Lung Injury Prediction Model in Bone Marrow Transplantation: A Multicenter Cohort Study.

RATIONALE: Pulmonary complications contribute significantly to non-relapse mortality following hematopoietic stem cell transplantation (HCT). Identifying high-risk patients can help enroll such patients into clinical studies to better understand, prevent and treat post-transplant respiratory failure syndromes.

OBJECTIVE: Develop and validate a prediction model to identify those at increased risk of acute respiratory failure after HCT.

METHODS: Patients underwent HCT between January 1, 2019, and December 31, 2021. Those in Rochester, Minnesota formed the derivation cohort and those from Scottsdale, Arizona or Jacksonville, Florida formed the validation cohort. The primary outcome was development of acute respiratory distress syndrome (ARDS), with secondary outcomes including need for invasive/noninvasive ventilation. Predictors were based on prior case-control studies.

MEASUREMENTS AND MAIN RESULTS: Of 2450 patients undergoing stem cell transplantation, there were 1718 hospitalizations (888 patients) in the training cohort and 1005 hospitalizations (470 patients) in the test cohort. A 22-point model was developed, with 11 points from pre-hospital predictors and 11 points from post-transplant or early (<24h) in-hospital predictors. The model performed well for predicting ARDS (C-statistic = 0.905, 95%CI: 0.870-0.941) and need for invasive/noninvasive ventilation (C-statistic = 0.863, 95%CI: 0.828-0.898). The test cohort differed markedly in demographic, medical and hematologic characteristics. The model performed well in this setting as well for predicting ARDS (C-statistic = 0.841 (95%CI: 0.782-0.900) and need for invasive/noninvasive ventilation (C-statistic = 0.872, 95%CI: 0.831-0.914).

CONCLUSION: A novel prediction model incorporating data elements from the pre-transplant, post-transplant and early in-hospital domains can reliably predict development of post-HCT acute respiratory failure.