The Association Between Early Arterial Oxygen Tension and in-Hospital Mortality in Patients With Aneurysmal Subarachnoid Hemorrhage Admitted to ICUs in Australia and New Zealand: A Retrospective Multicenter Cohort Study

BACKGROUND: Variable systemic oxygenation may affect mortality in critically ill patients with aneurysmal subarachnoid hemorrhage (aSAH). METHODS: In this multicenter retrospective cohort study, data were extracted from the Australian and New Zealand Intensive Care Society Adult Patient Database. Adult patients with aSAH admitted to an intensive care unit (ICU) between January 2018 and June 2024 were included. The arterial oxygen tension during the initial 24 hours in ICU (D1 PaO2), which yielded the highest Acute Physiology and Chronic Health Evaluation III-J oxygenation subscore, was the primary exposure variable. The primary outcome was in-hospital mortality. A multivariable mixed-effects logistic regression model was constructed to determine the adjusted probability of in-hospital mortality based on D1 PaO2. RESULTS: A total of 6446 patients were included. The median D1 PaO2 was 92 mm Hg (IQR: 74 to 138), in-hospital mortality was 20.0% (n=1291), and 12-month mortality was 25.6% (n=1647). A U-shaped multivariable relationship was observed between D1 PaO2 and in-hospital mortality, with 79 mm Hg being associated with a nadir of 16.2%. This varied according to age (65 y and over: 69 mm Hg; below 65 y: 83 mm Hg), admission type (medical: 77 mm Hg; surgical: 89 mm Hg), intubation status (not intubated: 90 mm Hg; intubated: 143 mm Hg), and Glasgow Coma Score (GCS) (GCS 15: 75 mm Hg; GCS 13 to 14: 65 mm Hg; GCS 7 to 12: 96 mm Hg; GCS 3 to 6: 131 mm Hg). CONCLUSIONS: In critically ill patients with aSAH, a U-shaped relationship was noted between D1 PaO2 and in-hospital mortality. This varied according to patient demographics and illness severity.