Predictors of Mortality Post Lung Transplantation: Systematic Review and Meta-Analysis

Purpose In the field of lung transplantation , prognostic factors help in identifying characteristics of recipients and donors that may influence the survival of the transplanted organ or the patient's life. Understanding such factors can guide clinical decision making and better matching of recipients and donors. Our systematic review and meta-analysis aimed to provide best estimates of the association between possible predictors and 1-year all-cause mortality in adult primary lung transplant recipients. Methods We systematically searched 5 bibliographic databases for eligible primary studies assessing the association between any potential risk factor (related to lung donor, recipient, or the transplant operation), and 1-year recipient mortality. We pooled effect estimates using a random effects or fixed effects framework, as appropriate, according to a pre-specified protocol. This review utilized the GRADE approach to assess the quality of the evidence. Results High quality evidence identified older recipient age (HR 1.19 per 10-year increase, 95% CI 1.12 to 1.26), higher recipient BMI (HR 1.05 per 5-kg/m2 increase, 95% CI 1.01 to 1.10), female donor sex (HR 1.09, 95% CI 1.01 to 1.18), single vs bilateral lung transplant (HR 1.25, 95% CI 1.19 to 1.32) and post-transplant use of dialysis (HR 7.96, 95% CI 6.87 to 9.22) as predictors of 1-year mortality (Table 1). High quality evidence also excluded associations of pre-transplant lung disease etiology, recipient diabetes , pre-transplant systolic pulmonary artery pressure (PAP), pre-transplant mean PAP, donor age, blood type compatibility, and height mismatch between recipient and donors with 1-year mortality. Conclusion Recipient age, recipient BMI, post-transplant dialysis, donor sex, and transplant type predict 1-year graft survival . With the exception of post-transplant need for dialysis, the impact of each factor is, however, modest, suggesting the potential value for a predictive model that would incorporate multiple factors.