Arterial hyperoxia and in-hospital mortality after resuscitation from cardiac arrest

Table 5 Multiple logistic regression model with in-hospital mortality as dependent variable using EMShockNet model variables^a

Variable	OR (95%CI)	P value
Acute renal failure	3.3 (2.9 to 3.7)	<0.0001
Hypotension in first 24 hours^b	1.9 (1.7 to 2.0)	<0.0001
Age, decile	1.1 (1.1 to 1.1)	<0.0001
Emergency department origin	1.6 (1.4 to 1.7)	<0.0001
High heart rate^c	1.5 (1.3 to 1.6)	<0.0001
Hypoxia/poor O₂ exchange versus normoxia	1.4 (1.3 to 1.6)	<0.0001
Hyperoxia versus normoxia	1.5 (1.3 to 1.8)	<0.0001
Cancer	2.0 (1.5 to 2.5)	<0.0001
Cirrhosis	2.2 (1.5 to 3.1)	<0.0001
Female sex	1.2 (1.1 to 1.3)	<0.0001
Chronic renal	1.4 (1.1 to 1.6)	0.001
Chronic respiratory disease	1.3 (1.1 to 1.5)	0.002
Hepatic failure	2.7 (1.3 to 5.9)	0.01

^aEMShockNet, Emergency Medicine Shock Research Network; OR, odds ratio; 95% CI, 95% confidence interval. The following variables (OR (95% CI), P value) were removed from the model for nonsignificance (P < 0.01): immunosuppression (1.3 (1.0 to 1.7), P = 0.04), indigenous status (1.2 (0.9 to 1.5), P = 0.13), chronic cardiovascular disease (1.1 (0.9 to 1.2), P = 0.34), chronic liver disease (0.6 (1.0 to 3.3), P = 0.56) and hospital source prior to admission being from home (1.0 (0.9 to 1.1), P = 0.61). ^bDefined as any systolic blood pressure of less than 90 mmHg in the first 24 hours; ^cindicates highest value for first 24 hours in the intensive care unit (1 = exceeds median and 0 = median or lower).

ISSN: 1364-8535