Is it time for new initiatives in the blood center and/or the hospital to reduce bacterial risk of platelets?

Platelets (PLTs) are stored at room temperature because they survive better than when stored in refrigerators.1 Room temperature storage, however, permits growth and transmission of certain bacteria.2-4 Blood facilities have implemented procedures to reduce risk of bacterial transmission with PLTs. Whole blood and apheresis blood from which PLTs are made are now collected in kits that have a pouch that diverts the first 30 to 50 mL of blood from the donor capturing some skin-contaminating bacteria and reducing the risk (46%-71%) of bacterial contamination of PLTs.5-7 Better skin cleansing procedures and materials have further reduced the risk of transmitting bacteria.7-9 Most North American and European blood centers now culture PLTs early in their storage period to detect contaminating bacteria.10-17 Rates of bacterial detection using early culture vary by testing and manufacturing procedure. Recent reports show a range of 1 in 4329 to 1 in 7210 rate of detection.14,18,19 Early testing plus surveillance testing performed at the end of PLT storage (5-7 days) suggest that the rate of true bacterial contamination is likely higher, with rates of detection in the range of 1 in 1072 to 1 in 1183.18,20 The implication of these findings is that current QC culture-based screening of PLTs appears to have a sensitivity as low as 24% to 40% and that a significant number of bacterially contaminated PLTs escape early detection and are transfused. The residual risk for septic transfusion reactions in patients as a result of this breakthrough has been a challenge to quantify. Septic transfusion reaction risk in recent years has been reported as high as 1 in 6437 and as low as 1 in 83,333.6,21,22 Primarily contributing to this large difference in reported septic transfusion reaction rate is the method of surveillance. Voluntary reports of adverse clinical events are received at a rate that is much lower than expected based on estimated breakthrough events, but passive surveillance is known to lead to underrecognition of adverse events.23 Some definitions of septic transfusion reaction have increased the reported rate of reaction, because these definitions include mild symptoms only and/or lack criteria that are normally used to define clinical sepsis.24 Today, despite some uncertainty surrounding the extent of the clinical problem, the residual risk of transfusion-transmitted bacterial sepsis remains even with recent significant interventions. This reality has raised both the commercial and the clinical interest in a point-of-issue test for bacteria in PLT units. Three studies published in this issue of TRANSFUSION document the risk remaining after early testing in various fashions. The study by Jenkins and colleagues25 is a comprehensively chronicled report on a series of interventions intelligently applied by the Canadian Blood Services to reduce risk and improve quality. It unfortunately documents that these interventions have not eliminated residual risk. The study by Ramirez-Arcos and colleagues,26 also from Canadian Blood Services, using the Verax PLT pan genera detection (PGD) test to evaluate outdated buffy coat–derived PLTs has shown that contaminated units continue to escape detection by the early testing technology; this study also demonstrates that the use of the bioMérieux anaerobic bottle would result in significant PLT wastage due to the high number of false-positive results. Finally, in a report of a large commercially sponsored study, Jacobs and coworkers27 proposed the application of the PGD test to reduce this residual risk by testing on the “day of transfusion” in the hospital. The FDA licensed the Pall eBDS and the bioMérieux BacT/ALERT tests as “QC tests” for transfusable PLT components and recommended their use. Most blood centers culture every PLT unit in hope of interdicting contaminated units before transfusion. While both tests are sensitive, neither can detect all potentially contaminated PLT units because the test is performed at approximately 24 hours of storage, and units with bacterial contamination below the limits of detection at 24 hours can have a contamination level of clinical significance by the time they are transfused on Day 3, 4, or 5 of storage. One way to look at what we have learned from our experience with the application of the current collection and testing procedures is that PLTs tested early with eBDS or BacT/ALERT have a safe period of X hours during which they are presumed to have an acceptable level of risk from sepsis, after which bacteria that may have been present at undetectable levels may propagate to achieve clinical significance. When the tests were first introduced, it was hoped that the safe period (X) would be at least 96 hours (4 days), indicating that the risk of sepsis was acceptable for the entire duration of PLT storage. At one point, it was actually hoped that X would be 6 days; hence the Passport study of 7-day PLTs, which was designed to potentially allow extension of the shelf life of cultured PLTs. We have since learned that X is shorter than 96 hours when eBDS or BacT/ALERT cultures are performed at 24 hours, a period not adequate to achieve near-zero risk. Similarly, previously cultured PLTs tested with PGD have a safe period of (Y) hours after testing. Because the PGD test is approximately 1000-fold less sensitive than eBDS and BacT/ALERT, the PGD safe period Y is shorter than X and perhaps 1 day or less. The PGD test was designed and licensed to reduce the risk of sepsis from apheresis PLT units previously tested by culture at 24 hours. The PGD test is also used within 4 hours of transfusion for whole blood–derived PLTs pooled in an open system, the standard shelf life for PLTs processed in an open system. This test produces a negative or positive result more quickly than eBDS and BacT/ALERT.23,28,29 An idea behind the PGD test was to emphasize simplicity and rapid turnaround time, thereby making it a point-of-issue test, but at a cost of less analytic sensitivity compared to culture tests. Analytic sensitivity is less important when the transfusion is administered within a very short time after the test is initiated. Thus, clinical sensitivity of a point-of-issue test may be sufficient to prevent adverse consequences depending on how soon after testing the PLT is transfused. The manufacturer of the PGD test states that the analytical sensitivity or limit of detection is in the range of 8.2 × 103 to 8.6 × 105 colony-forming units (CFUs)/mL depending on the organism, although some reports indicate that the sensitivity for certain strains of Gram-negative organisms (Escherichia coli and Krebsiella pneumoniae) are poorer than this claim30 and some organisms are missed despite being at a CFUs/mL concentration greater than the limit of detection.27 It is known that contamination at or above 105 CFUs/mL has caused serious symptoms and even death.21 The manufacturers of the eBDS and the BacT/ALERT test have indicated sensitivity in the range of 1 to 10 CFUs/mL.23,31,32 The PGD test would not be appropriate to replace cultures as a test to be initiated 24 hours after collection because of its relative lack of analytic sensitivity. Similarly, the eBDS and the BacT/ALERT would detect a greater proportion of contaminated units if they are done late rather than early in the storage period. Jacobs and colleagues27 showed that when QC-negative PLTs (10,424 units) are recultured at the same time the PGD test is applied (day of transfusion), culturing detects contaminated units (5 units) that are missed by the initial QC test. The PGD test detected 60% (3) (rate = 2.9/10,000) of the 5 units (rate, 4.8/10,000) detected by simultaneous culture. In their study, Jacobs and colleagues applied the PGD test within the day of transfusion, which is unfortunately too ill defined to definitively establish safety gain relative to practical application of the test.27 Interpreting the data from this study is further complicated by the fact that 30% of the PLTs tested (including 3 of the 9 true-positive units) were tested after the PLTs were transfused. Including the data from the posttransfusion testing in calculating benefit is questionable. Although it is clear that when the test is used before transfusion to interdict contaminated units the test reduces risk, without knowing the times of testing and the times of transfusion, it is difficult to characterize the duration of the safer period (Y) after PGD testing and the true benefit of implementing the test. In the study by Jacobs and coworkers, none of the participating hospitals PGD tested 100% of PLTs in time to interdict all positive units, so ease of test performance allowing universal application has not been documented and is a serious issue. Heaton and colleagues33 aware of the benefits of current good manufacturing practices (cGMP), described the system applied at North Shore University Hospital. Although it was the plan to test products scheduled on the day of transfusion, complicating issues included placing PLTs received from the blood center on hold (quarantine), generating appropriate labels, sampling in an appropriate fashion, utilization of appropriate positive and negative run controls, relabeling, resolving initially reactive results by retesting in duplicate (which in the study by Jacobs' and coworkers reduced the false-positive rate by approximately 50%), handling split components, and ensuring timely communication with the supplying blood center for cocomponent and donor management. This last point is essential because rapid communication from the testing hospital to the collection site will ensure the timely quarantine of potentially infectious cocomponents (red blood cells [RBCs], PLTs, and plasma can be collected with apheresis PLTs). The goal of Heaton's team was to provide PGD-tested PLTs for routine orders, but not for irregular orders. Utilizing this system, 59% of the PLTs issued during the study period were PGD tested. The US FDA has driven improvement in the safety and quality of the US blood supply partially by emphasizing use of cGMP, including adherence to FDA-approved manufacturers' instructions. The requirement for cGMP has facilitated the increase in blood center component production. Among the reasons that RBCs and PLTs are filtered in blood centers and not in hospitals is the fact that the application of cGMP ensures that each unit is treated according to standard procedures that achieve high levels of leukoreduction and confidence that white blood cell contamination is below a specific threshold. While the circumstances are quite different, the same concerns apply to the ability of busy hospitals to test PLTs at the appropriate time and in a manner to interdict the transfusion of positive (and false-positive) units. Hospitals operate in a clinical environment and they expect blood centers to apply cGMP to provide components that are sufficiently safe for transfusion without further testing. With the current data available in the refereed literature, including the study by Jacobs and coworkers, it is not apparent that the application of a pretransfusion test in hospitals is the only logical or feasible next step to reduce the risk of sepsis for PLT recipients. Day-of-transfusion testing (PGD or other) has potential advantages and it may be possible to document effectiveness in the future. Some hospitals might conclude that it is impossible to perform a day-of-transfusion test in a fashion to interdict 100% of the potentially contaminated units, but still wish to utilize the test. If so they may introduce a day-of-transfusion test with a strategy that some but not all PLTs will be tested, which would reduce risk but would lead to different levels of safety within the hospital and the problem of selecting and identifying which patients would receive day-of-transfusion–tested PLTs, or hospitals may wish only to perform initial testing and not repeat positive samples in duplicate, which would effectively remove 1% of PLTs from the inventory and create cocomponent management issues. While such a step would improve safety by eliminating contaminated PLTs, it would complicate the manufacturing process, lead to the loss of safe PLTs, and increase costs. Further consideration should be given to alternative strategies. More can be done to reduce the risk of sepsis from PLT transfusion, but additional data are needed to determine whether in-hospital or in-blood-center interventions (or combinations) are the most desirable. France and other developed and developing countries have adopted pathogen reduction to reduce the risk of sepsis and other risks from PLT transfusion.34-36 While this may be the most effective intervention, it will not be feasible to apply this intervention in the United States in the near future, regardless of its advantages for patients. Canada is also evaluating pathogen reduction technology and has implemented the preparation of pooled buffy coat PLTs to improve effectiveness and/or clinical outcomes and reduce risks from PLT transfusion. If pathogen reduction can be implemented in Canada, the residual risk of sepsis may be reduced nearly to zero without the need for point-of-issue testing of PLTs (and perhaps without the need for early culturing as in some European countries). What additional approaches to risk reduction might be taken by regional blood centers? A centralized approach could produce products of uniform safety in a community and has the potential to be less expensive and less disruptive than introducing blood component safety procedures in the hospital transfusion service. If blood centers are not using a diversion pouch and culturing with larger sample volumes, when applicable, they should. A more complicated intervention would be to culture a specimen from each split unit when multiple PLT units are collected by apheresis from one donor during one donation. This step would have the advantage of increasing the cultured sample size relative to the transfused PLT volume and thus increase the sensitivity and positive predictive value19,26 of the system as well as provide needed information when one split is found to be initially reactive. Jenkins and colleagues and other investigators have documented a trend toward increasing sensitivity with an increase in the percentage of the PLT volume tested.6,19,25 Theoretically one would expect an increase in sensitivity by an additional 25% to 45% (depending on a double or triple product) based on modeling studies that show the adequacy of the sample is a function of the bacterial concentration.6,37,38 These advantages would be balanced by disadvantages that include decreasing the split rate because of the increased volume taken from the unit and the cost of an increased number of tests. An even more complicated intervention would be to perform later sampling of PLTs in the blood center inventory (second, third, or fourth day) if they are not needed immediately. Testing later has the advantage of increased sensitivity to contamination. How much longer would the safe period (X) be for PLTs tested with a more sensitive culture test compared to the period Y for simultaneously PGD tested PLTs? Jacobs and colleagues posit that Y for the PGD is approximately 1 day.27 If indeed, the PGD test can provide 1 day (Y) of safety, it is possible that culture tests could provide a safer period (X) longer than 1 day. If the application of the culture tests later in the PLT storage period is considered, the time to negativity or positivity may be an issue. For example, when PLTs are tested 24 hours after collection using 8 mL in an aerobic BacT/ALERT bottle, 100% of the true positives are detected within 18 hours, and 60% are detected within 12 hours.19 In the study by Ramirez-Arcos and coworkers,26 a PLT contaminated with Staphylococcus epidermidis at 2.8 × 107 CFUs/mL, a level of contamination more likely to be seen later in the storage than at 24 hours, produced a positive BacT/ALERT bottle in 4 hours. BacT/ALERT testing of 48-hour-old PLTs showed a mean time to positivity, which was approximately 2 hours (6 hr vs. 8 hr) shorter than when the same PLTs were sampled at 24 hours.28 Times to positivity are shorter if Day 2, Day 3, and/or Day 4 units are tested,39,40 but not likely to be as short as the 30 to 90 minutes needed to perform one or multiple PGD tests.33,41 The PGD test could also be performed in the blood center routinely or for emergency release. These efforts might move the burden of increased safety from the hospital to the blood center, where cGMP is already utilized to improve safety and control costs. Specific examples may be useful. Blood centers could inoculate all PLT cultures at 48 hours instead of 24 hours and they could be released if negative not more than 12 hours after inoculation. This would increase sensitivity because the units contaminated with low levels of bacteria may increase in concentration to the level of detectability with the extra 24 hours of presampling incubation. In this example there would be a tradeoff of PLT function for bacterial safety. An alternative would be to test a portion of the PLTs collected each day at 24 hours leaving some PLTs unscreened. Each evening the blood center would test a sufficient number of previously unscreened 2-, 3-, 4-, or 5-day-old PLTs at a time to allow appropriate incubation before release that would result in a suitable safer period for patient therapy. If blood centers managed their inventories to do later culture or PGD testing, it would simplify procedures for hospitals, control costs, and reduce the risk to patients. Such procedures might support the safe release of 7-day PLTs. Testing later would not increase the number of test kits used, but would provide some challenges for information management and labeling that need to be resolved. Do we need a next step—an additional intervention to make PLTs safer? The current risk to patients of sepsis from PLT transfusions is lower now than it was 10 years ago. Hospital clinicians are not demanding improvements in PLT safety at present. Nevertheless, the residual risk of septic reactions is troubling to those clinicians and blood bankers who recognize that the risks (per component transfused) of transmitting hepatitis B virus (1 in virus (1 in and hepatitis virus (1 in are all lower than the risk of PLTs transmitting bacteria that result in are other serious issues to transfusion that are at present and that will hospital or blood center is less likely to than it was 10 years in most North American and European but remains a serious to patients and there is more to be transfusion reactions are less than they were 10 years but they also a serious to reactions will significant hospital but the improvements in patient outcomes should be Hospitals and should be In we have made a of in the safety of PLT transfusions bacterial contamination the improvement in skin the diversion of the first 30 to 50 mL of and the application of early PLT culture. The of the PGD test (or alternative point of issue tests in with rapid time to negativity a potential way to further improve safety. utilization of the that have safety to and that cost is an additional should as are The application of pathogen reduction technology is very and could be a but the in the United States is not While blood center testing of each split unit at 24 hours would increase the increase in safety from testing the split units should be Blood centers should also the way the current culture are used to include testing later in the PLT storage period to improve safety and shelf Hospitals and blood centers should these interventions and the that a current or point-of-issue test might be developed for use in blood centers or in The have of of this to Verax has been by the FDA to the PGD test in a although the test is The PGD is a for the detection of aerobic and anaerobic and Gram-negative bacteria in apheresis within 24 hours to transfusion of as a safety testing with a quality control test by FDA for components and of to 6 reduced and reduced whole blood that are pooled within 4 hours of transfusion as a quality control test. this of the PGD test as a safety we have the not to the but it is that the a safer period X and after the test during which the patient risk is good The What data are to determine the duration of the safer Testing and transfusion times were not documented in the report by Jacobs and of the negative PGD results were with units which were tested 4 hours to issue for One of the was with a septic transfusion reaction, but the time of the test in to the transfusion was not it that PLT units tested with a point-of-issue test be and found to be negative 24 hours later to establish a safe the Pall and bioMérieux to a safer period the use of their culture-based tests sensitive than when they are applied to previously tested (or apheresis or of whole or 7-day PLTs be safe to if they been tested with the PGD test (or with eBDS or within 24 The to reduce risk of bacterial transmission with units are but the in safety remains difficult to quantify.

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.