Inhaled Isoflurane Sedation Did Not Lower Mortality in Hypoxemic Respiratory Failure. SAVE-ICU Found More Adverse Events.
In the 780-patient SAVE-ICU trial, published in JAMA on Oct. 14, hospital mortality was 42.5 percent with an isoflurane-based strategy and 43.8 percent with intravenous sedation. All five serious adverse events occurred in the inhaled arm.
Volatile anesthetics delivered through the ventilator circuit have been promoted as an alternative to intravenous sedatives such as propofol for mechanically ventilated patients. A pragmatic Canadian trial in acute hypoxemic respiratory failure has now tested whether the approach saves lives, and it did not find a survival benefit.
The SAVE-ICU investigators, led by Angela Jerath of Sunnybrook Research Institute in Toronto, reported in JAMA on Oct. 14 that hospital mortality was 42.5 percent (161 of 379) with an isoflurane-based inhaled sedation strategy and 43.8 percent (171 of 390) with usual intravenous sedation (risk ratio 0.98; 95% CI, 0.84 to 1.14; P = .85).
What the trial tested
SAVE-ICU was a pragmatic, open-label, superiority trial in 14 intensive care units in Canada. It enrolled adults on mechanical ventilation for acute hypoxemic respiratory failure, either non-COVID with a PaO2:FiO2 ratio below 300 mm Hg or COVID-19, who had been sedated for less than 72 hours. Enrollment ran from September 2020 to March 2025. Of 786 patients randomized, 780 were analyzed; mean age was 61.1 years, 73.7 percent had non-COVID respiratory failure and the mean PaO2:FiO2 ratio was 152.6 mm Hg.
The inhaled arm received isoflurane, or sevoflurane at two sites, with intravenous sedatives added as clinically needed. In practice 99.4 percent of those who received the intervention got isoflurane. Clinical teams titrated sedation to a target score and continued it until extubation or death. The trial was powered to detect a 10-percentage-point absolute reduction from an assumed 40 percent control mortality; observed control mortality was higher.
Secondary outcomes and safety
ICU-free, ventilator-free and hospital-free days were similar, as were quality-of-life and disability scores at three months. There was no effect modification by age, sex or COVID-19 status. The inhaled group used more adjunctive therapies for hypoxemia (63.9 percent vs. 55.0 percent) and had fewer delirium- and coma-free days through day 14 (difference, -0.95 days; 95% CI, -1.69 to -0.22). The authors say the gap was driven mainly by coma days, and offer three possible explanations: a learning curve with a new drug, deeper sedation to allow hypoxemia adjuncts, or chance across many unadjusted secondary outcomes.
Adverse events occurred in 35 patients (9.1 percent) in the inhaled group and 17 (4.3 percent) in the intravenous group. All five serious adverse events were in the inhaled group and all led to stopping the intervention. Among the events, the inhaled arm had abnormal liver function in five patients, bradyarrhythmia with QT prolongation in four and hypercapnia with respiratory acidosis in four; one patient on both inhaled and intravenous therapy developed propofol infusion syndrome. The authors write that it is unclear whether these events were related to isoflurane or to patient and disease factors.
SAVE-ICU tested a strategy, not a molecule, and the strategy did not save lives. That is enough to keep intravenous sedation as the default in this population.
Limits
The trial was open-label. Intravenous sedatives were common in the inhaled arm, given on any day in the first 14 days to 49.9 to 55.4 percent of patients, so the authors say the results cannot separate isoflurane's own effect from the overall strategy. Only 88.5 percent of the inhaled group actually received the intervention. Proficiency varied across centers, much of enrollment fell during the pandemic, and daily sedative doses and full sedation scores were not collected. The authors contrast their results with the earlier SESAR trial, in which sevoflurane in moderate to severe ARDS was associated with fewer ventilator-free days and higher mortality, and attribute the difference partly to SESAR's sicker cohort and deeper protocolized sedation. JAMA paired the trial with an editorial titled "New Evidence for Debate."
What it means in clinic
For intensivists, SAVE-ICU removes mortality as an argument for adopting inhaled sedation in acute hypoxemic respiratory failure. Units already using it can review their protocols against the trial's signals: liver function tests, QT monitoring, CO2 retention and depth of sedation, especially coma. Another JAMA trial the same day tested a different ICU adjunct, intraoperative dobutamine in septic shock, with a positive kidney result.
Reported, fact-checked and edited by The Script Pad's editorial staff.
This article is for professional education and does not replace clinical judgment. Treatment decisions should be based on the individual patient and current guidelines.
