By Elaine Chen, MD

Assistant Professor, Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, Section
of Palliative Medicine, Rush University Medical Center, Chicago

Dr. Chen reports no financial relationships relevant to this field of study.

SYNOPSIS: In the LOCO2 study, a conservative oxygen strategy with SpO2 goals of 88% to 92% was not shown to improve mortality over a liberal oxygen strategy as hypothesized, but rather was found to have a worrisome signal of increased mortality and increased mesenteric ischemia.

SOURCE: 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.

In the management of acute respiratory distress syndrome (ARDS), clinicians traditionally target a partial pressure of arterial oxygen (PaO2) between 55 mmHg and 80 mmHg, or pulse oximetry (SpO2) between 88% and 95%.1 However, these oxygenation goals are not based on robust experimental data. Based on a feasibility study in the Netherlands, Barrot et al hypothesized that a conservative oxygen therapy would reduce mortality compared with a liberal oxygen strategy.2

The LOCO2 (Liberal Oxygenation versus Conservative Oxygenation in Acute Respiratory Distress Syndrome) trial was a prospective, multicenter, randomized, open-label trial involving patients with ARDS. It was conducted in 13 intensive care units (ICUs) in France between June 2016 and September 2018. Eligible patients were intubated and on mechanical ventilation who met criteria for the Berlin definition of ARDS. Patients were randomized to the liberal-oxygen group (PaO2 target of 90 mmHg to 105 mmHg, or SpO2 greater than 96%) or the conservative-oxygen group (PaO2 target between 55 mmHg to 70 mmHg, or SpO2 88% to 92%) for the first seven days of mechanical ventilation with specified ventilation management and adjunctive therapy protocols. The primary outcome was death at 28 days. Secondary outcomes included death at day 90, Sequential Organ Failure Assessment (SOFA) scores, other medical complications, ventilator weaning success, and neurologic status.

Of 396 eligible patients, 205 underwent randomization, and the final analysis included 99 in the conservative-oxygen group and 102 in the liberal-oxygen group. During the protocol days, the conservative-oxygen group received lower FiO2 compared to the liberal-oxygen group and also had statistically significant lower PaO2 and SpO2 levels. There was less use of prone positioning and lower positive end expiratory pressure (PEEP) levels in the conservative-oxygen groups. At 28 days, 34/99 patients in the conservative-oxygen group had died (34.3%) compared with 27/102 (26.5%) in the liberal-oxygen group, but this was not statistically significant (95% confidence interval [CI], -4.8 to 20.6). At 90 days, mortality was significantly higher in the conservative-oxygen group compared to the liberal-oxygen group (44/99 or 44.4% vs. 31/102 or 30.4%; 95% CI, 0.7 to 27.2). The conservative-oxygen group also had significantly more mesenteric ischemia events compared to the liberal-oxygen group (5 vs. 0).

The authors’ hypothesis of reduced mortality at 28 days with a conservative-oxygen approach was not confirmed in this study. Rather, they found a worrisome signal of increased mortality at 90 days and mesenteric ischemia. Because of these findings, the trial was stopped early.


LOCO2 was a well-designed, prospective, multicenter study, aiming to optimize oxygenation targets in ARDS. The authors were unable to support their hypothesis that conservative (lower) oxygenation at 88% to 92% would improve outcomes compared with liberal (higher) oxygenation at ≥ 96%, but actually found that outcomes might be worse. The authors described limitations that results could not be masked and that the lower oxygen targets may have exposed patients to unexpected and undetected hypoxemia.

This was not the first study to compare liberal and conservative oxygen targets. In the OXYGEN-ICU3 and CLOSE4 studies, the conservative groups had better outcomes compared to the liberal groups. The OXYGEN-ICU study, which targeted higher oxygen levels overall, with SpO2 goals of 94% to 98% and PaO2 goals of 70-100 in the conservative group, showed lower mortality in the conservative compared to a standard group that allowed PaO2 values up to 150 mmHg or SpO2 between 97% and 100%. The CLOSE study was a feasibility study targeting similar goals as the LOCO2 study, with SpO2 goals of 88% to 92% vs. ≥ 96%, and no significant differences were found. However, the conservative-oxygen arms of both of those trials had greater oxygen exposure compared with LOCO2, which had more hypoxemia. Subgroup analyses in OXYGEN-ICU found the lowest mortality in those with median time-weighted PaO2 between 87 mmHg and 93 mmHg, and an increased mortality in the group with median time-weighted PaO2 between 54 mmHg and 81 mmHg.

Based on LOCO2, I no longer confidently request that oxygen be lowered to target as near a PaO2 of 55 mmHg or an SpO2 of 88% as possible. In the era of COVID-19, our group has cautiously adopted a practice of targeting SpO2 of approximately 92% and higher, and PaO2 greater than about 60 mmHg. Although it represents a bit of extrapolation of data, it has not felt like a significant change in clinical practice, and I do not think it has delayed ventilator weaning or extubation. It has reminded me to be flexible, and flexibility in many areas has been key as we adapt to changes during COVID-19.

Based on the available data, I wonder if we should aim for PaO2 targets of 60 mmHg to 80 mmHg, 70 mmHg to 90 mmHg, and SpO2 targets of 92% to 96%, or perhaps something else. I hope that further study will help us clarify exactly what oxygen level we should target, but I recognize that fine-tuning a study to target such small differences would be challenging to design and implement.


  1. Acute Respiratory Distress Syndrome Network, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301-1308.
  2. Helmerhorst HJF, et al. Effectiveness and clinical outcomes of a two-step implementation of conservative oxygenation targets in critically ill patients: A before and after trial. Crit Care Med 2016;44:554-563.
  3. Girardis M, et al. Effect of conservative vs conventional oxygen therapy on mortality among patients in an intensive care unit: The Oxygen-ICU randomized clinical trial. JAMA 2016;316:1583-1589.
  4. Panwar R, et al. Conservative versus liberal oxygenation targets for mechanically ventilated patients: A pilot multicenter randomized controlled trial. Am J Respir Crit Care Med 2016;193:43-51.