Table 3. Multiple logistic regression model with favourable neurological outcome at hospital discharge as the dependent variable.

Independent variablea	Odds ratio	95% confidence interval	p value
Pulmonary artery catheter in place at time of arrest	9.85	2.00–48.65	0.005
Intravenous antiarrhythmics in place at time of arrest	2.93	1.38–6.24	0.005
Favourable neurological status 24 h before cardiac arrestb	2.88	1.72–4.85	<0.001
Myocardial infarction, this admission	2.73	1.43–5.22	0.002
VFc/ Pulseless VTd	2.58	1.50–4.47	0.001
Male	1.89	1.08–3.30	0.025
Age (year)	0.98	0.96–0.99	0.009
CPRe duration (min)	0.94	0.91–0.96	<0.001
Cardiopulmonary resuscitation during or before 2010	0.56	0.32–0.97	0.037
Renal insufficiency	0.45	0.26–0.77	0.004
Baseline depression in CNSf function	0.29	0.15–0.57	<0.001
Metastatic or hematologic malignant disease	0.08	0.02–0.32	<0.001

Goodness-of-fit assessment: n = 796, adjusted generalised R² = 0.4; the estimated area under the receiver operating characteristics curve = 0.86, with 95% confidence interval 0.83–0.90; and the Hosmer and Lemeshow goodness-of-fit Chi-Squared test p = 0.71.

^a The display of independent variables is arranged by descending order of odds ratio. The bold-typed variables were also components of Cardiac Arrest Survival Post-resuscitation In-hospital (CASPRI) score.

^b Patients with cerebral performance score 1 or 2

^c VF, ventricular fibrillation

^d VT, ventricular tachycardia

^e CPR, cardiopulmonary resuscitation

^f CNS, central nervous system