HOT-ICU

Clinical Question

In patients in the intensive care unit requiring supplemental oxygen, does a lower partial pressure of arterial oxygen (PaO2) target reduce mortality compared to a higher target?

Bottom Line

In patients with hypoxic respiratory failure, targeting a lower PaO2 of 60 mmHg did not improve outcomes in terms of 90-day mortality.

Major Points

Several trials have tried to evaluate optimal oxygen targets for patients requiring supplemental oxygen in the intensive care unit with varying results. While it is known that excessive oxygenation is harmful to patients prone to hypercapnea such as those with COPD, it is unclear if these same mechanisms related to free-radical generation and vasodilation leading worsening V/Q mismatch apply to patients with hypoxic respiratory failure. The Handling Oxygenation Targets in the ICU (HOT-ICU) trial compared a target PaO2 of 60 mmHg to 90 mmHg in terms of 90-day mortality, and did not find a significant difference. This is in line with prior trials such as the 2020 ICU-ROX study that did not find a difference in 180-day mortality or ventilator free-days comparing a conservative and liberal oxygenation approach.

Guidelines

As of May 2021, no guidelines have been published that reflect the results of this trial.

Design

• Multicenter, parallel-group, randomized, controlled trial
• Analysis performed with data monitoring committee blinded to oxygenation targets, but participants and clinicians were not blinded to trial group assignments
• N=2928
  • Lower oxygenation group(n=1447)
    • 21 excluded for loss to follow-up or consent
  • Higher oxygenation group (n=1441)
    • 19 excluded for loss to follow-up or consent
• Setting: 35 ICUs in Denmark, Switzerland, Finland, Netherlands, Norway, the United Kingdom, and Iceland.
• Enrollment: June 20, 2017 to August 3, 2020.
• Follow-up: 90-day
• Analysis: Intention-to-treat
• Primary outcome:

Population

Inclusion Criteria

• Age >/= 18 years
• Hypoxemia respiratory failure receiving at least 10L O2/min in an open system or FiO2 >/= 0.5 in a closed system

Exclusion Criteria

• Unable to randomize within 12 hours of ICU admission
• On chronic mechanical ventilation or home use of supplemental oxygen
• previous treatment with bleomycin
• organ transplant
• pregnant
• carbon monoxide, cyanide, or paraquat poisoning; methemoglobinemia
• Sickle cell disease

Baseline Characteristics

For Lower-oxygenation group

• Median age: 70 years
• Male sex: 63.7%
• Chronic obstructive pulmonary disease: 19.1%
• Chronic heart failure: 9.6%
• Ischemic heart disease: 14.1%
• Acute illness
  • Pneumonia: 57.7%
  • ARDS: 12.3%
  • Cardiac arrest: (10.3%)
  • Myocardial infarction: 5.8%
• Invasive Ventilation: 57.4%

Interventions

• Randomized to target PaO2 of 60 mmHg or PaO2 of 90 mmHg
  • oxygen targets achieved by adjusting FiO2

Outcomes

Comparisons are lower-oxygenation vs higher-oxygenation groups.

Primary Outcomes

90-day mortality

42.9% vs 42.4% (HR 1.02; 95% CI 0.94-1.11; P=0.64)

Secondary Outcomes

Median percentage of days alive without life support

87.8 vs 84.4 (p=0.10)

Subgroup Analysis

There was no evidence of heterogeneity of treatment effect in subgroups by presence of shock, patients on invasive ventilation, type of admission, presence of COPD, or presence of cardiac arrest. Heterogeneity was investigated for patients with traumatic brain injury but the small sample size limits any conclusions.

Adverse Events

Serious adverse events (shock, MI, ischemic stroke, intestinal ischemia)

36.1% vs 38.1% (HR 0.95; 95% CI 0.84-1.07; P=0.24)

Criticisms

• The lower oxygenation group was targeted to PaO2 60 mmHg, but the median PaO2 was 70.8 mmHg while the median PaO2 in the control arm was similar to the target of 90 mmHg.
• Clinicians were not blinded to treatment group