Secondary Logo

Journal Logo

Clinical Science Aspects

Improving Patient Selection for Refractory Out of Hospital Cardiac Arrest Treated with Extracorporeal Life Support

Chouihed, Tahar∗,§,||; Kimmoun, Antoine†,§,¶,#; Lauvray, Adrien; Laithier, Francois-Xavier; Jaeger, Deborah; Lemoine, Simon; Maureira, Juan Pablo; Nace, Lionel; Duarte, Kevin||,¶,#; Albizzati, Stéphane; Girerd, Nicolas§,||,¶,#; Levy, Bruno†,§,¶,#

Author Information
doi: 10.1097/SHK.0000000000000941



Over the past decade, the management of out-hospital cardiac arrest (OHCA) has been focused on its earlier recognition by bystanders and the teaching of basic cardiopulmonary resuscitation (CPR). Despite the aforementioned improvements, patients with OHCA still suffer from poor survival, at approximately 12% (1). The absence of return of spontaneous circulation (ROSC) 20 min after initiation of advanced CPR defines refractory out of hospital cardiac arrest (ROHCA), which is systematically associated with death in the absence of invasive measures (2).

French guidelines for CPR consider Extracorporeal Life Support (ECLS) as one option in ROHCA patients with a no-flow less than 5 min and the absence of spontaneous circulation 30 min after initiation of advanced CPR (3). Duration of both pre-CPR arrest (no-flow) and CPR (low-flow) has been systematically highlighted as crucial prognostic factors in all observational studies focused on ROHCA (4).

To shorten the time to ECLS initiation, the most recent European Resuscitation Council guidelines recommend, in eligible ROHCA patients, a fast track access to ECLS implantation.

In light of these guidelines, our institution elaborated an operational strategy entitled “OSCAR-ECLS” (Out of hoSpital Cardiac ARrest–ExtraCorporeal Life Support), which was designed to improve the enrolment of eligible ROHCA patients and to reduce the delay time between recognition and ECLS initiation. The objective of the present study was to assess the delaysand patient survival before and after the implementation of the “OSCAR-ECLS” strategy.


This observational single-center study was conducted in a tertiary teaching hospital in France. The study was performed in the Greater Nancy metropolitan district, an area encompassing 143 km2 with 266,000 residents. All ROHCA patients eligible for ECLS and managed by the Mobile Intensive Care Unit (MICU) were included in the study. In this center, surgical ECLS implantation and primary coronary intervention (PCI) can be performed 24 h a day, 365 days a year. In accordance with French law, there was no need to obtain an IRB approval for a retrospective study. The study was submitted and approved by the Nancy Hospital Ethics Committee (CRENHU).

Two periods were individualized. The first period, between January 2010 and December 2011, was referred as the “before OSCAR-ECLS period,” whereas the second period, between August 2013 and July 2016, was referred as the “OSCAR-ECLS period.” Management changes were initiated after the very poor results obtained during the “before OSCAR-ECLS period.” The delay between December 2011 and August 2013 was required for the training of all the members of the prehospital and hospital teams (rescuers, physicians, and nurses).


For the “before OSCAR-ECLS period,” timing and decision for the implantation of an ECLS device in the event of ROHCA was mainly at the discretion of the prehospital emergency physician. No age limit was clearly defined for implantation.

The management of OHCA involves basic life support (BLS) and MICU, simultaneously dispatched to provide BLS and advanced life support (ALS) according to international guidelines. The MICU team consists of a driver, a dedicated nurse, and an emergency physician. The mean response time from «call to arrival on site» is 12 min. The OSCAR-ECLS period was essentially characterized by the rapid recognition of ROHCA (10 min) and transport by the MICU to the hospital such that the ECLS intervention could be performed, which is in contrast with the current ROHCA definition of the French guidelines (30 min) and the European Resuscitation Council guidelines (20–30 min) (2, 3).

This procedure encompasses two phases: identification of eligible ROHCA (Fig. 1) and fast track access to ECLS implantation. The OSCAR-ECLS procedure specifies that ROHCA must be witnessed with immediate onset of satisfactory bystander CPR and an immediate call to Emergency Dispatch. Furthermore, to minimize any loss of time, all qualified geographical locations are directly incorporated in the regulation software of the center for medical emergency control [Appli SAMU (4D Engine) version 7.0.8].

Fig. 1:
Description of the OSCAR-ECLS procedure and expected timing of each phase.

Fast track access to ECLS implantation

Patients suitable for ECLS implantation were identified, based on the aforementioned criteria, as early as possible by the Medical Dispatcher at the Emergency Call Dispatch Center (SAMU) (Fig. 1). Bystanders pursued CPR until MICU arrival. The mobile intensive care team verified at arrival that no-flow was under 1 min and that CPR was effective. Thereafter, the MICU team initiated advanced CPR, in which the patient was intubated and mechanically ventilated. The patient also received 1 mg of intravenous adrenaline every 5 min. On site, all members of the MICU team have a predefined role. An automated chest compression device (Autopulse, Zoll, Germany) was used to minimize interruption periods. The resuscitation was pursued during the transfer to hospital according to resuscitation guidelines (2).

The intensivists and cardiac surgeons stood ready to implant the ECLS surgically in the catheterization laboratory where a percutaneous coronary intervention (PCI) was subsequently performed. Before the OSCAR-ECLS period, all patients received 5,000 IU of heparin at ECLS initiation, whereas during the OSCAR-ELCLS period the heparin dose was begun 4 h after the ECMO run based on coagulation tests. The OSCAR-ECLS procedure ended when ECLS implantation was achieved.

Postresuscitation care

After ECMO implantation, the ICU physician inserted an echo-guided radial or brachial arterial catheter in the right arm. FiO2 and airflow of the ECMO oxygenator were adapted to maintain PaO2 between 60 and 150 mmHg and PaCO2 between 35 and 45 mmHg, with a prior adaptation of the ventilator for a protective ventilation associated with the lowest FiO2 using the right arm arterial catheter. Monitoring of heart function, and in particular the impact of ECMO on left ventricular afterload, was performed in all patients by echocardiography. Systemic anticoagulation was achieved by intravenous administration of unfractionated heparin; however, the timing of initiation of anticoagulation was dependent on the presence of clinical signs of bleeding and hemostasis parameters. Volume expansion was used in case of decreased blood flow or in case of jerking or shaking movements of the cannulae.

Mean arterial pressure was maintained between 65 and 75 mmHg by administering norepinephrine. In patients with no other obvious cause of cardiac arrest, PCI was performed immediately after ECMO implantation and hemodynamic stabilization. Intra-aortic balloon pump and low doses of dobutamine were used to unload the left ventricle during the “OSCAR-ECLS period.” Moderate hypothermia (33°–34°) was induced in all patients during 24 h when possible.

Variables and assessment

The following variables were collected from the medical report: baseline characteristics (date, patient identity, date of birth, telephone number of the family), witnessed status, bystander CPR, initial ECG rhythm, prehospital defibrillation, arrest location, and elapsed time between collapse and call to the Emergency Dispatch Center (SAMU), collapse and initiation of bystander CPR (no-flow duration), departure of the MICU and arrival on site, arrival on site and departure to hospital (time on site), departure and arrival to the catheterization laboratory (transfer time), arrival at the catheterization laboratory and ECLS implantation for the OSCAR population, initiation of CPR to ECLS run (low-flow period). The time between hospital arrival and ECMO run was missing for all non-OSCAR patients (n = 14). In this sensitivity analysis, these missing values were imputed by the mean calculated in the OSCAR-ECLS population (mean equal to 19 min).

The following assessments were made during the two study periods: survival at hospital discharge and neurological status in survivors, defined a priori according to the Glasgow Pittsburgh Cerebral Performance Category (CPC) (5) The CPC scale ranges from 1 to 5 with 1 representing intact function and 5 representing brain death. CPC 1 and CPC 2 are considered as good outcomes (6). The CPC score was performed by a senior intensivist 6 months after hospital release.

Statistical analysis

Statistical analysis was performed using SAS software version 9.4 (SAS Institute Inc, Cary, NC). Continuous data are described as mean (SD) or median [interquartile range] as appropriate and compared with the Wilcoxon test. Categorical variables are described as proportions and compared with Fisher exact test. Logistic regression analysis was performed to assess the associations between the OSCAR-ECLS period and characteristics. To highlight the key factors associated with the OSCAR-ECLS period, a logistic regression model was constructed using a stepwise-forward selection of variables with a threshold set at 5%. A significance threshold of P < 0.05 was adopted for all statistical analyses.


A total of 46 consecutive patients were included, 14 during the “before OSCAR-ECLS period” and 32 during the “OSCAR-ECLS period” (Table 1). Mean age was 46 (38–53) years, with most patients being male (38/46, 83%). Ventricular fibrillation was the initial rhythm in 37 of 46 patients. The median elapsed time between collapse and arrival to the catheterization laboratory was 80 (71–88) min before the OSCAR-ECLS procedure vs. 60 (51–72) min during the OSCAR-ECLS procedure (P < 0.0003) (Table 1). The time spent on site by the mobile intensive care team was significantly shortened from 48 (40–54) min before the OSCAR-ECLS period to 24 (20–28) min during the OSCAR-ECLS period (P < 0.0001). Transfer duration from cardiac arrest location to the catheterization laboratory was lower during the OSCAR-ECLS period: 20 (14–22) min before the OSCAR-ECLS procedure vs. 15 (9–18) min during the OSCAR-ECLS procedure (P = 0.033) (Table 1). Low-flow time was significantly lower during the OSCAR-ECLS procedure: 80 (65–94) min vs. 99 (90–107) min before the OSCAR-ECLS procedure (P < 0.0003). Similar results were obtained when comparing survivors versus nonsurvivors in the OSCAR-ECLS population (Table 2), namely a reduction in delay to hospital and a reduction in no-flow plus low-flow. Survival at hospital discharge was 7% (1/14) before the OSCAR-ECLS procedure and 25% (8/32) during the OSCAR-ECLS procedure (P = 0.24). Only one patient survived with a CPC score = 1 before the OSCAR-ECLS procedure, whereas during the OSCAR-ECLS procedure, eight patients (25%) survived, six with a CPC score = 1, one with a CPC score = 2, and one with a CPC score = 3 (Table 2). Before the OSCAR-ECLS procedure, 8 of 10 patients died within the first 24 h of ECLS. The mean initial ECLS flow was 4.3 L/min (3.4–5.2).

Table 1:
Demographics, arrest characteristics, and prehospital management
Table 2:
No- and low-flow duration in survivors versus nonsurvivors

Logistic regression

In univariable analysis, the variable most associated with the OSCAR-ECLS period was the time on site [odds ratio = 0.82 (0.73–0.92), P = 0.0009]. After adjustment for this time variable, the only variable significantly associated with the OSCAR-ECLS period was the time from collapse to call to the Emergency Dispatcher Center, although with a borderline significant effect [0.34 (0.12–0.99), P = 0.049] (supplemental Table 1,

Moreover, two predictors of the OSCAR-ECLS period were identified using a stepwise-forward selection procedure with a threshold set at 5%: the time on site and the time from collapse to call to the Emergency Dispatcher Center, both of which were associated with elapsed time to ECLS implantation (supplemental Table 2,


The main result of the present study is that, compared with before the OSCAR-ECLS procedure, prehospital delay times were significantly reduced and time spent on site was shortened by approximately 50% after implementation of the OSCAR-ECLS procedure.

Very few studies have focused on prehospital delays. In previous reports, most teams included patients after 20 or 30 min of advanced CPR with no ROSC, such as in the CHEER protocol (4).

To our knowledge, the present study is one of the first studies attempting to shorten the delay from cardiac arrest recognition to arrival to the catheterization laboratory in the event of ROHCA. Importantly, given the poor outcome observed before the OSCAR procedure, it was decided to markedly reduce the threshold time to identify candidates to ECLS, Indeed, during the OSCAR period, patients who underwent 10 min of advanced CPR on site by the mobile intensive care team (i.e., much lower than the current definition of ROHCA according to European guidelines, which requires 20–30 min of advanced CPR) were considered as candidates for ECLS.

Second, patient survival improved from 7% to 25% after the initiation of the “fast track access” to ECLS for ROHCA. This improvement in outcome may be explained by the rigorous patient selection in conjunction with the use of the OSCAR-ECLS strategy and by the reduction in low-flow delays. ECLS should be considered early by the medical dispatcher physician receiving the first call for suspected OHCA and its indication must be confirmed quickly by the on-site emergency physician performing the CPR if no ROSC is obtained within the first 5 min after MICU arrival. These results concur with the latest trial evidence, which emphasized the critical impact of precannulation delay on survival (7, 8). All of the OSCAR patients underwent a PCI which may be a factor in increasing the survival rate (9, 10). In the ”before OSCAR-ECLS” population, only 36% of the patients underwent PCI mainly because ECLS did not run (major ischemia–reperfusion syndrome). Another potential explanation may be related to the fact that the number of ECLS performed in our center has dramatically increased during the OSCAR-ECLS period (from 15 per year to 100 per year). Other studies report the same results on identical populations, inviting us to continue our evaluation (11, 12). Tonna et al. (13) argue for a strict definition of ECLS criteria for ROHCA (i.e., age, duration of no-flow, low-flow, and so on).

The present protocol is also a true call for multidisciplinary coordination, highlighting the concept of the “cardiac arrest team” as described in our protocol. This new paradigm revolves around two important notions. The first crucial key point of the OSCAR-ECLS procedure is the strict selection of patients. This observation was also reported in a recent meta-analysis by Kim et al. which also encourages physicians to implement «rigorous criteria for candidate selection» for ECLS (14). The second key point is to shorten the delay in setting up the ECLS. Indeed, ECLS should be considered early by the medical dispatch physician receiving the first call for suspected OHCA and its indication must be confirmed quickly (10 min after MICU arrival) by the on-site emergency physician.

The most noticeable novelty of this approach is probably the anticipation by the medical dispatcher that a patient could eventually have ROHCA as early as during the initial emergency call (Fig. 1), with only 10 min of advanced CPR on site triggering the entire ECLS process, including the immediate activation of the ECLS implantation and ICU teams. Thus, the two key points, the very early identification of selected candidates (e.g., prescreening before MICU arrival and confirmed within 10 min of MICU arrival) and the fast initiation of the ECLS, could be the cornerstone of ECLS outcome improvement defining the “golden hour concept.”

Study limitations

However, the present study is only a preliminary proof of concept pilot study and thus the results cannot be generalized. Cannulation time was not taken into consideration during the pre-OSCAR-ECLS period. Nevertheless, we may hypothesize that, at least, the implantation time was similar (same surgeons but less experienced). The surgeon estimated the median cannulation time at approximately 20 min for the “before OSCAR-ECLS period,” thus quite similar to the time needed to implant ECLS during the OSCAR-ECLS procedure. As a result, the only significantly reduced time owing to the OSCAR paradigm is the time spent on site. However, the dramatic decrease in this specific delay time was sufficient to ensure a less than 1-h delay between collapse and arrival in the catheterization laboratory. Using the same type of study, Dennis et al. (15) described the ECLS experience of two Australian ECLS centers. Thirty-five percent of the patients survived after refractory cardiac arrest and therefore the authors highlighted the importance of patient selection for this therapy.

In summary, an ECLS protocol aimed at enhancing patient selection and reducing prehospital delays significantly shortened low-flow duration. The very early selection of patients suitable for ECLS implantation, before MICU arrival, enabled that most patients (65%) had an elapsed time from collapse to arrival in the catheterization laboratory of less than 1 h.


To Pierre Pothier for editing the manuscript, to Dr. François Mougeolle, the Health Agencies of the «Meurthe et Moselle» department, the «Meurthe et Moselle» firefighter Rescuers for their participation in achieving this «OSCAR-ECLS» procedure, to the emergency and resuscitation team (SMUR Nancy), and the intensive care, cardiology and cardiac surgery department nurses of our hospital.


1. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, de Ferranti S, Després JP, Fullerton HJ, Howard VJ, et al. Heart disease and stroke statistics—2016 update: a report from the American Heart Association. Circulation 2016; 133 4:e38–e360.
2. Monsieurs KG, Nolan JP, Bossaert LL, Greif R, Maconochie IK, Nikolaou NI, Perkins GD, Soar J, Truhlář A, Wyllie J, et al. European Resuscitation Council Guidelines for Resuscitation 2015: Section 1. Executive summary. Resuscitation 2015; 95:1–80.
3. Adnet F, Baud F, Cariou A, Carli P, Combes A, Devictor D, Dubois-Randé JL, Gérard JL, Gueugniaud PY, Ricard-Hebon A, et al. Conseil français de réanimation cardiopulmonaire, Société française d’anesthésie et de réanimation, Société française de cardiologie, Société française de chirurgie thoracique et cardiovasculaire, Société française de médecine d’urgence, Société française de pédiatrie, et al. Guidelines for indications for the use of extracorporeal life support in refractory cardiac arrest. French Ministry of Health. Ann Fr Anesth Reanim 2009; 28 2:182–190.
4. Stub D, Bernard S, Pellegrino V, Smith K, Walker T, Sheldrake J, Hockings L, Shaw J, Duffy SJ, Burrell A, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation 2015; 86:88–94.
5. Jacobs I, Nadkarni V, Bahr J, Berg RA, Billi JE, Bossaert L, Cassan P, Coovadia A, D’Este K, Finn J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. Resuscitation 2004; 63 3:233–249.
6. Phelps R, Dumas F, Maynard C, Silver J, Rea T. Cerebral performance category and long-term prognosis following out-of-hospital cardiac arrest. Crit Care Med 2013; 41 5:1252–1257.
7. Jouffroy R, Lamhaut L, Philippe P, An K, Carli P, Vivien B. A new approach for treatment of refractory ventricular fibrillation allowed by extra corporeal life support (ECLS)? Resuscitation 2014; 85 8:e118.
8. Wang C-H, Chou N-K, Becker LB, Lin J-W, Yu H-Y, Chi N-H, Hunag SC, Ko WJ, Wang SS, Tseng LJ, et al. Improved outcome of extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest—a comparison with that for extracorporeal rescue for in-hospital cardiac arrest. Resuscitation 2014; 85 9:1219–1224.
9. Dumas F, Bougouin W, Geri G, Lamhaut L, Rosencher J, Pène F, Chiche JD, Varenne O, Carli P, Jouven X, et al. Emergency percutaneous coronary intervention in post-cardiac arrest patients without ST-segment elevation pattern. JACC Cardiovasc Interv 2016; 9 10:1011–1018.
10. Dumas F, Cariou A, Manzo-Silberman S, Grimaldi D, Vivien B, Rosencher J, Empana JP, Carli P, Mira JP, Jouven X, et al. Immediate percutaneous coronary intervention is associated with better survival after out-of-hospital cardiac arrest: insights from the PROCAT (Parisian Region Out of Hospital Cardiac Arrest) Registry. Circ Cardiovasc Interv 2010; 3 3:200–207.
11. Ha TS, Yang JH, Cho YH, Chung CR, Park C-M, Jeon K, Suh GY. Clinical outcomes after rescue extracorporeal cardiopulmonary resuscitation for out-of-hospital cardiac arrest. Emerg Med J 2017; 34 2:107–111.
12. Schober A, Sterz F, Herkner H, Wallmueller C, Weiser C, Hubner P, Testori C. Emergency extracorporeal life support and ongoing resuscitation: a retrospective comparison for refractory out-of-hospital cardiac arrest. Emerg Med J 2017; 34 5:277–281.
13. Tonna JE, Johnson NJ, Greenwood J, Gaieski DF, Shinar Z, Bellezo JM, Becker L, Shah AP, Youngquist ST, Mallin MP, et al. Practice characteristics of Emergency Department extracorporeal cardiopulmonary resuscitation (eCPR) programs in the United States: the current state of the art of Emergency Department extracorporeal membrane oxygenation (ED ECMO). Resuscitation 2016; 107:38–46.
14. Kim SJ, Kim HJ, Lee HY, Ahn HS, Lee SW. Comparing extracorporeal cardiopulmonary resuscitation with conventional cardiopulmonary resuscitation: a meta-analysis. Resuscitation 2016; 103:106–116.
15. Dennis M, McCanny P, D'Souza M, Forrest P, Burns B, Lowe DA, Gattas D, Scott S, Bannon P, Granger E, et al. Extracorporeal cardiopulmonary resuscitation for refractory cardiac arrest: a multicentre experience. Int J Cardiol 2017; 231:131–136.

Extracorporeal membrane oxygenation; hypothermia; induced; out-of-hospital cardiac arrest

Supplemental Digital Content

Copyright © 2017 by the Shock Society