Patients who develop respiratory failure requiring mechanical ventilation after hematopoietic stem cell transplantation (HSCT) have very high mortality. Several investigators have identified prognostic features that can be used to identify a subset of these patients who are virtually certain to die, yet these have never been prospectively assessed. The objectives of this study were to determine the accuracy of published prognostic features for mortality and to determine the survival of patients who recover from respiratory failure. A systematic review of the literature was undertaken to identify reported poor prognostic features and survival rates. The study validated the reported poor prognostic features on a prospective, multicenter inception cohort of 226 patients with respiratory failure requiring mechanical ventilation after HSCT. The main outcome measures were determination of a baseline probability of death, drawn from literature review; likelihood ratio of mortality for each prognostic feature determined from the validation cohort; conditional probability of death in the presence of each feature; and 6-month survival of those who recover. Patients requiring mechanical ventilation after HSCT have a baseline probability of death of 82% to 96%. In the setting of combined hepatic and renal dysfunction, the probability of death rises to 98% to 100%. Other previously reported prognostic features are less strongly associated with mortality. For patients who recover from respiratory failure, the proportion surviving 6 months or longer ranges from 27% to 88%. It was concluded that in patients requiring mechanical ventilation after HSCT, the presence of combined hepatic and renal dysfunction is highly predictive of death. The presence of this feature may justify the recommendation to withdraw life-sustaining measures.