understanding of the role of PLT transfusion in clinical medicine. In a report published in 1910, Duke4 reported on three thrombocytopenic, bleeding patients. Their bleeding ceased and PLT counts improved with transfusions of PLT-rich fresh whole blood. Since that time we have been challenged with how best to provide PLTs in a practical, efficacious, and safe manner. Thus, this capability would await the advent of the plastic storage bag, more than 40 years later, for PLT transfusion therapy to become therapeutically practical.5 Several more years would elapse, however, for investigators to determine that PLT function is optimally maintained by room temperature rather than refrigerated storage.6 The fact that room temperature was required for optimal PLT function provided early concern, which would prove prescient, for the possibility and occurrence of the growth of bacteria in PLT storage bags over time.6 This concern was quickly realized and subsequently validated by the work of Buchholz and coworkers. They demonstrated the predilection for PLT concentrates, stored at room temperature, to serve as a growth medium for bacteria in up to 2.5% of their study units.7, 8 These investigators were also the first to demonstrate that the incidence of PLT bacterial contamination is correlated with the storage age of the PLT unit and to suggest that initial contamination is the result of only a few contaminating bacterial organisms that subsequently proliferate throughout the room temperature storage period to become clinically significant.8 Thus, the older the PLT unit the more likely it is to contain an increased quantity of bacteria and to elicit clinically significant septic events associated with transfusion.9-11 This finding highlights one of the ongoing and daunting challenges related to PLT bacterial contamination—that early detection strategies, especially bacterial culture from samples procured from PLT units less than 2 days of storage age, now common practice based on CAP and AABB accreditation requirements, are greatly limited in their ability to detect the often small numbers of bacteria present in PLT units at that time. The sensitivity of bacterial detection in freshly stored units is estimated at only approximately 30% of the sensitivity for bacterial detection of units stored for 7 or more days.12-14 Thus, bacterial screening of PLTs, as is currently employed, is of limited clinical utility. The observation that older PLT units harbor a greater quantity of bacteria has led some groups to preferentially transfuse relatively fresh PLTs to reduce the occurrence of clinically overt PLT bacterial contamination.15, 16 Unfortunately, transfusion of relatively fresh PLTs, given their limited availability, is not practical as a widespread and uniform solution to reduce the occurrence of clinically significant PLT bacterial contamination. In addition while this strategy may avoid some instances of potentially clinically significant contamination by Staphylococcus epidermidis and other slow-growing organisms, it will not prevent transfusion of PLTs contaminated by fast-growing, mainly Gram-negative, organisms in which the level of bacteria reaches clinically significant quantities early in storage.17, 18 Finally, the use of PAS rather than plasma during PLT storage may alter the kinetics of bacterial growth, as described below, so that an increased quantity of bacteria is present earlier during PLT storage allowing, potentially, for improved detectability by early culture methods1, 19 as well as more efficient elimination by pathogen inactivation.20 Most of the work on the kinetics of bacterial growth in stored PLTs is based on the paradigm of storage in plasma. Data from in vitro models of bacteria inoculated into fresh whole blood21 or PLTs22 demonstrates an initial decrement phase in which the number of bacteria are reduced by naturally occurring bacteriostatic and bactericidal substances.23, 24 After this early reduction, there is a variable lag phase, in which the number of bacteria apparently remains stationary; a log growth phase in which the bacteria multiply exponentially; a plateau phase in which the bacteria reach an apparent equilibrium with their environment and exhibit no further growth; and a death phase in which bacteria, no longer sustainable by their environment, begin to die. Coagulase-negative staphylococci and, in particular, S. epidermidis, the most common organism implicated in PLT bacterial contamination25 is characterized by a prominent lag phase which often lasts at least 2 days during in vitro PLT storage.17 Traditionally, the lag phase is presumed to represent a period during which bacteria, adjusting to their new environment, do not actively grow.26 More recently, however, the lag-phase growth period of S. epidermidis in PLT concentrates has been investigated and challenged. Studies utilizing epifluorescence and scanning electron microscopy demonstrated the lag phase as a period during which the bacteria are actively growing—not as free-floating planktonic organisms in the fluid phase but rather as complex aggregates of PLTs and bacteria reaching sizes, during the first 3 days of PLT storage of up to 10,000 µm2.27 During this time sampling of these PLTs for bacterial quantitation by traditional culture techniques revealed no increase in the number of colony-forming units (CFUs) per mL present. Motoyama and coworkers27 thus note that the term “lag phase” is better termed a “pseudo-lag phase.” The authors suggest that these aggregates or microcolonies likely represent precursors of bacterial biofilm that form on the inner surface of bacterially contaminated PLT unit storage bags as described by Greco and coworkers.1-3 The interaction of PLTs with bacteria, in which activated PLTs actively surround and engulf bacteria via the open canalicular system,28 likely contributes to the pathogenesis of these microcolonies and aids in the transport of PLT-bacterial aggregates to multiple sites throughout the body. The concept of bacterial biofilm is of fundamental importance to understanding the life cycle of most bacteria in nature, as well as those involved in human infections, and has recently been extended to our understanding of PLT bacterial contamination.1-3 The natural state of most bacteria—either in the environment or within our bodies—is within biofilms.29 Bacteria differ in their ability to form biofilms. Those most able to do so are considered more virulent and exhibit a variety of genetic-based virulence factors. A bacterial biofilm is a sophisticated three-dimensional aggregate of bacterial organisms adherent to an underlying structure and to each other, protected by an extracellular polysaccharide polymeric matrix (giving these structures an appearance of slime) providing marked resistance to antimicrobial agents and having an ability to communicate via a quorum sensing system. The latter is thought to determine the overall size and shape of the biofilm. A bacterial biofilm may be composed of multiple different bacterial species (as in dental plaque) or as a single species, as in most instances of S. epidermidis.30 Cells that break off from a biofilm may maintain their virulence factors and initiate a new biofilm at a different location or circulate in a fluid phase as single, more vulnerable planktonic cells. At present, it is only these planktonic cells that we detect and measure when we sample PLT units for the presence of contaminating bacteria despite the fact that biofilms readily form on inner surface of plastic PLT storage bags, particularly at the air-liquid interface.3, 24 There are no validated detection strategies for bacterial biofilms in contaminated PLT units, albeit sometimes these are plainly visible in PLT units to the human eye as elegantly portrayed by Greco's group at the air-liquid interface.1 Thus, the development of new strategies to either detect or reduce biofilm formation during PLT storage should decrease the risk of transfusion-transmitted bacteria. Storage of PLTs in PAS rather than plasma1 reduces biofilm formation and favors planktonic growth,19 which should promote earlier detection of bacteria by current culture methods. Biofilm formation by coagulase-negative staphylococci, particularly S. epidermidis, has been extensively studied.30-35 The skin harbors at least 18 species of coagulase-negative staphylococci with at least 50% representing S. epidermidis and with an estimated 75% of clinical isolates from the skin comprising coagulase-negative staphylococci.33 Because of the frequency with which S. epidermidis colonizes the skin, it represents the most frequent organism to contaminate medical devices, such as catheters or phlebotomy devices that penetrate or in some way come in contact with the skin.30 PLT units are analogous to indwelling vascular catheters in being susceptible to contamination with and subject to biofilm formation by staphylococci. Although S. epidermis biofilms are the most common on plastic surfaces including PLT bags, many other bacterial species are capable of biofilm formation on plastic polymers including Staphylococcus aureus, Pseudomonas aeruginosa, Escherichia coli, Serratia marcescens, and Serratia liquefaciens.1, 3 In the same way that patients harboring a biofilm-containing catheter, even those associated with clinical sequelae, may fail to elicit a positive blood culture,3, 36 PLT units harboring biofilm may fail to elicit positive cultures yet represent a potential danger during transfusion. This observation emphasizes the importance of biofilm formation during PLT storage and the need to interdict this process. Isolates of S. epidermidis differ in their ability to form biofilms. While all or nearly all strains, depending on environmental conditions, are able to form some amount of biofilm, the presence of specific genetic factors is correlated with enhanced biofilm-forming ability.31, 33 The most important factor involved in biofilm formation is believed to be the polysaccharide intercellular adhesion (PIA) gene, which is part of the ica (intercellular adhesion) operon. The PIA gene product, a β-1,6-linked glucosaminoglycan, has been identified as a cell surface component of S. epidermidis and serves to mediate intercellular adhesion important in biofilm formation. While its absence does not preclude biofilm formation by alternate genetic mechanisms, its presence is believed to correlate with the virulence of S. epidermidis in clinical isolates.31, 33 Work by Greco's group2 demonstrates that isolates of S. epidermidis from bacterially contaminated PLTs identified clinically similarly differ in their biofilm genotype. It remains uncertain, however, if differences in the biofilm genotype correlate clinically with the severity of transfusion reactions associated with S. epidermidis or other bacterially contaminated PLT units. Work on this concept is ongoing.35, 37 The use of PAS, in various formulations, has been ongoing in Europe for more than a decade. The underlying driver was to create a solution, similar to that used for red blood cells, to optimize PLT storage in vitro. Not only is the ability to store PLTs for extended periods of time at room temperature limited due to the propensity of bacterial contamination, storage is also limited by loss in PLT quality—referred to as the PLT storage lesion.38 With the development of strategies to limit PLT bacterial contamination, it was hoped that attention could be directed to dealing with ways to delay the onset of the PLT storage lesion by use of one or more PAS formulations.39 Other incentives for the use of PAS include the ability to procure additional plasma for transfusion or fractionation, a reduction in adverse effects of transfusion linked to plasma such as allergic reactions, hemolysis secondary to passive infusion of hemolytic antibodies, and possibly transfusion-related acute lung injury.39 Finally, another driver for the development of PAS, in Europe and elsewhere,40 has been the development of PLT pathogen inactivation technologies, which function best when the amount of plasma in the storage medium is reduced. Pathogen inactivation technologies appear to be partially inhibited by plasma proteins. PASs replace approximately two-thirds of the original plasma providing a better platform for pathogen inactivation of viruses and bacteria.21, 41 Regarding bacteria, when standardized low-level quantities of bacteria typically implicated in PLT bacterial contamination42 were inoculated into apheresis PLT units, split into plasma and PAS units, and subjected to pathogen reduction using the Mirasol PRT System (CaridianBCT, Lakewood, CO), there was more successful eradication of bacteria in PAS with 35% plasma compared to 100% plasma. Specifically, while Klebsiella pneumoniae and E. coli were equally eradicated in both, one of two inocula of S. epidermidis was eradicated in plasma compared to six of six in PAS, and zero of two inocula of Streptococcus pyogenes was eradicated in plasma compared to four of six in PAS.20 Although very preliminary, it appears that PAS provides a better environment for eradication of bacteria at levels simulating those occurring in routine practice. A further potential advantage of PAS regarding PLT bacterial contamination is a reduction in biofilm formation during in vitro PLT storage compared to storage in plasma. Conceptually, one would anticipate greater biofilm formation when surfaces are coated with an increased concentration of plasma since bacteria, notably S. epidermidis, have surface proteins able to adhere to fibrinogen and fibronectin, which are present in large quantities in plasma. Such surface adherence is the necessary first step in biofilm formation.33 A reduction in the concentration of plasma, as in PAS, would reduce the quantities of these proteins and reduce, potentially, the ability of bacteria to attach to a given surface. This appears to be the case as recently demonstrated by Greco and coworkers.1 Two model bacterial strains—one Gram negative (S. liquefaciens) and one Gram positive (S. epidermidis)—selected for their slow growth characteristics in contaminated PLT units and corresponding biofilm-forming abilities—were inoculated at quantities of 10 and 100 CFUs/mL into split buffy coat PLT units containing PAS (SSP+, MacoPharma Canada, Montreal, Quebec, Canada) with 35% plasma or PLT units containing 100% plasma. Although on Day 5 of storage biofilms had formed in all PLT bags, there was an appreciable qualitative difference in the appearance by scanning electron microscopy, as well as an approximately 80% semiquantitative difference, by crystal violet staining, of the intensity of biofilm formation in the plasma bags compared to the PAS bags. In addition to a reduction in biofilm formation in PAS compared to plasma, S. liquefaciens exhibited a lag phase that extended approximately 24 hours more in plasma compared to PAS. This observation suggests, as reported by Motoyama and coworkers,27 the occurrence of a “pseudo-lag” phase in plasma in which growth is occurring, not by the planktonic organisms in suspension as measured by standard culture methods, but by biofilm formation on the surface of the PLT bag. Reduction or elimination of the lag phase in PAS compared to plasma suggests, as noted by the authors, more planktonic bacteria present earlier in storage which can be identified by traditional culture methods. Unfortunately, Greco and coworkers could not, despite an appreciable reduction in biofilm formation, demonstrate a similar reduction in the lag phase for S. epidermidis. Nonetheless, preliminary work19 using PAS InterSol (Fenwal, Inc., Lake Zurich, IL), recently licensed by the FDA for storage of Amicus PLTs43 demonstrated a significant difference in the lag phase overall for those bacteria growing in 100% plasma (9.5 ± 7.2 hr) compared to those growing in PAS with 35% plasma (4.8 ± 5.8 hr). This study further corroborates the work of Greco and coworkers1 and their suggestion that use of PAS may improve the safety of PLTs by allowing for earlier culture detection. Hence, to the list of potential advantages of PAS should be added the potential for improved bacterial detection. While we await further clinical and laboratory investigations on ways to reduce bacterial contamination and improve detection, it is noteworthy that a single multicenter study demonstrated that storage of buffy coats in the PAS T-Sol (Baxter, Unterschleißheim, Germany; or MacoPharma, Langen, Germany) resulted in the fewest units issued on average before the first positive growth signal using the BacT/ALERT (bioMérieux, Durham, NC), 45%, compared to buffy coat plasma units, 54%.44 While not significant, it is nonetheless consistent with possible faster planktonic growth of bacteria in PAS units. As PAS enters the marketplace in the United States, we should take the opportunity to closely monitor PLT bacterial contamination to determine if the potential advantages of PAS in improving the safety of PLT products are realized. However, continued vigilance, preferably with testing for bacterial contamination at or near time of use,45 will be needed until pathogen inactivation technology is available.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.