42

Abstract

estimated for these hospitalizations. The primary outcome variables of interest included in-hospital mortality (IHM), length of stay (LOS), and hospitalization charges (HC [inflation-adjusted to year 2010). Multivariable linear regression analyses were used to examine the association between MV and LOS and hospital charges. Multivariable logistic regression was used for in-hospital mortality. Since LOS and HC were skewed, they were log transformed and used as dependent variables in the regression models. The effects of age, gender, race, type of admission, co-morbid burden, and hospital region were adjusted in the regression models. Results: During the study period a total of 1,737 hospitalizations had acute respiratory failure following bone marrow transplantation. The mean age of hospitalizations was 44.5 years. Males comprised 55% of all hospitalizations. MV was needed for less than 96 consecutive hours in 21.2% of patients and 48.9% had MV for 96 consecutive hours or more. A total of 996 patients died in hospitals (overall IHM rate of 57.4% for all hospitalizations). IHM rates were 35.1%, 62.1%, and 69.1% among those who did not require MV, who had MV for less than 96 consecutive hours, and had MV for 96 consecutive hours or more, respectively. The mean LOS and HC for those who did not have any form of MV were 43.5 days and $518,990 respectively. The mean LOS and HC for those who had MV for less than 96 consecutive hours were 38.6 days and $483,410, respectively. The mean LOS and HC for those who had MV for 96 consecutive hours or more were 53.8 days and $743,197, respectively. Following adjustment for all confounding factors, those who had MV for less than 96 hours (OR=3.07, p<0.01) and those who had MV for 96 hours or greater (OR=4.10, p<0.01) were associated with significantly higher odds for IHM. Following adjustment for all confounding factors, those who had MV for 96 hours or greater were associated with significantly higher hospitalization charges and longer hospital LOS (p<0.0001). Conclusions: Mechanical ventilation is an independent predictor of mortality in hospitalized bone marrow transplant patients with respiratory failure. Mortality was highest in those needing the mechanical ventilation greater than 96 hours. In hospital resource utilization was significant.