T R A N S F U S I O N C O M P L I C AT I O N S Residual plasma in red blood cells and transfusion-related acute lung injury Litchia L. Weber,1 Laura D. Roberts,2 and Joseph D. Sweeney1,2
BACKGROUND: Transfusion-related acute lung injury (TRALI) is the most common cause of death from blood transfusion and red blood cells (RBCs) now account for approximately 50% of these fatalities. RBCs from female donors have been implicated in large series and HLA Class II antibodies to cognate recipient antigens identified in small series and case reports. The absolute volume of residual plasma in these RBCs is unknown. STUDY DESIGN AND METHODS: Two confirmed cases of RBC-associated TRALI in which the implicated donors had Class II antibodies were investigated, and the antibody strength against recipient cognate antigens was assessed using a fluorescent bead assay. RBCs in additive solution (AS) were studied on Day 42 of liquid storage to calculate residual anticoagulated plasma. RESULTS: Both RBC units were stored in AS-1 and were from female donors who had anti-HLA Class II antibodies of high strength against cognate antigens in the recipients. Anti-DR4 was identified in both cases. Nineteen AS-1 RBC units manufactured from whole blood donations using a hard spin had a mean (±1SD) residual plasma content of 38 ± 8 mL, and 26 AS-3 RBC units manufactured using a soft spin had 66 ± 13 mL (p < 0.01). CONCLUSION: RBCs continue to be manufactured from female donors of unknown or even known antiHLA status. The residual plasma content of these RBCs may approach 100 mL. A combination of a high-strength antibody and large residual plasma volume could explain severe or even fatal RBC-associated TRALI.
ransfusion-related acute lung injury (TRALI) remains the most common cause of transfusion-related mortality in the United States, accounting for 37% of all such deaths in the 5-year period 2008 through 2012.1 TRALI has most commonly been associated with high-plasma-volume components, since the plasma may contain either antiHLA Class I or anti-HLA Class II alloantibodies or alloantibodies directed against cognate antigens within the human neutrophil antigen (HNA) systems. There is no universal agreement regarding the threshold volume of plasma that constitutes a “high-plasma-volume” component, but in practice, this usually refers to plasma, apheresis platelets (PLTs) in plasma, or buffy coat pools suspended in plasma where the volume of plasma is typically in the range between 170 and 300 mL. Red blood cells (RBCs) are manufactured by removal of plasma and storage in an additive solution (AS) and are not considered a high-plasma-volume component. In the United States, three ASs are approved that differ slightly in formulation. The volume of the AS is fixed (either 100 or 110 mL) and, hence, the supernatant of the stored RBCs consist of the AS and any residual (carryover) anticoagulated plasma remaining after centrifugation(s). The volume of this plasma will depend on the centrifugational conditions of the initial (or only) centrifugation, which in turn is related to the manufacture of plasma-only components (RBCs and PLT-poor plasma or RBCs,
ABBREVIATION: TACO = transfusion-associated circulatory overload. From the 1Roger Williams Hospital and 2Rhode Island Hospital, Providence, Rhode Island. Address reprint requests to: Joseph D. Sweeney, MD, FACP, FRCPath, Coagulation and Transfusion Medicine, Rhode Island Hospital, 593 Eddy Street, Providence, RI 02903; e-mail: [email protected]
Received for publication November 13, 2013; revision received February 13, 2014, and accepted February 23, 2014. doi: 10.1111/trf.12662 © 2014 AABB TRANSFUSION 2014;54:2425-2430. Volume 54, October 2014 TRANSFUSION
WEBER ET AL.
cryosupernatant plasma, and cryoprecipitate) or a PLT component (RBCs, PLT concentrate, and PLT-poor plasma) using the PLT-rich plasma method in the United States. Most likely, more residual plasma will be present when a PLT concentrate is manufactured since the primary centrifugation is a “soft-spin,” which will result in a greater degree of plasma trapping. Estimates of the volume of residual plasma in RBCs vary between 10 and 25 mL,2 “less than 29 mL,”3 between 5 and 30 mL,4 and between 30 and 100 mL.5 One study showed mean residual plasma between 4 and 77 mL depending on the processing schema of the whole blood.6 The volume of residual plasma in RBCs does not have any clear specification or standard and is not subjected to any routine quality control. However, the residual plasma volume could be important for allergic reactions or TRALI. In the 5-year period (2008-2012) RBCs accounted for 29 fatal TRALI cases reported to the Food and Drug Administration while the high-volume plasma components (freshfrozen plasma [FFP], 2FP24, and apheresis PLTs) together accounted for 35 cases.1 Hence, RBC-associated TRALI remains a significant cause of transfusion mortality emphasizing that residual plasma may have direct relevance. The purpose of this article is to describe two cases of RBC-associated TRALI observed over a 3-year period and to measure the residual plasma in a sample of outdating RBCs, which are representatives of what would routinely be transfused in our community.
MATERIALS AND METHODS Two cases of RBC-associated TRALI from RBCs were observed over a 3-year period. Data were collected regarding the clinical aspects in the recipient, the components manufactured from the donation, the serologic findings in the donor, and the phenotype of the recipient. Class I HLA phenotyping was performed using a standard National Institutes of Health microlymphocytotoxicity (Centers for Disease Control and Prevention) assay. Class II HLA phenotyping was determined by polymerase chain reaction (sequence-specific primer) typing. HLA antibodies in donor plasma were tested using a solid-phase assay (Luminex, Gen-Probe Life Sciences, Bedford, MA). The study of residual plasma in stored RBCs was approved by the Institutional Review Board at Roger Williams Hospital (Providence, RI). RBCs were manufactured at the Rhode Island Blood Center from 550-mL whole blood donations using standard procedures and liquid stored in di(2-ethylhexyl)phthalate-polyvinyl chloride containers. All studies were performed on Day 42 (day of outdate). Each RBC unit was weighed, and the weight of the plastic container was tared. The volume of the unit was obtained by multiplying the tared weight in grams × 1.065 to derive the volume of the RBC suspension in milliliters. A sample was then taken from the container and the hema2426
TRANSFUSION Volume 54, October 2014
tocrit (Hct) measured using a cell counter (ABX Micro 60, Horiba, Irvine, CA). With the Hct and total volume, the volume of the RBCs and supernatant suspending fluid was obtained. A quantity of 110 mL was subtracted from the suspending fluid volume to obtain the volume of residual (carryover) plasma, as this is the fixed volume of AS used with a 500-mL whole blood collection. Data were entered into a software program and summarized as descriptive statistics. Statistical analysis was performed using independent t tests. A p value of less than 0.05 was considered significant.
RESULTS Two confirmed cases of RBC-associated TRALI were observed over this 3-year period during which 56,000 RBC units were transfused for an observed prevalence of 1:28,000 (95% confidence interval, 1:6500-1;197,000).
Case 1 Patient A 65-year-old female, group B, with adenocarcinoma of the lung metastatic to the brain was admitted complaining of weakness and fatigue. The hemoglobin (Hb) was 7.9 g/dL. She was transfused with 2 units of group B RBCs in AS-1. Shortly after completion of the second unit, she became acutely short of breath with erythema and hypotension. A chest X-ray showed bilateral pulmonary edema and O2 saturation was 67% on 3 L of O2. She was transferred to the intensive care unit but recovered the next day with pulse oximetry of 100% O2 saturation on room air. The posttransfusion Hb was 11.6 g/dL.
Donor HLA antibody and patient HLA phenotype The donors of the 2 RBC units returned for testing. Both were negative for HLA Class I antibodies but one of the two donors was a female with a history of pregnancy who was positive for HLA Class II antibodies. The HLA Class II antibodies detected were as follows: DR4, DR1, and DR53. For the DR1 antibody, four of four DR1 beads present were positive, with a mean positive fluorescence channel (MFI) of 5750 (range, 4801-6858). For the DR4 antibody, four of four DR4 beads were positive; mean positive MFI was 4757 (range, 2974-5852). For the DR53, 10 of 10 DR53 beads were positive; mean positive MFI was 3992 (range, 7737053). The recipient phenotyped as HLA Class I A2, 28; B14,60; HLA Class II DRB1*01,04; DRB4*present; DQB1*03,05.
Products The implicated RBC product from the female donor was derived from a whole blood donation, collected into CPD, and filtered using a RZ200 leukoreduction whole blood filter. The donor Hct was 44 predonation and the volume collected was approximately 550 mL. This gives an
RBCs AND TRALI
TABLE 1. Characterization of 45 RBC units in AS* Volumes Total volume (mL) RBC volume (mL) Hct Supernatant volume (mL) Residual plasma (mL) Minimum plasma (mL) Maximum plasma (mL)
AS-1 (n = 19) 340 ± 25 192 ± 23 56 ± 3 148 ± 8 38 ± 8 24 57
AS-3 (n = 26) 369 ± 32 193 ± 21 52 ± 2 176 ± 13 66 ± 13 29 98