OUP user menu

Effects of Prestorage vs Poststorage Leukoreduction on the Rate of Febrile Nonhemolytic Transfusion Reactions to Platelets

Rachel R. Wang MD, Darrell J. Triulzi MD, Lirong Qu MD, PhD
DOI: http://dx.doi.org/10.1309/AJCP5H7EKZTGGBKZ 255-259 First published online: 1 August 2012


The purpose of this study was to assess the incidence of febrile nonhemolytic transfusion reactions (FNHTRs) to concurrent transfusions of prestorage-leukoreduced (PreSLR) pooled platelets, poststorage-leukoreduced (PostSLR) pooled platelets, nonleukoreduced (NonLR) pooled platelets, and apheresis single-donor platelets (SDPs) to compare the rates of FNHTRs to PreSLR vs PostSLR pooled platelets. Reported transfusion reactions to platelets at 15 hospitals for a period of 45 months were retrospectively reviewed. Reaction rates to different types of platelet products were calculated and compared. During the study period, 70,015 platelet transfusions were administered. Among these, 152 (0.22%) FNHTRs and 111 (0.16%) allergic transfusion reactions were seen. Reported rates of FNHTRs were 0.07% (SDP), 0.16% (PreSLR), 0.30% (PostSLR), and 0.20% (NonLR) (P < .05 for PreSLR vs PostSLR). Rates of allergic reactions were 0.16% (SDP), 0.17% (PreSLR), 0.18% (PostSLR), and 0.11% (NonLR) (P > .05). The rates of reported FNHTRs were low for all types of platelet transfusions. SDPs and PreSLR pooled platelets were associated with a slightly lower rate of FNHTR compared with PostSLR pooled platelets.

Key Words
  • Transfusion reactions
  • Leukoreduction
  • Platelets

Febrile nonhemolytic transfusion reactions (FNHTRs) are characterized by fever (≥1°C elevation) that may be accompanied by chills, rigors, hypertension, tachycardia, and dyspnea without another clinical explanation.1 Leukocytes and cytokines released by leukocytes in the transfused allogeneic blood components are important causes of FNHTRs,1,2 but other mediators such as platelet-derived soluble CD154 (CD40 ligand) are also likely involved in its etiology.3,4 Leukoreduction removes leukocytes from cellular blood components and has been shown to reduce the rate of FNHTRs to RBC and platelet transfusion.1,58 Leukoreduction of pooled whole blood platelet concentrates can be performed at any point during their storage. Prestorage leukoreduction (PreSLR) refers to removal of WBCs soon after component preparation (ie, before storage) whereas poststorage leuko-reduction (PostSLR) usually occurs just before the products are issued for transfusion. Single donor platelets (SDPs) are typically leukoreduced at the time of apheresis collection. Leukoreduction, especially when performed at the time of issue from the blood bank, has limited efficacy in reducing the rate and severity of FNHTRs to platelets because the level of cytokines in the platelet concentrates is correlated with the age of the platelets.911 Therefore, to effectively prevent or decrease the frequency of FNHTRs to platelet transfusion, leukoreduction should ideally be performed at the time of collection or shortly thereafter. Leukoreduction would not reduce the amount of CD154 elaborated by the platelets themselves, thus it is unlikely that leukocyte filtration by itself can completely eliminate the risk of FNHTRs.

Although PostSLR has been the mainstay of leukoreduction in our transfusion service, in July 2008 we began using a device approved by the US Food and Drug Administration which permits prestorage pooling, leukoreduction, and bacterial detection of whole blood platelets (Acrodose, Pall Medical, East Hills, NY). This study was performed to test the hypothesis that the rate of FNHTRs to PostSLR pooled platelets was higher than that to PreSLR pooled platelets or SDPs.

Materials and Methods

Patients and Clinical Data

The centralized transfusion service (CTS) provides blood components and transfusion medicine clinical services to 15 local hospitals in the greater Pittsburgh (PA) area. Selective leukoreduction is used for patients with clinical indications for leukoreduced blood components. The most common indications are organ or stem cell transplantation, hematologic or solid organ malignancy, cardiac surgery, or a history of FNHTR. All patients at the local children’s hospital receive leukoreduced blood components. SDPs are primarily reserved for patients who require human leukocyte antigen (HLA)–matched platelets. All other patients receive nonleukoreduced (NonLR) blood components. For this study, data were collected on all platelet transfusions between August 2007 and February 2010 at 15 hospitals served by CTS.

Suspected transfusion reactions were reported to the hospital’s blood bank by the clinical team who used a standard form across all hospitals; the form includes pertinent patient information as well as signs and symptoms observed during the reaction. Investigation for hemolysis included clerical check, visual inspection for free plasma hemoglobin, and direct antiglobulin test (DAT). Transfusion reactions were evaluated by a single team of transfusion medicine physicians who reviewed the patient’s clinical and laboratory data and rendered an interpretation. For this report, transfusion reactions were classified as acute hemolytic, delayed hemolytic, febrile nonhemolytic, allergic (mild, moderate, and anaphylactic), septic, unrelated to transfusion, or caused by the patient’s underlying disease according to established criteria.1 Allergic transfusion reactions to the different platelet types, which are caused by a mechanism distinct from FNHTR and are not affected by the timing of leukoreduction, were also enumerated to serve as a control for any potential reporting bias between the different platelet types. Reaction rates were calculated by dividing the number of reactions to each type of platelet component by the total number of platelets of that type transfused during the study period. The protocol was approved by the University of Pittsburgh Medical Center’s Total Quality Council.

Platelet Components

SDPs were collected through apheresis and leukoreduced at the time of collection (Trima, CoridianBCT, Lakewood, CO). PreSLR pooled platelets (Acrodose) were manufactured from whole blood platelets within 36 to 48 hours after the donation. PostSLR pooled whole blood platelets were leukoreduced using the PXL8c filter (Pall Medical) in the blood bank immediately before issue. Patients with an indication for leukoreduced pooled platelets received exclusively PostSLR pooled platelets before July 2008. In July 2008, the transfusion service added PreSLR pooled platelets to its inventory and they have been used interchangeably with PostSLR platelets for patients with indications for leukoreduced blood components. The standard issuing practice was to use the oldest platelets first. The standard pool size for PreSLR, PostSLR, and NonLR platelet doses was 5 whole blood platelet units. Transfusion of a 5-unit platelet pool or 1 SDP was considered 1 transfusion episode.

Statistical Analysis

The rates of FNHTRs to the different types of platelet products were compared using the χ2 test. The ages of the platelets associated with transfusion reactions were compared with the ages of platelets that were not associated with reactions using 1-way analysis of variance. A P value of less than .05 was considered significant.


During the study period spanning 45 consecutive months, 70,015 platelet doses were transfused Table 1. There were 152 (0.22%) FNHTRs and 111 (0.16%) allergic transfusion reactions reported. The rates of FNHTR to SDPs, PreSLR, PostSLR, and NonLR pooled platelets were 0.07%, 0.16%, 0.30%, and 0.20%, respectively (Table 1). The rate of FNHTRs to SDPs and PreSLR pooled platelets was significantly lower than that to PostSLR pooled platelets Figure 1. The rate of FNHTRs to NonLR platelets was also lower (P = .045) than that to PostSLR platelets (Figure 1). No significant differences were found in the rates of allergic reactions among the different types of platelet products (P > .05 between any groups) (Table 1).

View this table:
Table 1

For PostSLR and NonLR platelets, the average age of platelets associated with FNHTRs were older than that of platelets not associated with reactions Figure 2. The average age of the SDP, PreSLR, PostSLR, and NonLR platelets that were implicated in FNHTR was 4.1, 4.6, 4.3, and 4.4 days, respectively. The proportion of older (4- to 5-day-old) platelets in pools associated with FNHTRs was significantly higher than that of pools not associated with FNHTRs for the PostSLR and nonLR pooled platelets Figure 3.

Figure 1

Febrile nonhemolytic transfusion reaction (FNHTR) rates by platelet type. The χ2 test was performed to determine the significance of differences in the rates of FNHTR among the 4 types of platelet products. The difference between the nonleukoreduced (NonLR) and the other 3 types of leukoreduced platelets was P = .008 (for single-donor [SDP]), P = .382 (for prestorage leukoreduction [PreSLR]), and P = .045 (for poststorage leukoreduction [PostSLR]).

Figure 2

Mean age of platelets by type and reaction. The average ages of platelets associated with no reaction, febrile nonhemolytic transfusion reaction (FNHTR), and allergic transfusion reactions (ATR) are shown. For pools containing platelets of different ages, the mean age of all units in the pool was used. The age difference between platelets implicated in FNHTR and platelets associated with no transfusion reactions was compared for each of the 4 types of platelets using analysis of variance. NonLR, nonleukoreduced; PostSLR, poststorage leukoreduction; PreSLR, prestorage leukoreduction; SDP, single-donor platelets.

Figure 3

Proportion of 4- to 5-day-old units in pools by platelet type and reaction. For the 3 types of pooled platelets, the proportion of 4- to 5-day-old platelets in each of the transfused pools (doses) are shown. The difference between those implicated in febrile nonhemolytic transfusion reaction (FNHTR) and those not associated with transfusion reaction was compared using analysis of variance. ATR, allergic transfusion reaction; NonLR, nonleukoreduced; PostSLR, poststorage leukoreduction; PreSLR, prestorage leukoreduction.


We analyzed more than 70,000 platelet transfusions, making this study one of the largest platelet transfusion data sets on FNHTRs to date. Four types of platelets (SDP, Pre-SLR, PostSLR, and NonLR) were concurrently available and used during the 45-month study period. The study showed that the rates of FNHTRs to all platelet components were low. The rates of FNHTRs to SDPs and PreSLR platelets were similar to those seen in 2 previously published studies6,7 and lower than one recently published study.12 A 2-fold increase was observed in the rate of FNHTRs to PreSLR pooled platelets (0.16%) compared with SDPs (0.07%), but the difference was not statistically significant (P = .067) because of the small number of reactions. This may be because the PreSLR platelets underwent leukoreduction between 36 and 48 hours after donation, which, even in that short interval, permits the accumulation of pyrogenic cytokines.9 A recent study revealed that some cytokines start accumulating in the donated whole blood within hours after donation.13

This study demonstrated a nearly doubled rate of FNHTRs between PreSLR and PostSLR pooled platelets but both rates are quite low. Previous studies have shown that Pre-SLR apheresis and whole blood–derived platelets were more effective in decreasing the frequency of FNHTRs to platelet transfusion than plasma removal, which was in turn more effective than poststorage leukoreduction.10,14 This study revealed a clear temporal correlation between the timing of leukoreduction and the rates of FNHTRs SDPs that underwent leukoreduction at the time of collection had the lowest rates of FNHTRs, PreSLR that underwent leukoreduction 36 to 48 hours after collection had more than twice the rate of FNHTRs compared with SDPs, and PostSLR that underwent leukoreduction at the time of product issue had the highest rates of FNHTRs (Table 1 and Figure 1). Furthermore, Post-SLR and NonLR pooled platelets associated with FNHTRs were on average older compared with the same type of platelet product that was not associated with FNHTRs. These findings are also consistent with previous studies that demonstrated an increased rate of FNHTRs to platelets that had been stored for at least 3 days.11,1517

A selective leukoreduction strategy was practiced at all CTS hospitals. In this study, patients were given transfusions with either leukoreduced or NonLR blood products exclusively. The patients who received NonLR platelets were general medical and surgical patients who usually do not require long-term transfusion support. Although the rates of FNHTR to NonLR pooled platelets were similar to that of PreSLR (P = .382) pooled platelets and lower than that of PostSLR (P = .045) pooled platelets, we do not believe any clinically meaningful conclusions can be drawn from these numbers because of the inherent differences in the patient populations given transfusions. It should also be emphasized that leukoreduction provides other important benefits such as minimizing HLA alloimmunization and reducing the risk of transfusion-transmitted infections (such as cytomegalovirus) to specific patient populations, in addition to reducing the incidence of FNHTRs.1

This study had several limitations. Because reaction rates were calculated from a retrospective review of passively reported reactions, underreporting of reactions could have led to lower calculated rates of reactions. To assess whether there was a bias in reporting reactions to one type of platelet component vs another, we analyzed the rates of allergic reactions. No significant differences were seen in the reported rates of allergic reactions among the 4 platelet types, which suggests that a reaction reporting bias for patients receiving different types of blood products was not present. Because this was a retrospective review, we did not have the opportunity to measure the levels of pyrogenic cytokines in the platelets involved in the transfusion reactions. CTS hospitals do not have a universal premedication policy, and the premedication status of the patients receiving transfusions was unknown. However, a recent study has shown that premedication is ineffective in preventing FNHTRs; thus the reduction in FNHTRs caused by premedication is probably minimal.18

In summary, we demonstrated increasing rates of FNHTRs to transfusions of SDPs, PreSLR pooled platelets, and PostSLR pooled platelets. PostSLR and NonLR platelet concentrates associated with FNHTR were on average older than the same type of products not associated with FNHTR. Our study supports and expands on previously published data9,10 demonstrating that the timing of leukocyte removal during platelet storage likely influences the amount of accumulated cytokines in the transfused platelet products and plays an important role in the pathogenesis of FNHTR to platelet transfusion.


We thank Ronald Radocay at The Institute for Transfusion Medicine for his support in platelet data collection and Mark Yazer, MD, for his valuable comments on the manuscript.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
View Abstract