1. Hypothesis The early elevation of the lower extremities during out-of-hospital cardiopulmonary resuscitation increases survival to one month by improving cardiac preload and blood flow to the heart and brain during chest compression.

1. Background The majority of sudden death cases have a cardiac origin and occur unexpectedly, often outside hospital. Attwood et al.[1] estimated the incidence of and survival from EMS-treated OHCA in Europe and found, for "all-rhythm" CA, an incidence of 37.72 per 100,000 person-years. Survival was 10.7% in "all-rhythm" CA. If these results were applied to the European population, approximately 275,000 persons would experience an all-rhythm, EMS-treated OHCA, with 29,000 persons surviving to hospital discharge[1].

   To resuscitate a person, without neurological damage, various efforts, which are described as the four links in the chain of survival (early call, early CPR, early defibrillation and early advanced life support), have to be optimal[2]. During the last decade, the quality and continuity of chest compressions have been increasingly highlighted[3]. The reason is that blood flow and coronary perfusion during cardiac arrest are related to the quality and continuity of chest compressions[4]. A coronary perfusion pressure (CPP) above 15mmHg, at defibrillation, also appears to be necessary for the return of spontaneous circulation (ROSC)[5]. Consequently, different methods and devices to improve blood flow to the heart (coronary perfusion) and brain during CPR, such as different types of mechanical compressor and impedance threshold device[6-9], have been studied.

   The initial CPR guidelines[10-12] stated that the "elevation of the lower extremities may promote venous return and augment artificial circulation during external cardiac compression". However, in the 1992 guidelines[13], this statement was removed. The reason for this decision was a lack of clinical evidence. During the last five years, the debate on how PLR may improve the outcomes of the resuscitation manoeuvres in CPR has been re-opened.

   According to Préau et al.[14], the effect of PLR is equivalent to a rapid intravenous volume expander by shifting blood from the lower extremities towards the intra-thoracic compartment. A 45° leg elevation for four minutes increases right and left ventricular preload and, by definition, the stroke volume, if the heart is preload dependent[15]. This makes PLR predictive of fluid responsiveness among patients with circulatory failure, e.g. sepsis and acute pancreatitis[14-17], and it has been recommended as part of haemodynamic monitoring in recent international recommendations[18]. Other researchers have also shown the benefit of using PLR to increase resistance to blood flow[19], thereby shifting fluid from the lower extremities to the central circulation[20, 21].

   The present study design is based on a pilot study recently conducted in Gothenburg, Sweden. This pilot study concluded that a 20° leg elevation during CPR improved the levels of end-tidal carbon dioxide (EtCO2) during CPR[22]. It has previously been concluded that EtCO2 correlates well with blood flow and that PLR induces an increase in descending aortic blood flow of at least 10% or in echocardiographic sub-aortic flow of at least 12%[23-26]. In other studies, EtCO2 has been shown to be quantitatively predictive of stroke volumes[27]. EtCO2 has also been described as an important value for predicting ROSC and CPR quality[22, 28, 29]. The resuscitation in the Gothenburg pilot study was performed using both manual and mechanical compressions made by LUCAS TM 2 (Lund University Cardiac Assist System), but the effect of PLR appeared to be greater during manual compressions. It was only possible to speculate on the reason for this, but the EtCO2 value started from a higher level in the mechanical group. The possible reason for this could be the "active decompression" creating a larger preload.

   Dragoumanos et al.[30] found in their animal study that the coronary perfusion pressure (CPP) also increased when PLR was performed during CPR and auto-transfusion of the aorta by PLR was the explanation. It is unclear whether this mechanism can be transferred to humans. The literature also includes some case reports and letters advocating PLR during CPR[31, 32]. However, no studies showing that PLR during CPR will increase survival have been conducted.

2. Method and design:

   A prospective, randomised, controlled trial in which all patients (>18 years) receiving out-of hospital CPR are randomised by envelope to be treated with either PLR or in a flat position. The ambulance crew use a special folding stool, which allows the legs to be elevated about 20 degrees.

   The PLR manoeuvre needs to be performed immediately (within five minutes) after the arrival of the first ambulance. Leg elevation has to be maintained while the patient receives chest compressions during CPR and has to be stopped when the patient has an ROSC or when a medical decision is made to interrupt these manoeuvres. PLR is to be performed at an angle of between 20 and 45 degrees (approximately 35 to 40 cm). An instruction video is produced for training prior to the study; the aim of using a specially designed folding stool is to standardise the intervention as much as possible.

   In the start-up phase between June 2013 and April 2014, the study has only been conducted in the City of Tarragona and the surrounding areas. In all, 13 mobile units (12 BLS and one ALS unit) will attend (attended) in the start-up phase. Since April 2014, a further 56 units, the whole province, have been participating in the study. The study will continue for three years.

3. Patient selection and randomization:

   Inclusion/exclusion Inclusion: All patients of both sexes who suffer an out-of-hospital cardiorespiratory arrest and require CPR and who are attended by the BLS and/or ALS units in the Tarragona area.

   Exclusion: Patients aged < 18 will be excluded from the study. Allocation concealment is ensured via opaque, numbered and sealed envelopes. The random allocation lists are generated by a web-based automated randomization system. To guarantee a numeric balance across conditions the randomisation will be performed separately in random permuted blocks of hundred. The allocation list will be kept in a remote secure location and an independent person randomly allocates the envelopes.

4. Endpoints:

   Primary end-point: survival to one month Secondary end-point: survival to hospital admission to one month and to one year with acceptable cerebral performance classification (CPC) 1-2 [33]

5. Evaluation of other result:

   Sub-group analysis: the result will also be analysed in relation to rhythm (shockable/non- shockable rhythm), age (more and less than 65 years), gender and ambulance delay (more and less than 10 minutes).

   Statistical analysis Group comparisons (PLR/flat position) will be performed using Fisher's non-parametric permutation test, the Mann-Whitney U test for continuous/ordered variables and Fisher's exact test for dichotomous variables.

   All tests will be two-tailed and p-values below 0.05 will be considered statistically significant.

6. Study time: April 1, 2012 -Mars 31 2015 (Sweden) Jun 8, 2013 - Jun 2016 (Catalonia)

Inclusion Criteria:

• Cardiac arrest where the rescue team perform chest compression.

Exclusion Criteria:

• Persons under 18 years