Transfusion Related Acute Lung Injury (TRALI): A Tertiary Care Hospital Experience.

Background: The 500 bed tertiary care hospital provides Level 1 trauma services, solid organ transplant services and an extensive cardiovascular surgery program.

Methods: Transfusion reaction records for January 2000 – June 2004 were reviewed.

Results: 39 suspected transfusion related acute lung injury reactions reported during this time at a rate of 1:1036 transfused blood products (1:4961 for RBC, 1:1031 for Apheresis Platelets, and 1:522 for FFP). There were 26 male patients, average age of 62.2 years (17–87 years) and 13 females, average age of 62.5 years (24–82 years).

All 39 patients developed acute lung injury within 6 hours of transfusion characterized by acute onset of hypoxia, without evidence of volume overload and chest xrays (performed in 34/39) showing evidence of new bilateral pulmonary infiltrates.

Patients received an average of 2.5 blood products (1–8), and 13/39 (33%) received only one product. FFP was 66% (65/98 units) of implicated product transfused, leukoreduced RBC 24% (24/98, all additive solution products), and leukoreduced platelet pheresis 9% (9/98 units). Of the patients that only received one product, 8/13 (62%) were transfused with a RBC.

Many (74%, 29/39) patients experiencing possible TRALI reactions had underlying risk factors (some patients had more than one underlying problem) including: probable sepsis (5), multiple trauma (2), cardiopulmonary bypass (6), underlying pulmonary disease (5), liver transplant (2), cardiac transplant (1), hematologic malignancy (4), TTP (3), recent surgery (5).

The reaction occurred during or immediately following transfusion in 60% (23/39) of patients, and in 77% (30/39) within 30 minutes of completion of transfusion. Of the recipients that reacted immediately 39% (9/23) of the implicated donors had HLA antibodies (Class I and Class II), and of those reacting within 30 minutes 47% (14/30) of donors had HLA antibodies. 7.7% (3/39) reactions were fatal.

Of implicated donors, 59% (58/98) are female, and 41% male (40/98). 10 donors gave a history of transfusion (1 male, 9 females). 74% (42/57 (1 female couldn’t be contacted)) had been pregnant, with the average number of pregnancies 4.0 (range 1–9), with 67% (28/42) of these donors having 3 or more pregnancies.

Donor antibody testing (Granulocyte Agglutination Assay (GA), Granulocyte Immunofluorescence Assay (GIF), and HLA Antibody Screen (HLA)) was performed only on donors that had been pregnant and/or had received a transfusion, therefore only one male donor was tested and was negative. Of the female donors 62% (36/58) were tested, 29% (17/58) did not meet testing criteria and 9% (5/58) were not available for testing. Of those tested, 31% (11/36) were negative. 69% (25/36) had HLA antibodies. 8% (2/25) with HLA antibodies also had antibodies with granulocyte specificity.

Conclusion: The etiology(ies) and prevention of TRALI reactions require additional investigation of both blood donors and recipients, although HLA antibodies appear to be involved.