Survival following cardiopulmonary resuscitation (CPR) from out-of-hospital cardiac arrest (OHCA) depends on numerous prehospital and in-hospital variables and interventions. The aim of this study was to develop a score to predict the resuscitation outcome after OHCA at hospital discharge.

All patients suffered OHCA between 01.01.2010 and 31.12.2016 with ROSC or ongoing CPR at hospital admission in Emergency Medical Service (EMS) systems with good quality in documentation in the German Resuscitation Registry (GRR) were included. The study population was divided into development dataset (5,775) and validation dataset (1,457) by random. Binary logistic regression analysis was used to derive the score. Hospital discharge with good neurological function (CPC 1-2 or mRS 0-2) was used as dependent variable, and various combination of potential predictor variables were used to create the model.

This study is a retrospective analysis of 8,603 prospectively documented OHCA patients between 01.01.2010 and 31.12.2016 within the German Resuscitation Registry (GRR),4 attended by EMS. The GRR represents currently approximately 160 emergency medical systems who record data on out-of-hospital resuscitation attempts throughout Germany, thus encompassing approximately 30 million citizens (total population of Germany counts 85 million).

The German Resuscitation Registry for out-of-hospital cardiac arrest is divided into two different datasets:

1. The 'Preclinical care' dataset originated from the Utstein-style template aiming at documentation of pre-hospital logistic issues, presumed aetiology, resuscitation therapy and patient's initial outcome including 118 variables.12
2. The 'Postresuscitation care' dataset is aimed at documentation of in-hospital post-resuscitation efforts. The participating hospitals can choose between a basic version which includes the use of coronary angiography, temperature management and the status at hospital discharged and an extended version. This includes 156 variables and inquires especially the status at admission, initial blood gas analysis, temperature management, coronary angiography and survival at 24h after cardiac arrest, 30 days and at hospital discharge exactly.12 Participation in the registry is voluntary. The participating emergency medical services and hospitals submit their data anonymously into a central database via a web-based application. Multiple plausibility checks have been implemented into this application in order to improve data quality. The registry is organised and funded by the German Society of Anaesthesiology and Intensive Care Medicine (DGAI).

Inclusion Criteria:

• OHCA between 01.01.2010 and 31.12.2016
• ROSC or ongoing CPR on hopsital admission
• high quality documentation with more than 75% documentation of post resuscitation care in hospital
• Patients cases documented in German Resuscitation Registry

Between 01.01.2010 and 31.12.2016 the 'Preclinical care' dataset contained 8,603 out-of-hospital CA patients with return of spontaneous circulation (ROSC) or ongoing CPR at hospital admission in EMS systems with good documentation quality. Good quality of documentation was defined by documented post-resuscitation care in more than 75%.

Exclusion Criteria:

• incomplete dataset
• age less than 18 years
• unknown initial ECG
• unknown age
• unknown neurological status on hospital discharge

1,371 patients were excluded from further analysis because of incomplete data in terms of age, neurological status at hospital discharge, unknown initial EKG or age < 18 years.

Between 01.01.2010 and 31.12.2016 the 'Preclinical care' dataset contained 8,603 out-of-hospital CA patients with return of spontaneous circulation (ROSC) or ongoing CPR at hospital admission in EMS systems with good documentation quality.