Background

Transfusion-related acute lung injury (TRALI) has been the leading cause of transfusion-related morbidity and mortality in the UK and the USA in recent years. A threshold mechanism of TRALI has been proposed in which both patient factors (type and/or severity of clinical insult) and blood product factors (strength and/or concentration of antibodies or biological response modifiers) interact to surpass a threshold for TRALI development (Bux et al. Br J Haematol; 2007; 136: 788-99). The risk of developing antibody-mediated TRALI has been minimised by the introduction of risk-reduction strategies such as limiting the use of plasma from female donors. In contrast, there are no strategies currently in place to mitigate the development of non-antibody mediated TRALI as the mechanisms remain largely undefined. Previous studies have implicated non-polar lipids such as arachidonic acid and various species of hydroxyeicosatetranoic acid (HETE) in the development of non-antibody mediated TRALI (Silliman et al. Transfusion; 2011; 51: 2549-54), however the contribution of these lipids to the development of an inflammatory response in TRALI is poorly understood.

Methodology

Standard leucodepleted packed red blood cell (PRBC) units were sampled at either day (D) 2 (n=75) or at D42 (n=113). PRBC-supernatants were obtained via centrifugation, pooled (D2, D42) and levels of arachidonic acid, 5-, 12- and 15-HETE determined using commercial ELISA kits. In an in vitro transfusion model, fresh human whole blood (“recipient”; n=8 for each lipid) was mixed with combinations of culture media (control) or lipopolysaccharide (LPS, 0.23 µg/mL) as a first hit. A range of concentrations of either 5-HETE (200; 1,000; 2,500; 10,000; 250,000 pg/mL), 12-HETE (1,500; 5,000; 62,500; 250,000 pg/mL) or 15-HETE (150; 1,000; 2,000; 8,000 pg/mL) were added as the second hit, and incubated for 6 hours with the addition of a protein transport inhibitor. Neutrophil- and monocyte-specific inflammatory response was assessed using multi-colour flow cytometry (panel: IL-6, IL-8, IL-10, IL-12, IL-1α, IL-1β, TNF-α, MCP-1, IP-10, MIP-1α, MIP-1β). Significance was determined as P < 0.05 by one-way ANOVA with Bonferonni's correction used to determine dose response (indicated by asterisks).

Results

5-, 12- and 15-HETE were all detectable in both of the PRBC supernatant pools, with levels increased in D42 compared to D2 (5-HETE: 20,347 vs. 3,449; 12-HETE: 240,967 vs. 1,572; 15-HETE: 7,900 vs. 934; all levels in pg/mL). Arachidonic acid was not detectable in either of the PRBC supernatant pools. In the absence of LPS as a first hit, the addition of non-polar lipids had a predominantly immunosuppressive effect in the transfusion model. 12-HETE suppressed monocyte production of MIP-1α* and neutrophil production of IL-6, IL-8 and IL-12. Also, 15-HETE modulated monocyte IL-8 production and reduced neutrophil production of IL-8, IL-12, IP-10, MIP-1α, MIP-1β and TNF-α. In contrast, in the presence of LPS as a first hit, a predominantly pro-inflammatory response was evident to these lipids. 12-HETE increased monocyte production of IL-1α, IL-8* and MIP-1β* as well as neutrophil production of IL-1α*, IP-10*, MCP-1, MIP-1α* and MIP-1β. In addition, 15-HETE increased neutrophil expression of IL-1α and IL-6, and 5-HETE modulated monocyte production of MIP-1β.

Conclusions

These data suggest that the non-polar lipid mediators investigated here, in particular 12-HETE, may contribute to TRALI pathogenesis. A storage related accumulation of 5-, 12- and 15-HETE was evident in leucodepleted PRBC units. The in vitro model indicated that exposure to these lipid mediators supressed the recipient inflammatory responses in the absence of LPS, but contributed to a pro-inflammatory profile in the presence of LPS as a first hit. Together these data provide further evidence of the importance of both patient (first hit) and blood component (second hit) factors in the development of TRALI. Furthermore, the dose-associated response observed for a number of inflammatory markers is consistent with the threshold hypothesis of TRALI pathogenesis.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution