Prevalence and Prognosis of Updated Phenotypes of Chronic Lung Allograft Dysfunction

PURPOSE The recently updated ISHLT chronic lung allograft dysfunction (CLAD) consensus report establishes four CLAD phenotypes. The proportion of patients who meet criteria for these phenotypes and their respective survival have not been described. METHODS We retrospectively determined the CLAD phenotype at the time of CLAD onset for all patients who received a bilateral lung transplant at the University of Michigan based on the presence of obstruction (FEV1/FVC <70%), restriction (TLC <90% of post-transplant baseline, when TLC data was not available FVC <80% of post-transplant baseline), and radiographic opacities (CT scan with parenchymal opacities or pleural thickening). Survival analyses were performed using Cox proportional hazards and presented as Kaplan Meier curves. RESULTS Among 327 eligible patients, 142 (43.4%) developed a definitive FEV1 decline not attributable to another cause, consistent with CLAD. At CLAD onset, 46 (32.4%) patients had BOS, 22 (15.5%) patients had undefined CLAD, 12 (8.5%) had RAS, 8 (5.6%) patients had mixed CLAD; 54 (38.0%) patients could not be classified as any CLAD phenotype as defined by ISHLT guidelines (Figure, panel A). Compared to patients with BOS, patients with undefined CLAD, RAS, and mixed CLAD all had shorter survival after CLAD onset (p<0.001, p=0.007 & p<0.001, respectively), but these were not significantly different from each other (Figure, panel B). CONCLUSION Patients with undefined CLAD, RAS, and mixed CLAD, as defined by the updated ISHLT guidelines, have worse survival than patients with BOS. A substantial portion of patients with CLAD could not be categorized into a defined phenotype.