Are characteristics of hospitals associated with outcome after cardiac arrest? Insights from the Great Paris registry.

Abstract

As post-cardiac arrest care may influence patients' outcome, characteristics of receiving hospitals should be integrated in the evaluation of survival. We aimed at assessing the influence of care level center on patients' survival at hospital discharge using a regional registry of out-of-hospital cardiac arrest patients (OHCA). We retrospectively analysed a Utstein and in-hospital data prospectively collected for all non-traumatic OHCA patients, in whom a successful return of spontaneous circulation (ROSC) had been obtained, from a large metropolitan area (Great Paris). Receiving hospitals were categorized in 3 groups as follows: A centers (High-case volume with cath-lab 24/7), B centers (Intermediate-case volume with cath-lab partly available) and C centers (Low-case volume and no cath-lab) We compared patients' characteristics and outcome in the 3 groups and performed a multivariate logistic regression using survival to discharge as primary endpoint. Between May 2011 and December 2013, 1476 patients were admitted in 48 hospitals (group A: n=917; group B: n=428; group C: n=91). Overall survival rate at discharge was 433/1436 (30%). Patients' baseline characteristics significantly differed, as hospitals from group A received younger patients with a higher rate of shockable cardiac rhythms (p<0.001). Unadjusted survival rate differed significantly among the 3 groups of hospitals (respectively 34%, 25% and 15.4% for A-C, p<0.01). In multivariate analysis, the category of receiving hospital was no longer associated with survival, even in the subgroup of witnessed arrest and shockable patients. In this population-based study, characteristics of receiving hospitals are not associated with survival rate at discharge. This might be partially explained by the prehospital triage organization used in France.