Oxygenation response to positive end-expiratory pressure predicts mortality in acute respiratory distress syndrome. A secondary analysis of the LOVS and ExPress trials

Am J Respir Crit Care Med. 2014 Jul 1;190(1):70-6. doi: 10.1164/rccm.201404-0688OC.


Rationale: Previous trials of higher positive end-expiratory pressure (PEEP) for acute respiratory distress syndrome (ARDS) failed to demonstrate mortality benefit, possibly because of differences in lung recruitability among patients with ARDS.

Objectives: To determine whether the physiological response to increased PEEP is associated with mortality.

Methods: In a secondary analysis of the Lung Open Ventilation Study (LOVS, n = 983), we examined the relationship between the initial response to changes in PEEP after randomization and mortality. We sought to corroborate our findings using data from a different trial of higher PEEP (ExPress, n = 749).

Measurements and main results: The oxygenation response (change in ratio of arterial partial pressure of oxygen to fraction of inspired oxygen: P/F) after the initial change in PEEP after randomization varied widely (median, 9.5 mm Hg; interquartile range, -16 to 47) and was only weakly related to baseline P/F or the magnitude of PEEP change. Among patients in whom PEEP was increased after randomization, an increase in P/F was associated with reduced mortality (multivariable logistic regression; adjusted odds ratio, 0.80 [95% confidence interval, 0.72-0.89] per 25-mm Hg increase in P/F), particularly in patients with severe disease (baseline P/F [less-than-or-equal-to] 150 mm Hg). Changes in compliance and dead space were not associated with mortality. These findings were confirmed by a similar analysis of data from the ExPress trial.

Conclusions: Patients with ARDS who respond to increased PEEP by improved oxygenation have a lower risk of death. The oxygenation response to PEEP might be used to predict whether patients will benefit from higher versus lower PEEP.

Keywords: positive end-expiratory pressure; respiratory distress syndrome, adult; ventilator-induced lung injury.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Female
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Oxygen Consumption / physiology*
  • Positive-Pressure Respiration / methods
  • Positive-Pressure Respiration / mortality
  • Positive-Pressure Respiration / statistics & numerical data*
  • Prognosis
  • Randomized Controlled Trials as Topic
  • Respiratory Distress Syndrome / mortality
  • Respiratory Distress Syndrome / therapy*
  • Severity of Illness Index
  • Tidal Volume / physiology
  • Treatment Outcome
  • Ventilator-Induced Lung Injury / etiology
  • Ventilator-Induced Lung Injury / mortality
  • Ventilator-Induced Lung Injury / prevention & control*