• F Dumas, C Farhenbruch, C Hambly, R T Donohoe, P Carli, A Cariou, and T D Rea.
• Emergency Medical Services Division of Public Health for Seattle and King County, Seattle, WA 98104, USA. florence.dumas@htd.aphp.fr
• Resuscitation. 2012 Jan 1;83(1):134-7.

AimsOptimal care for out-of hospital cardiac arrest (OHCA) patients may depend on the underlying aetiology of OHCA. Specifically chest compression only bystander CPR may provide greater benefit among those with cardiac aetiology and chest compressions plus rescue breathing may provide greater benefit among those with non-cardiac aetiology. The aim of this study was to generate a simple predictor model to identify OHCA patients with non-cardiac aetiology in order to accurately allocate rescue breathing.MethodsWe used two independent cohorts of OHCA patients from a randomized pre-hospital trial and a prospective hospital registry (total n=3086) to assess whether the characteristics of age, gender and arrest location (private versus public) could sufficiently discriminate non-cardiac aetiology. We used logistic regression models to generate a receiver operator curve and likelihood ratios.ResultsOverall, 965/3086 (31%) had a final diagnosis of a non-cardiac cause. Using 8 exclusive groups according to age, gender, and location, the frequency of non-cardiac aetiology varied from a low of 16% (55/351) among men >age 50 in a public location up to 58% (199/346) among women <60 in a private location. Although each characteristic was predictive in the logistic regression model, the area under the curve in the receiver operating curve was only 0.66. The associated positive likelihood ratios ranged from 1 to 3 and the negative likelihood ratios ranged from 1 to 0.4.ConclusionThe results highlight the challenge of accurately identifying non-cardiac aetiology by characteristics that could be consistently used to allocate bystander rescue breathing.Copyright © 2011 Elsevier Ireland Ltd. All rights reserved.

26 keywords...

hide…