Michaela R Anderson, Robert Gallop, A Russell Localio, Michelle Oyster, Laurel Kalman, Stephen Griffiths, Matthew Hartwig, Laurie D Snyder, Michael Shashaty, Vivek Ahya, Maria Crespo, Mauer Biscotti, Gabrielle Mezochow, Krishna Pandya, Julia Gasior, Daniel Calabrese, John R Greenland, Luke Benvenuto, Gundeep Dhillon, Michael P Combs, John McDyer, Chadi Hage, David Roe, Pali Shah, Christian Merlo, Jonathan P Singer, Scott M Palmer, Jason D Christie
A revised PGD definition stratified by mode of oxygen delivery identifies groups with differential mortality. Chest radiography is critical to excluding alternative causes of lung injury in recipients with severe hypoxemia.
BACKGROUND: Primary graft dysfunction (PGD) is severe lung injury following lung transplantation. Variable radiograph interpretation and changes in mechanical ventilation and extracorporeal membrane oxygenation (ECMO) support for hypoxemia may limit the current PGD definition's validity.
METHODS: We assessed whether changes in the PGD definition accounting for oxygen delivery method (intubated, non-intubated, ECMO) were associated with differential graft survival. Severe PGD was grade 3 at 48 or 72 hours after transplant. We estimated crude and standardized time from transplantation to death or re-transplantation with each alternate defining criterion. We secondarily evaluated whether excluding chest radiography led to severe PGD re-classification.
RESULTS: Among 3170 recipients, 749 (24%) had severe PGD by current definition. By revised oxygen delivery definition, 182 non-intubated recipients and 372 intubated recipients had severe PGD, and 177 recipients had severe PGD requiring ECMO. Recipients with severe PGD requiring ECMO had highest 1-year mortality (cumulative incidence of death 0.20, 95%CI 0.17-0.24), followed by intubated severe PGD (0.15, 95%CI 0.13-0.17). Non-intubated recipients had similar mortality with and without severe PGD (non-intubated with PGD: 1-year cumulative incidence of death 0.11, 95%CI 0.09-0.13; non-intubated without severe PGD: 0.10, 95%CI 0.09-0.11). Exclusion of radiographs reclassified 814 recipients, including 89 (11%) re-classified as severe PGD. On review of radiographs from recipients re-classified as severe PGD, most frequent findings were effusions, atelectasis, and pneumonia.
CONCLUSION: A revised PGD definition stratified by mode of oxygen delivery identifies groups with differential mortality. Chest radiography is critical to excluding alternative causes of lung injury in recipients with severe hypoxemia.