Donor screening as a TRALI risk reduction strategy

Since its first description by Popovsky and Moore in 1985,1 transfusion-related acute lung injury (TRALI) has become recognized as the most frequent cause of mortality and morbidity after blood transfusion.2-5 There have been continuing efforts to raise awareness of the condition among physicians and improve its diagnosis and treatment. Most importantly, blood services are now aiming to implement interventions to prevent its occurrence. Since the predominant mechanism of severe cases of TRALI is the transfusion of plasma from female donors with HLA- or granulocyte-specific antibodies,6-9 strategies for TRALI reduction should logically target high-plasma-volume blood components. These include fresh-frozen plasma (FFP) made from whole blood or apheresis, plasma frozen within 24 hours after its collection (FP-24), platelet (PLT) concentrates prepared by apheresis, buffy coat–derived PLTs resuspended in the plasma of one of the donors in the pool, and whole blood. The policy of sourcing plasma from males is already showing signs of success in TRALI reduction. Another possible intervention is to prospectively screen female donors for the causative antibodies of TRALI, and this approach may be particularly applicable to donors of apheresis PLTs where the loss of female donors would compromise the PLT supply. The definition of TRALI agreed at a consensus conference in Toronto in 2004 is widely accepted as an international standard.10 The clinical diagnosis of TRALI was based on the American-European Consensus Conference (AECC) definition of acute lung injury11 and a temporal relationship to transfusion. The definition rightly prioritized accurate documentation of the history, physical examination findings, and investigations performed before and after an episode of hypoxia, noting the timing of transfusion. It has been suggested that there may be less florid cases than those defined above resulting from similar mechanisms, which may result in less severe symptoms and signs and/or a delayed onset.12 In the 2003 Serious Hazards of Transfusion (SHOT) annual hemovigilance report in the United Kingdom, TRALI was implicated in 8 of 12 (66%) deaths resulting from transfusion, with an overall mortality of 25% for cases referred as TRALI.2 Of the 19 cases with documented white blood cell (WBC) incompatibility, FFP and PLTs were implicated on eight occasions each and whole blood, cryoprecipitate, and buffy coats each in one event. No donors of red blood cells (RBCs) transfused to patients with TRALI had detectable WBC antibodies. In 2005 and 2006 in the United States, 64 of 125 (51%) deaths resulting from transfusion reported to the Food and Drug Administration were attributed to TRALI.4 Plasma products were implicated in 39 of 64 (61%) cases, apheresis PLTs in 6 of 64 (9%), RBCs in 10 of 64 (17%), and a mixture of different components in 9 of 64 (14%). Like the United Kingdom, this shows a greater risk from components with a significant amount of donor plasma. The initial description of TRALI by Popovsky and Moore1 implicated antibodies against WBCs. The infusion of WBC antibodies without any other obvious precipitant causes TRALI in ex vivo and in vivo animal models as well as otherwise healthy individuals, supporting their hypothesis.13, 14 However, the pathogenesis of TRALI remains unclear.15 Some cases do not appear to be associated with WBC antibodies, and the presence of biologically active lipids in stored blood components, causing neutrophil priming, has been associated with the presence of TRALI.16 Patients with hematologic and cardiac disease have been shown to be at higher risk than other patients.17 Two events may be needed to cause TRALI: a predisposing clinical condition and a transfusion containing WBC antibodies or biologically active lipids. As recently reported in TRANSFUSION, an analysis of 96 TRALI cases reported to the SHOT hemovigilance scheme in the United Kingdom found donor antibodies against cognate antigens in the affected recipient in 73 cases (65%), with HLA Class I antibodies in 25 of those (40%), HLA Class II antibodies in 38 (62%), and granulocyte antibodies in 12 (17%).18 As expected, WBC antibodies were found significantly more frequently in donors implicated in a case of TRALI than in the normal donor population in a pilot study performed to assess the impact of introducing donor screening.19 In this pilot UK study of 1416 donors, 15% of random female donors were positive for HLA Class I and II antibodies, compared to 2.4% of males. The percentage of female donors with HLA antibodies increased with a greater number of pregnancies. Granulocyte-specific antibodies were found in 1.7% of females and 1.4% of male donors. Approximately half of the granulocyte-specific antibodies were specific for a defined antigen of the HNA system. In 2004, the Canadian Consensus Panel recommended that interventions to reduce the risk of TRALI should be evaluated.10 The Consensus Panel noted that the two main hypotheses for the cause of TRALI suggest very different strategies for its prevention. The antibody hypothesis points to exclusion of plasma from donors with pathogenic antibodies, whereas the biologically active lipid, neutrophil priming hypothesis supports a reduction in the storage age of blood components or washing blood components regardless of donor characteristics. The Consensus Panel noted that both approaches have the potential to reduce the blood supply by excluding safe donors (antibody hypothesis) or safe products (neutrophil priming hypothesis). It did not recommend a specific risk reduction strategy, although it did agree that a general strategy of practice guidelines should be adopted for the use of blood components to minimize inappropriate blood use. Recent efforts for TRALI reduction have focused on excluding plasma from donors with pathogenic antibodies. No countries appear to have implemented a strategy for a reduction in the storage age of blood components or washing blood components.20 The preferential use of male plasma for FFP was introduced in the United Kingdom in late 2003, and the annual number of cases of TRALI documented by the SHOT hemovigilance scheme has decreased since then.18 The data for the 2008 reporting year for SHOT have recently become available; there were 17 cases of TRALI in 2008, a reduction from the 24 cases reported in 2007.4 Although four patients died, these were cases considered “unlikely” to be caused by TRALI but due to other causes. There were therefore no deaths from TRALI in 2008, for the first time since SHOT reporting began. The three cases of proven TRALI associated with FFP all occurred due to the transfusion of FFP from female donors in England, where the use of male plasma for FFP is not universal. Policies of supplying FFP from predominantly male donors or use of pooled solvent/detergent–treated FFP, which appears not to cause TRALI,21, 22 have been widely implemented worldwide.20 In the United Kingdom, female plasma is not used for the resuspension of buffy coat–derived PLTs. A similar policy of removing female donors from plateletpheresis panels would clearly have a serious impact on supply of PLT concentrates. An alternative is to screen female apheresis donors for WBC antibodies. In this issue of TRANSFUSION, Triulzi and colleagues23 report a prospective study, titled the Leukocyte Antibody Prevalence Study (LAPS), which was designed to measure the prevalence of HLA Class I and Class II antibodies in a large number of blood donors with or without a history of pregnancy or blood transfusion. The target enrollment for the LAPS was 5700 females, 1100 transfused males, and 1100 nontransfused males from six US blood centers. Females and nontransfused males were enrolled without any specific recruitment strategy, but targeted recruitment was required to enroll sufficient numbers of transfused males. Pregnancy history was established using a validated questionnaire to elicit a history of all pregnancies, including miscarriages and terminations. Donors were asked if they had ever had a blood transfusion, the number of transfusions, and the date of the last transfusion. Testing for HLA Class I and Class II antibodies was carried out in a central laboratory using a fluorescein-tagged multiantigen bead assay. Testing was also carried out on a subset of donors for granulocyte-specific antibodies but, disappointingly, the results were not reported in this publication. Participation of donors was more than 90%, and the final donor sample was 7841, including 3992 females with a history of pregnancy, 1816 never pregnant females, 1138 nontransfused males, and 895 transfused males. HLA antibodies were detected in 24.4% of females with a history of pregnancy, 1.7% never pregnant females, 1.0% of nontransfused males, and 1.7% of transfused males. The overall incidence of HLA antibodies was higher than the UK study, but the increase in the prevalence of HLA antibodies in women with a greater number of pregnancies was similar: 11.2% (one), 22.5% (two), 27.5% (three), and 32.2% (four or more pregnancies). Transfusion alone did not result in an increase in the prevalence of HLA antibodies; the small difference between transfused and nontransfused males was not significant. However, the time since the last transfusion in the donors in the LAPS was very long (median, 22 years). The study did not evaluate recently transfused individuals, and therefore the possibility that a recent transfusion history is associated with a significant prevalence of HLA antibodies was not excluded, even if the antibodies subsequently become undetectable in a matter of months. The reason for HLA antibody formation in never pregnant females and nontransfused males is uncertain, as is their clinical significance. They could be due to false-positive results, antibodies cross-reacting with bacterial antigens, or immunization associated with vaccination. A limitation of these studies is the lack of a gold standard for HLA antibody testing that would allow the selection of a definitive cutoff for positivity in screening tests. The use of different techniques and the selection of different cutoffs probably explains the variation in HLA antibody positivity between studies. What are the implications of these results? How can they be used to determine a policy for TRALI reduction associated with PLT transfusion? Recent studies indicate that 38% of apheresis donors in the United States are female24 and that 33% have had no pregnancies, 12% one pregnancy, 23% two, and 32% three or more pregnancies.9 Based on the results of the LAPS, the loss of plateletpheresis donors would be limited to 6% by the exclusion of previously pregnant females with HLA antibodies as a TRALI reduction strategy. In 2004, the Canadian Consensus Panel suggested that blood services explore options for a reduction in the risk of TRALI, resulting in a benefit that would outweigh any possible disadvantages such as a decrease in the blood supply. The preferential use of male plasma for FFP is one measure that has been introduced in many countries with emerging evidence of its effectiveness in reducing the risk of TRALI, but no countries have yet introduced a national policy of screening donors with a history of pregnancy or transfusion for HLA- and granulocyte-specific antibodies. The LAPS provides valuable data indicating that a policy of testing female plateletpheresis donors with a history of one or more pregnancies for HLA antibodies is feasible and would not result in a major loss of donors if those with detectable HLA antibodies are excluded. Such a policy has already been successfully implemented in a regional blood center in Spain with a similar loss of female plateletpheresis donors as in the LAPS.25 The ultimate confirmation of the success of TRALI risk reduction strategies is required from clinical studies, which will most realistically come from national and international hemovigilance data, although a clinical study of products from implicated donors in the LAPS is under way. There will clearly be operational challenges in the implementation of the inclusion of a history of pregnancy in the screening of female plateletpheresis donors and testing for HLA antibodies, but these should not prove more difficult or costly to overcome than other successfully implemented, and often extremely costly, interventions to minimize the risk of transfusion-transmitted infection.26 Blood services need to weigh carefully the potential benefits of TRALI reduction strategies against their costs and risks such as their impact on the blood supply, but continually bear in mind that TRALI is currently the most frequent serious adverse event associated with blood transfusion. The authors declare no conflict of interest.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.