Abstract
Whether low peripheral oxygen saturation (SpO(2)) directed oxygen therapy is associated with lower mortality in critically ill patients needs further exploration. Adult critically ill patients from 11 intensive care units in China were screened. Participants were randomly assigned to the low SpO(2) (90%-95%) group or the high SpO(2) (≥96%) -group. The primary outcome was 28-day all-cause mortality. The secondary outcomes were hours free from ventilators and from renal replacement therapy (RRT) within 14 days. Note that 857 patients in the low SpO(2) group and 849 in the high SpO(2) group were included. In the low SpO(2) group versus the high SpO(2) group, the time-weighted average of the fraction of inspired oxygen (FiO(2)) was significantly lower (33.5 ± 9.7% vs. 39.6 ± 9.3%, p < 0.001), and so was the time-weighted average of SpO(2) (95.9 ± 1.8% vs. 98.0 ± 1.9%, p < 0.001). Within 28 days after randomization, 172 (20.1%) in the low SpO(2) group and 193 (22.7%) in the high SpO(2) group died (p = 0.180). Ventilator-free time and RRT-free time were not significantly different within 14 days. In critically ill patients, low SpO(2)directed oxygen therapy did not decrease 28-day mortality, 14-day ventilator-free time, or 14-day RRT-free time.