ABSTRACT
BACKGROUND: Providing supplemental oxygen is common in the management of critically ill patients, yet the optimal oxygen regimen remains unclear. OBJECTIVES: To explore the optimal range of PaO2 in critically ill patients. METHODS: This is a retrospective study conducted in the Medical Information Mart for Intensive Care III (MIMIC-III) database. The patients with a least 48 h of oxygen therapy were included. Nonlinear regression was used to analyze the association between PaO2 and mortality. We derived an optimal range of PaO2 and evaluated the association between the proportion of PaO2 measurements within this range and mortality. RESULTS: In total, 8401 patients were included in the study. A J-shaped relationship was observed between median PaO2 and hospital mortality. Compared with the reference group of 100-120 mmHg, patients with values of 80-100 mmHg and 120-140 mmHg had higher hospital mortality (adjusted odds ratio [aOR], 1.23; 95% CI, 1.05-1.43 and 1.29; 95%CI, 1.08-1.54, respectively). Similarly, mortality rates were significantly higher for PaO2 <80 mmHg and ≥140 mmHg (aOR, 1.97; 95%CI, 1.58-2.45 and 1.42; 95%CI, 1.19-1.69, respectively). Patients spent a greater proportion of time within 100-120 mmHg tended to have a lower mortality rate. CONCLUSION: Among critically ill patients, the relationship between median PaO2 and hospital mortality was J-shaped. The lowest rates of mortality was observed in those with PaO2 levels within 100 to 120 mmHg.