RISK FACTORS, CLINICAL CHARACTERISTICS, AND OUTCOMES OF TRANSFUSION-RELATED ACUTE LUNG INJURY: A SYSTEMATIC REVIEW OF RANDOMIZED CONTROLLED TRIALS AND PRIMARY STUDIES

Abstract

Introduction: Transfusion-related acute lung injury (TRALI) remains a leading cause of transfusion-associated morbidity and mortality. Evidence on its risk factors, clinical phenotype and prognosis is fragmented across heterogeneous definitions, clinical settings and surveillance systems, and has not been synthesised since the 2019 consensus redefinition (TRALI type I/type II). This review aimed to quantify patient-, donor- and transfusion-related risk factors, describe clinical, laboratory and radiological characteristics and underlying mechanisms, compare TRALI with transfusion-associated circulatory overload (TACO), and estimate clinical outcomes.

Methods: The study strictly adhered to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) 2020 guidelines. Randomised controlled trials (RCTs), prospective and retrospective cohort, nested case–control, cross-sectional and haemovigilance studies were eligible. Risk of bias was appraised with the Newcastle–Ottawa Scale (NOS) and Cochrane RoB 2. Odds ratios (OR), risk ratios (RR), mean differences (MD) and logit-transformed proportions were pooled using DerSimonian–Laird random-effects models with 95% prediction intervals, leave-one-out, cumulative and small-study-effect analyses; certainty was graded with GRADE.

Results: Thirty-nine studies (37 observational, 2 RCTs) from 14 countries were included, covering more than 16.8 million transfused patients or hospitalisations and more than 144 million blood components; 30 studies contributed to quantitative synthesis. Pooled TRALI incidence was 4.47% (95% CI 2.29–8.56) in high-risk/critical-care cohorts, 0.41% (0.11–1.47) in general peri-operative cohorts and 0.017% (0.007–0.045) in coded administrative databases, and 1.88 per 100,000 components (0.46–7.66). Exposure to high-risk donor plasma or leukocyte antibodies (OR 4.91, 2.43–9.94; I² = 0%), chronic lung disease (OR 2.72, 1.46–5.08), shock or sepsis (OR 2.54, 1.05–6.09), platelet exposure (OR 1.74, 1.08–2.78), female recipient sex (OR 1.24, 1.14–1.35) and each additional component transfused (OR 1.20, 1.12–1.29) were associated with TRALI. An antibody-mediated mechanism was confirmed in 68.7% (56.1–79.1) of investigated cases, and 69.2% of recent cases were TRALI type II. TRALI presented with fever (OR 4.84), hypotension and new mechanical ventilation (OR 32.2); 53.6% (28.2–77.2) of patients required invasive ventilation. In-hospital mortality was higher than in transfused controls (OR 5.22, 2.68–10.17; I² = 84%), with a pooled case-fatality of 27.3% (19.5–36.8), more ICU admissions (OR 3.16, 1.91–5.24) and longer hospital (MD +6.9 days) and ICU stay (MD +5.2 days). Compared with TACO, TRALI carried higher mortality (OR 2.48, 2.20–2.80), longer stay (MD +4.8 days), lower natriuretic peptide levels and fewer cardiovascular comorbidities. Mitigation strategies showed a non-significant reduction in incidence (RR 0.57, 0.28–1.18).

Discussion: The findings support a threshold two-hit model in which recipient inflammatory priming (shock, sepsis, chronic lung and liver disease, alcohol and tobacco exposure) interacts with donor leukocyte antibodies and non-antibody biological response modifiers in plasma-rich components. Recipient factors predominate in TRALI type II and possible TRALI, whereas donor antibodies drive type I disease. TRALI is clinically and prognostically distinct from TACO.

Conclusion: TRALI is not rare in high-risk patients and carries roughly a five-fold excess in-hospital mortality. Prevention should combine donor-based mitigation (male-predominant or antibody-screened plasma and platelets), restrictive patient blood management in primed recipients, early recognition using natriuretic peptides and imaging, and systematic haemovigilance reporting.