Epidemiology of Diffuse Alveolar Hemorrhage in Pediatric Allogeneic Hematopoietic Cell Transplantation Recipients
Geoffrey Cheng,Michael A Smith,Rachel Phelan,Ruta Brazauskas,Joelle Strom,Kwang Woo Ahn,Betty Hamilton,Andrew Peterson,Bipin Savani,Hélène Schoemans,Michelle Schoettler,Mohamed Sorror,Christine Higham,Sandhya Kharbanda,Christopher C Dvorak,Matt S Zinter
DOI: https://doi.org/10.1016/j.jtct.2024.07.022
Abstract:Diffuse alveolar hemorrhage (DAH) is a life-threatening pulmonary toxicity that can arise after hematopoietic cell transplantation (HCT). Risk factors and outcomes are not well understood owing to a sparsity of cases spread across multiple centers. The objectives of this epidemiologic study were to characterize the incidence, outcomes, transplantation-related risk factors and comorbid critical care diagnoses associated with post-HCT DAH. Retrospective analysis was performed in a multicenter cohort of 6995 patients age ≤21 years who underwent allogeneic HCT between 2008 and 2014 identified through the Center for International Blood and Marrow Transplant Research registry and cross-matched with the Virtual Pediatric Systems database to obtain critical care characteristics. A multivariable Cox proportional hazard model was used to determine risk factors for DAH. Logistic regression models were used to determine critical care diagnoses associated with DAH. Survival outcomes were analyzed using both a landmark approach and Cox regression, with DAH as a time-varying covariate. DAH occurred in 81 patients at a median of 54 days post-HCT (interquartile range, 23 to 160 days), with a 1-year post-transplantation cumulative incidence probability of 1.0% (95% confidence interval [CI], .81% to 1.3%) and was noted in 7.6% of all pediatric intensive care unit patients. Risk factors included receipt of transplantation for nonmalignant hematologic disease (reference: malignant hematologic disease; hazard ratio [HR], 1.98; 95% CI, 1.22 to 3.22; P = .006), use of a calcineurin inhibitor (CNI) plus mycophenolate mofetil (MMF) as graft-versus-host disease (GVHD) prophylaxis (referent: CNI plus methotrexate; HR, 1.89; 95% CI, 1.07 to 3.34; P = .029), and grade III-IV acute GVHD (HR, 2.67; 95% CI, 1.53-4.66; P < .001). Critical care admitted patients with DAH had significantly higher rates of systemic hypertension, pulmonary hypertension, pericardial disease, renal failure, and bacterial/viral/fungal infections (P < .05) than those without DAH. From the time of DAH, median survival was 2.2 months, and 1-year overall survival was 26% (95% CI, 17% to 36%). Among all HCT recipients, the development of DAH when considered was associated with a 7-fold increase in unadjusted all-cause post-HCT mortality (HR, 6.96; 95% CI, 5.42 to 8.94; P < .001). In a landmark analysis of patients alive at 2 months post-HCT, patients who developed DAH had a 1-year overall survival of 33% (95% CI, 18% to 49%), compared to 82% (95% CI, 81% to 83%) for patients without DAH (P < .001). Although DAH is rare, it is associated with high mortality in the post-HCT setting. Our data suggest that clinicians should have a heightened index of suspicion of DAH in patients with pulmonary symptoms in the context of nonmalignant hematologic indication for HCT, use of CNI + MMF as GVHD prophylaxis, and severe acute GVHD. Further investigations and validation of modifiable risk factors are warranted given poor outcomes.