By Samuel Nadler, MD, PhD

Critical Care, Pulmonary Medicine, The Polyclinic Madison Center, Seattle; Clinical Instructor, University of Washington, Seattle

SYNOPSIS: Optimizing positive end-expiratory pressure to minimize driving pressure may be a better strategy for ventilator adjustment than maximizing partial pressure of arterial oxygen/fraction of inspired oxygen (P/F) ratios to improve outcomes in patients with acute respiratory distress syndrome.

SOURCE: Yehya N, Hodgson CL, Amato MBP, et al. Response to ventilator adjustments for predicting acute respiratory distress syndrome mortality. Driving pressure versus oxygenation. Ann Am Thorac Soc 2021;18:857-864.

Lung protective ventilation has become the standard of care for patients with the acute respiratory distress syndrome (ARDS). Beyond this, the ventilator adjustments most closely associated with mortality remain unclear. Yehya et al re-evaluated the ALVEOLI and ExPress trials, two previous studies of higher vs. lower positive end-expiratory pressure (PEEP) in patients with ARDS.1,2 The relationships between changes in PaO2/FiO2 (ΔP/F) or changes in driving pressure (ΔΔP) on mortality were examined. Multiple sensitivity analyses further stratified patients by the presence of passive ventilation, level of pre-randomization PEEP, changes in tidal volumes (ΔVt) and changes in driving pressure attributable only to changes in Vt.

Within the ALVEOLI trial, positive ΔΔP (hazard ratio [HR], 1.50; 95% confidence interval [CI], 1.21-1.85; P < 0.001) but not negative ΔP/F (HR, 0.95; 95% CI, 0.86-1.04; P = 0.232) was associated with higher mortality among the 372 patients with complete data prior to randomization. This relationship remained when the analysis was restricted to patients with passive ventilation, baseline P/F 200, those in the high PEEP arm, and when adjusted for pre-randomization PEEP or changes in Vt. Neither ΔΔP nor ΔP/F correlated with mortality in the low PEEP arm.

A similar analysis of 596 patients from the ExPress trial showed a stronger association between mortality and positive ΔΔP (HR, 1.42; 95% CI, 1.14-1.79; P = 0.002) than with Δ P/F (HR, 0.95; 95% CI, 0.9-1.00; P = 0.040). This held true in patients with passive ventilation, those in the higher PEEP arm, and when adjusted for pre-randomization PEEP, but not in the low PEEP arm. When both data sets were combined into one complete model, ΔΔP was more strongly associated with mortality than ΔP/F (HR 1.48 vs. 0.95, respectively).


This study represents a complicated re-analysis of two previously published data sets that were developed to determine if higher or lower PEEP strategies affected mortality in patients with ARDS. Neither of these trials showed a significant benefit from higher PEEP strategies. Thus, the conclusions drawn from this re-analysis must be viewed in the context of the failure of these previous studies to show a benefit from protocols to adjust PEEP to optimize outcomes. However, within the data, there may be correlations that can be used to drive future studies. Although correlation does not mean causation, the improvements in mortality with lower driving pressure (ΔP) after changes in PEEP imply a more physiologic protocol of PEEP titration that minimizes ΔP, rather than adjusting to a fixed PEEP level, might improve outcomes.

Does the observed correlation make physiologic sense? Injury to the lung in ARDS can be quite heterogeneous. The effect of PEEP throughout the lung will vary as well. Increases in PEEP that cause decreases in ΔP imply improved overall compliance of the respiratory system or a more “open lung,” while increased PEEP associated with increased ΔP suggests worsening compliance or over-distention. If maximizing lung compliance as the parameter most closely associated with mortality is the goal, a driving pressure-based approach may be appropriate. Interestingly, neither ΔΔP nor ΔP/F was correlated with mortality in the low PEEP arms. This suggests these lower levels of PEEP do not commonly lead to over-distention but may not fully recruit lung tissue.

Other studies support the notion that decreased driving pressures are associated with improved outcomes. Amato et al similarly found that ΔP was more closely associated with mortality than PEEP or plateau pressure.3 More recently, Barrot et al found no difference in 28-day mortality in patients with ARDS randomized to a conservative vs. liberal oxygenation strategy.4 Thus, higher PaO2 or P/F ratios may not be good surrogates for ventilator titration, and changes in ΔP might be better.

More definitive studies of driving pressure-driven protocols for ventilator adjustment likely will be published. Until that time, paying attention to the ΔΔP as you adjust PEEP in patients with ARDS on the ventilator could improve their outcomes. 


  1. Brower RG, Lanken PN, MacIntyre N, et al; National Heart, Lung, and Blood Institute ARDS Clinical Trials Network. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med 2004:351:327-336.
  2. Mercat A, Richard JCM, Vielle B, et al. Positive end-expiratory pressure settings in adults with acute lung injury and acute respiratory distress syndrome: A randomized controlled trial. JAMA 2008;299:646-655.
  3. Amato MBP, Meade MO, Slutsky AS, et al. Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med 2015;372:747-755.
  4. Barrot L, Asfar P, Mauny F, et al. Liberal or conservative oxygen therapy for acute respiratory distress syndrome. N Engl J Med 2020;382:999-1008.