SESSION TITLE: ARDS/Lung Injury
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Monday, October 27, 2014 at 04:30 PM - 05:30 PM
PURPOSE: Multiple advances in the treatment of ARDS were been made in the past two decades. However, meta-analyses conducted to study trends in mortality from ARDS during this time period had conflicting results. Our goal was to examine ARDS mortality trends from 1996-2011 using the largest patient sample to date.
METHODS: We used the largest all-payer inpatient health care database in the United States, the National Inpatient Sample (NIS) database. Un-weighted, it encompasses data from approximately 8 million hospital stays per year. Weighted (when expanded to estimate country-wide discharges), it estimates data corresponding to approximately 40 million hospitalizations. We examined all patients in the database from 1996-2011 with ICD 9 codes of 518.5 and 518.82, including all possible patients with ARDS. Patients were then sub-classified by duration of ventilator support using procedural codes 96.70, 96.71 and 96.72. All patients with codes 518.5 or 518.82 plus 96.72 (possible ARDS with ventilation of 96 hours or more) were defined as truly having ARDS. We also defined patients with codes 518.5 or 518.82 plus 96.71 (possible ARDS with ventilation of less than 96 hours) as having true ARDS if they died within the first 96 hours of admission.
RESULTS: 174,180 patients fit our definition of ARDS. When weighted, this number can be expanded to 856,293 patients. Over this 16-year period, there was an absolute mortality reduction of 14.6 % (46.8% - 32.2%) and a relative reduction of 31%. Most striking, there was an 8.9 % absolute reduction from 2000-2005. Plotted as a joinpoint regression trend, the downward slope in mortality increased from -0.57 to -1.79 at the year 2000 (p= 0.02). After 2005, the rate of mortality decrease returned to the pre-2000 trend (slope -0.65).
CONCLUSIONS: In the largest study to examine mortality trends in ARDS from 1996-2011, we demonstrate a clear decline in ARDS mortality. Most notably, there was a sharp decrease in mortality in the period of 2000-2005. While we cannot prove causation for the decreased mortality, we believe that collaborative advances in critical care medicine contributed to the overall decline. However, we assert that the sudden and sharp decrease in mortality from 2000-2005 can be attributed to the practice of low tidal volume ventilation.
CLINICAL IMPLICATIONS: Our findings suggest that advances in the treatment of ARDS and critical care medicine have resulted in decreased mortality from ARDS. This study helps validate current practices in ARDS management.
DISCLOSURE: The following authors have nothing to disclose: Jared Radbel, Kathan Mehta, Neeraj Shah, Ronak Soni, Jasvinder Singh
No Product/Research Disclosure Information