Post-resuscitation arterial oxygen and carbon dioxide and outcomes after out-of-hospital cardiac arrest

Objective To determine if arterial oxygen and carbon dioxide abnormalities in the first 24 h after return of spontaneous circulation (ROSC) are associated with increased mortality in adult out-of-hospital cardiac arrest (OHCA). Methods We used data from the Resuscitation Outcomes Consortium (ROC), including adult OHCA with sustained ROSC ≥1 h after Emergency Department arrival and at least one arterial blood gas (ABG) measurement. Among ABGs measured during the first 24 h of hospitalization, we identified the presence of hyperoxemia (PaO2 ≥ 300 mmHg), hypoxemia (PaO2 < 60 mmHg), hypercarbia (PaCO2 > 50 mmHg) and hypocarbia (PaCO2 < 30 mmHg). We evaluated the associations between oxygen and carbon dioxide abnormalities and hospital mortality, adjusting for confounders. Results Among 9186 OHCA included in the analysis, hospital mortality was 67.3%. Hyperoxemia, hypoxemia, hypercarbia, and hypocarbia occurred in 26.5%, 19.0%, 51.0% and 30.6%, respectively. Initial hyperoxemia only was not associated with hospital mortality (adjusted OR 1.10; 95% CI: 0.97–1.26). However, final and any hyperoxemia (1.25; 1.11–1.41) were associated with increased hospital mortality. Initial (1.58; 1.30–1.92), final (3.06; 2.42–3.86) and any (1.76; 1.54–2.02) hypoxemia (PaO2 < 60 mmHg) were associated with increased hospital mortality. Initial (1.89; 1.70–2.10); final (2.57; 2.18–3.04) and any (1.85; 1.67–2.05) hypercarbia (PaCO2 > 50 mmHg) were associated with increased hospital mortality. Initial (1.13; 0.90–1.41), final (1.19; 1.04–1.37) and any (1.01; 0.91–1.12) hypocarbia (PaCO2 < 30 mmHg) were not associated with hospital mortality. Conclusions In the first 24 h after ROSC, abnormal post-arrest oxygen and carbon dioxide tensions are associated with increased out of-hospital cardiac arrest mortality.