'Event tree' analysis of out-of-hospital cardiac arrest data: confirming the importance of bystander CPR

Objective: The British National Service Framework (NSF) for heart disease commended the 'Utstein style' for auditing out-of-hospital cardiac arrests. The NSF also set standards for pre-hospital treatment and response times. To increase the flexibility of Utstein, an 'event tree' technique is proposed as an audit tool. Event trees consist of nodes and branches on which numbers, percentages or probability values are entered. Methods: Using the London Ambulance Service's (LAS) 1997 database on 3759 out-of-hospital cardiac arrests, 2772 arrests witnessed by lay bystanders or unwitnessed were analysed focusing on bystander cardiopulmonary resuscitation (BCPR) and response times. Results: The Utstein template showed that witnessed arrests in ventricular fibrillation (VF) or ventricular tachycardia (VT) who had received BCPR achieved a return of spontaneous circulation (ROSC) in the field significantly more often than non-BCPR recipients-26 versus 16% (P = 0.006). But the likelihood of being admitted to a hospital bed, and discharged alive, was only marginally better for BCPR recipients. To examine the influence of BCPR on the presenting rhythm an event tree showed that in 48% of witnessed BCPR cases the presenting rhythm was VF/VT, whereas, for witnessed non-BCPR cases, 27% were in VF/VT (P < 0.0001). With unwitnessed arrests, 31% of BCPR cases were in VF/VT compared with 18% for non-BCPR cases (P < 0.0001). Call to scene time was less than 8 min for 66% of all VF/VT arrests. Conclusion: The event trees, when combined with the Utstein template, demonstrated the importance of examining comprehensively datasets for both witnessed and unwitnessed cardiac arrests when monitoring performance standards. The analyses also emphasised the relevance of community programmes in Greater London for teaching basic life saving skills. (C) 2002 Elsevier Science Ireland Ltd. All rights reserved.