Abstracts / Resuscitation 96S (2015) 43–157 57 bystander-CPR and the probability of surviving a CA are partic- ularly low in the domestic setting. There is hardly any evidence why patients’ relatives at home start CPR measures rarely. One reason might be that the individuals capable of offering first aid in the domestic setting are older than in cases that occur in other locations. Our hypothesis was that awareness of CPR measures and self-confidence to initiate them are lower in older laypersons. To investigate this, a representative phone questionnaire sur- vey was carried out in Münster, a German city with a population of 300,000. Using computer-assisted telephone interviewing (CATI), we investigated whether the knowledge required to carry out bystander-CPR and self-confidence to do so differ between younger and older citizens. 2004 completed questionnaire where included in statistical analysis using Cramer’s V, Kendall’s tau-b coefficient and the chi-squared test as appropriate. When analysing the interviewees’ knowledge to carry out bystander-CPR, a lower level of information was found in older individuals. Among those under 65, for example, 82.4% gave the correct emergency number and 66.6% mentioned CPR as the pro- cedure needed. Among those aged over 64, these responses were only given by 75.1% and 49.5%, respectively (V = 0.082; P < 0.001 and V = 0.0157; P < 0.001). Questions regarding the participants’ self- assessments showed a similar picture: interviewees aged over 64 had a poorer assessment of their own abilities than those aged up to 65. Whereas 58.0% of the younger group of participants tended to think they would recognize a case of CA and 62.7% stated that they knew what needed to be done during CPR, interviewees aged over 65 only gave positive responses to these two statements in 44.6% and 51.3% of cases, respectively (V = 0.120; P < 0.001 and V = 0.103; P < 0.001). The data show that older citizens have less information about how to carry out lay CPR and are less confident about their ability to do so than younger individuals. The existence of differences of this type might provide an explanation for why older patients receive bystander-CPR less frequently, as potential providers of first aid in this group are potentially themselves also older. However, further investigation is necessary, as it is not possible to identify a causal connection in retrospective research. http://dx.doi.org/10.1016/j.resuscitation.2015.09.131 AP035 Cardiovascular MRI guides recommended hand position in CPR Sverre Nestaas 1,* , Knut Haakon Stensaeth 2 , Vigdis Rosseland 3 , Jo Kramer-Johansen 1 1 Division of Emergencies and Critical Care, Oslo University Hospital and University of Oslo, Oslo, Norway 2 Norwegian University of Science and Technology, Trondheim, Norway 3 Oslo University Hospital, Oslo, Norway Purpose of the study: Chest compressions during cardiopul- monary resuscitation (CPR) generate forward blood flow by compression of the ventricles. Compression of the left ventricular outflow tract (LVOT) or aortic root might impede this flow. Effec- tiveness of chest compressions can be altered by hand placement, but there is little scientific evidence regarding the optimum hand position. The European Resuscitation Council recommends plac- ing the hands on the lower half of the sternum, taught as “centre of the chest.” We aimed to relate structures of the heart to sur- face anatomy using MRI scans. We defined a hypothetical optimal Fig. 1. Axial scan illustrating how HOCP was defined. compression point (HOCP) in order to guide future studies of hand position recommendations. Materials and methods: Consecutive patients referred to car- diovascular MRI were recruited. We defined origo as the surface point of the centre of sternum at the level of the internipple line. Further, we defined the HOCP as the ventral surface point of a line in the axial scan perpendicular to the skin posteriorly with (1) both ventricles present; (2) the ventricular anterio-posterior diameters as large as possible; and (3) LVOT not present (Fig. 1). The distance from origo to HOCP was measured. Results: One hundred forty-four patients were enrolled. Patients were categorized into three subgroups based on conclu- sions from the cardiac MRI radiologist: cardiac disease (n = 74), aortic disease (n = 13) or no disease (n = 57). The mean (SD) distance from origo to HOCP was 32 (11) mm to the left and 16 (21) mm cau- dally. There was a non-significant increase in distance from origo to HOCP between the subgroups (no disease < cardiac disease < aortic disease). Conclusion: The centre of the chest might not be the optimum hand position for CPR. We found a hypothetical optimal compres- sion point 3 cm left of the centre of sternum. http://dx.doi.org/10.1016/j.resuscitation.2015.09.132 AP036 Size does matter: An incidence study of rib fractures due to CPR Youcef Azeli 1,* , Eneko Barbería 2 , Eva Valero-Mora 1 , María F. Jiménez-Herrera 3 , Christer Axelsson 4 , Alfredo Bardají 5 1 Emergency Medical System (SEM), Catalonia, Spain 2 Pathology Service, Institute of Legal Medicine, Catalonia, Spain 3 Nursin Department, Rovira i Virgili University, Tarragona, Spain 4 University of Borås, Borås/Va Götaland, Sweden 5 Cardiology Service, Joan XXIII University Hospital, Tarragona, Spain Purpose of the study: Rib fractures caused by chest com- pressions during cardiopulmonary resuscitation (CPR) have been associated with age, duration of CPR and a lower rate of return of spontaneous circulation. 1 Its relationship with anthropometric variables has rarely been described. The aim of this study is to analyze the relationship between some anthropometric variables and the incidence of rib fractures during CPR. Method: Clinical and Pathological prospective registry of out of hospital cardiac arrest that takes place in the region of Tarragona (ReCaPTa Study). For this study we included 53 non-traumatic car- diac arrest, over 18 years, receiving manual CPR and in whom an autopsy has been performed between April 2014 and January 2015.