Survival from in-hospital cardiac arrest in Auckland City Hospital

To describe in-hospital resuscitation outcomes and factors associated with survival at Auckland City Hospital, New Zealand.

The Utstein template for in-hospital cardiac arrests was used. A retrospective audit of all cardiac arrests 2004-06 determined patient demographics, resuscitation time intervals, interventions, survival and neurological outcome at 12 months. Factors associated with survival to discharge were explored with logistic regression.

There were 3470 in-hospital deaths. Resuscitation was attempted in 415 patients (12%), with survival to discharge 27.2%. Survival was higher in first rhythm VT/VF (52.7% vs 13.1%, χ(2) = 75.3, P < 0.001), when the arrest was 'In-Hours' (41.4% vs 17%, χ(2) = 30.1, P < 0.001) and with younger age (mean [SD] for survivors 59.4 [7.1]vs 69.1 [14] for non-survivors). These associations were independent predictors of survival after multivariate logistic regression, with OR 6.2 (95% CI 3.6-10.5), 3.1 (95% CI 1.8-5.4) and 1.04 (95% CI 1.02-1.06), respectively (all P < 0.001). Other univariate predictors of survival; cardiac arrest team on site, monitored arrest and time to CPR were not significant after multivariate logistic regression. Time intervals to arrest interventions were short. Twelve month neurological outcome was good (CPC1 or 2) in 97.1% (95% CI 91.6-99.4) of survivors.

Survival from cardiac arrest in our hospital compared well to similar centres and good neurological outcome was higher than reported previously. Reduced survival during the 'After-Hours' period is cause for concern, and further research into the factors underlying this is required.