Still Valid: Reassessing A Lung Transplant Recipient Risk of Death Model

Purpose Meticulous patient selection in lung transplantation remains a critically important driver of successful outcomes. We had previously created a validated risk-stratification model to aid in the pre-transplant assessment of potential recipients using only preoperative recipient data. We sought to assess the value of our model by comparing outcomes before and after the implementation of the model. Methods Data from lung transplant recipients at a single transplant institute from 2009-2018 were extracted from the UNOS database. Patients were divided into cohorts from before the implementation of the risk model (2009-2014) and after the implementation of the risk model (2015-2018). Patient and graft survival were assessed with Kaplan-Meier survival curves and compared using Cox proportional-hazards models. Results We analyzed 907 patients who underwent lung transplantation between 2009 and 2018, comparing the cohort between 2009 and 2014 ( n=633), before implementation of our risk model, to the cohort transplanted between 2015 and 2018 (n=274), after implementation of our risk model. Three-year survival was significantly different between cohorts (61.4% vs 77.7%, p<0.001) (Figure 1). Survival was also significantly different within the moderate-risk group between cohorts (n=253 vs n=136, 65.0% vs 75.8%, p=0.01). Differences in survival were most pronounced in the high-risk group between cohorts (n=166 vs n=47, 44.1% vs 79.5%, p<0.001). Stratification of recipients into low-, moderate- (HR 1.45, CI 1.01-2.08, p=0.01), and high-risk cohorts (HR 2.52, CI 1.70-3.73, p<0.001) continued to demonstrate significant differences in mortality. Conclusion Our lung transplant recipient risk-stratification model provides a valuable pre-transplant assessment tool, with significantly improved survival in our patient population after its implementation. In reassessing the model and its components, it continues to demonstrate effective discrimination of patients by risk of mortality.