Practice of excessive F(IO(2)) and effect on pulmonary outcomes in mechanically ventilated patients with acute lung injury.

BACKGROUND: Optimal titration of inspired oxygen is important to prevent hyperoxia in mechanically ventilated patients in ICUs. There is mounting evidence of the deleterious effects of hyperoxia; however, there is a paucity of data about F(IO(2)) practice and oxygen exposure among patients in ICUs. We therefore sought to assess excessive F(IO(2)) exposure in mechanically ventilated patients with acute lung injury and to evaluate the effect on pulmonary outcomes.
METHODS: From a database of ICU patients with acute lung injury identified by prospective electronic medical record screening, we identified those who underwent invasive mechanical ventilation for > 48 hours from January 1 to December 31, 2008. Ventilator settings, including F(IO(2)) and corresponding S(pO(2)), were collected from the electronic medical record at 15-min intervals for the first 48 hours. Excessive F(IO(2)) was defined as F(IO(2)) > 0.5 despite S(pO(2)) > 92%. The association between the duration of excessive exposure and pulmonary outcomes was assessed by change in oxygenation index from baseline to 48 hours and was analyzed by univariate and multivariate linear regression analysis.
RESULTS: Of 210 patients who met the inclusion criteria, 155 (74%) were exposed to excessive F(IO(2)) for a median duration of 17 hours (interquartile range 7.5-33 h). Prolonged exposure to excessive F(IO(2)) correlated with worse oxygenation index at 48 hours in a dose-response manner (P < .001.). Both exposure to higher F(IO(2)) and longer duration of exposure were associated with worsening oxygenation index at 48 hours (P < .001), more days on mechanical ventilation, longer ICU stay, and longer hospital stay (P = .004). No mortality difference was noted.
CONCLUSIONS: Excessive oxygen supplementation is common in mechanically ventilated patients with ALI and may be associated with worsening lung function.