Effect of a Lower vs Higher Positive End-Expiratory Pressure Strategy on Ventilator-Free Days in ICU Patients Without ARDS: A Randomized Clinical Trial

Abstract

Importance: It is uncertain whether invasive ventilation can use lower positive end-expiratory pressure (PEEP) in critically ill patients without acute respiratory distress syndrome (ARDS).

Objective: To determine whether a lower PEEP strategy is noninferior to a higher PEEP strategy regarding duration of mechanical ventilation at 28 days.

Design, setting, and participants: Noninferiority randomized clinical trial conducted from October 26, 2017, through December 17, 2019, in 8 intensive care units (ICUs) in the Netherlands among 980 patients without ARDS expected not to be extubated within 24 hours after start of ventilation. Final follow-up was conducted in March 2020.

Interventions: Participants were randomized to receive invasive ventilation using either lower PEEP, consisting of the lowest PEEP level between 0 and 5 cm H2O (n = 476), or higher PEEP, consisting of a PEEP level of 8 cm H2O (n = 493).

Main outcomes and measures: The primary outcome was the number of ventilator-free days at day 28, with a noninferiority margin for the difference in ventilator-free days at day 28 of -10%. Secondary outcomes included ICU and hospital lengths of stay; ICU, hospital, and 28- and 90-day mortality; development of ARDS, pneumonia, pneumothorax, severe atelectasis, severe hypoxemia, or need for rescue therapies for hypoxemia; and days with use of vasopressors or sedation.

Results: Among 980 patients who were randomized, 969 (99%) completed the trial (median age, 66 [interquartile range {IQR}, 56-74] years; 246 [36%] women). At day 28, 476 patients in the lower PEEP group had a median of 18 ventilator-free days (IQR, 0-27 days) and 493 patients in the higher PEEP group had a median of 17 ventilator-free days (IQR, 0-27 days) (mean ratio, 1.04; 95% CI, 0.95-∞; P = .007 for noninferiority), and the lower boundary of the 95% CI was within the noninferiority margin. Occurrence of severe hypoxemia was 20.6% vs 17.6% (risk ratio, 1.17; 95% CI, 0.90-1.51; P = .99) and need for rescue strategy was 19.7% vs 14.6% (risk ratio, 1.35; 95% CI, 1.02-1.79; adjusted P = .54) in patients in the lower and higher PEEP groups, respectively. Mortality at 28 days was 38.4% vs 42.0% (hazard ratio, 0.89; 95% CI, 0.73-1.09; P = .99) in patients in the lower and higher PEEP groups, respectively. There were no statistically significant differences in other secondary outcomes.

Conclusions and relevance: Among patients in the ICU without ARDS who were expected not to be extubated within 24 hours, a lower PEEP strategy was noninferior to a higher PEEP strategy with regard to the number of ventilator-free days at day 28. These findings support the use of lower PEEP in patients without ARDS.

Trial registration: ClinicalTrials.gov Identifier: NCT03167580.

Figure 1.. Flow of Participants in the Restricted vs Liberal Positive End-Expiratory Pressure in Patients Without ARDS Trial

In all analyses, patients were analyzed according to their randomization group, with the exception of those who withdrew informed consent or were lost to follow-up in the first 28 days. PEEP indicates positive end-expiratory pressure. ^aPatients could have more than 1 reason for exclusion; the main reason for exclusion is presented. ^bIncludes chronic obstructive pulmonary disease (COPD) stages III and IV in the Global Initiative for COPD (GOLD) classification and restrictive pulmonary disease. GOLD stage III COPD is defined as severe obstruction of the airways, with the ratio of forced expiratory volume in the first second to forced vital capacity (FEV₁/FVC) less than 70% and FEV₁ between 30% and 50% of predicted values. GOLD stage IV COPD is defined as very severe obstruction of the airways, with FEV₁/FVC less than 70% and FEV₁ less than 30% of predicted values. ^cIncludes ongoing cardiac ischemia due to cardiac infarction and failed revascularization, uncontrollable intracranial pressure, delayed cerebral ischemia after subarachnoid hemorrhage, necrotizing fasciitis, and severe untreated anemia. ^dIncludes Guillain-Barré syndrome, high spinal cord lesion, amyotrophic lateral sclerosis, multiple sclerosis, or myasthenia gravis.

Figure 2.. PEEP and Fio₂ Measures From Baseline Through Day 5

Fio₂ indicates fraction of inspired oxygen; PEEP, positive end-expiratory pressure. The figure shows the mean, lowest, and highest PEEP and Fio₂ values recorded from measurements taken every 6 hours while patients were undergoing invasive mechanical ventilation. Horizontal bars inside boxes represent medians; box tops and bottoms, interquartile ranges. Whiskers extend to 1.5 times the interquartile range beyond the first and third quartiles per the conventional Tukey method. Circles represent means. Dashed lines represent the limits imposed by the protocol in each group. Data labeled “after randomization” were recorded 1 hour after randomization (on day 0). The number of participants with available data decreases over successive study days because of deaths and discontinuation of invasive mechanical ventilation.

Figure 3.. Noninferiority Analysis of the Primary Outcome in the Overall Cohort and in Subgroup Analyses

APACHE indicates Acute Physiology and Chronic Health Evaluation (range, 0-286; higher scores indicate more severe disease and higher risk of death); Fio₂, fraction of inspired oxygen; IQR, interquartile range; LIPS, Lung Injury Protection Score (range, 0-32.5; higher scores indicate more severe condition and higher risk of acute respiratory distress syndrome); PEEP, positive end-expiratory pressure. Body mass index is calculated as weight in kilograms divided by height in meters squared. The primary outcome is the mean ratio (with 1-sided 95% CI) for the comparison of ventilator-free days at day 28 between the lower and higher PEEP groups, with the noninferiority margin set at 0.90. Absolute differences are mean differences. P values shown are from the β-binomial part of the model. P values for interaction for the binary logistic regression for 0 ventilator-free days at 28 days are P = .62 for type of admission, P = .14 for cardiac arrest, P = .22 for reason for intubation, P = .74 for body mass index, P = .66 for Pao₂/Fio₂, P = .65 for LIPS, and P = .74 for APACHE IV score.

Figure 4.. Kaplan-Meier Estimates for Patients in the Lower PEEP and Higher PEEP Groups

ICU, intensive care unit; IQR, interquartile range; PEEP, positive end-expiratory pressure. For each panel, an unadjusted hazard ratio and 95% CI calculated from a Cox proportional hazard model is presented. A, The median observation period for time to freedom from invasive ventilation was 2.0 (IQR, 1.7-2.6) days for the lower PEEP group and 2.0 (IQR, 1.6-2.4) days for the higher PEEP group; P = .14 for Schoenfeld residuals. B, The median observation period for time to discharge alive from the ICU was 7.0 (IQR, 6.0-9.0) days for the lower PEEP group and 7.0 (IQR, 6.0-8.0) days for the higher PEEP group; P = .85 for Schoenfeld residuals. C, The median observation time for 90-day mortality was not computed because the minimum observed value was 0.44 days; P = .89 for Schoenfeld residuals. D, The median observation period for time to discharge alive from the hospital was 22.0 (IQR, 20.0-27.0) days for the lower PEEP group and 22.0 (IQR, 20.0-25.0) for the higher PEEP group; P = .91 for Schoenfeld residuals.