Author + information
- Received May 14, 2017
- Revision received June 21, 2017
- Accepted June 22, 2017
- Published online August 21, 2017.
- Demetris Yannopoulos, MDa,∗ (, )
- Jason A. Bartos, MD, PhDa,
- Ganesh Raveendran, MDa,
- Marc Conterato, MDb,
- Ralph J. Frascone, MDc,
- Alexander Trembley, BSb,
- Ranjit John, MD, PhDd,
- John Connett, PhDe,
- David G. Benditt, MDa,
- Keith G. Lurie, MDa,
- Robert F. Wilson, MDa and
- Tom P. Aufderheide, MDf
- aDivision of Cardiology, Department of Medicine, University of Minnesota School of Medicine, Minneapolis, Minnesota
- bDepartment of Emergency Medicine, North Memorial Medical Center, Robbinsdale, Minnesota
- cDepartment of Emergency Medicine, Regions Hospital, St. Paul, Minnesota
- dDivision of Cardiothoracic Surgery, University of Minnesota School of Medicine, Minneapolis, Minnesota
- eDivision of Biostatistics, University of Minnesota, Minneapolis, Minnesota
- fDepartment of Emergency Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin
- ↵∗Address for correspondence:
Dr. Demetris Yannopoulos, University of Minnesota, UMN–Cardiology Division, 420 Delaware Street SE, MMC 508, Minneapolis, Minnesota 55455.
Background The prevalence of coronary artery disease (CAD) among patients with refractory out-of-hospital (OH) ventricular fibrillation (VF)/ventricular tachycardia (VT) cardiac arrest is unknown.
Objectives The goal of this study was to describe the prevalence and complexity of CAD and report survival to hospital discharge in patients experiencing refractory VF/VT cardiac arrest treated with a novel protocol of early transport to a cardiac catheterization laboratory (CCL) for extracorporeal life support (ECLS) and revascularization.
Methods Between December 1, 2015, and December 1, 2016, consecutive adult patients with refractory OH VF/VT cardiac arrest requiring ongoing cardiopulmonary resuscitation were transported by emergency medical services to the CCL. ECLS, coronary angiography, and percutaneous coronary intervention were performed, as appropriate. Functionally favorable survival to hospital discharge (Cerebral Performance Category 1 or 2) was determined. Outcomes in a historical comparison group were also evaluated.
Results Sixty-two (86%) of 72 transported patients met emergency medical services transport criteria. Fifty-five (89%) of the 62 patients met criteria for continuing resuscitation on CCL arrival; 5 had return of spontaneous circulation, 50 received ECLS, and all 55 received coronary angiography. Forty-six (84%) of 55 patients had significant CAD, 35 (64%) of 55 had acute thrombotic lesions, and 46 (84%) of 55 had percutaneous coronary intervention with 2.7 ± 2.0 stents deployed per patient. The mean SYNTAX score was 29.4 ± 13.9. Twenty-six (42%) of 62 patients were discharged alive with Cerebral Performance Category 1 or 2 versus 26 (15.3%) of 170 in the historical comparison group (odds ratio: 4.0; 95% confidence interval: 2.08 to 7.7; p < 0.0001).
Conclusions Complex but treatable CAD was prevalent in patients with refractory OH VF/VT cardiac arrest who also met criteria for continuing resuscitation in the CCL. A systems approach using ECLS and reperfusion seemed to improve functionally favorable survival.
The vast majority of all survivors of out-of-hospital cardiac arrest (OHCA) present to emergency medical services (EMS) with an initial shockable rhythm (ventricular fibrillation [VF]/pulseless ventricular tachycardia [VT]) (1–3). Even with this favorable presenting rhythm, >60% of patients with VF/VT are refractory to current treatment and never achieve return of spontaneous circulation (ROSC) or they die before they are admitted to the hospital (4).
Currently, the American Heart Association Advanced Cardiac Life Support (ACLS) guidelines recommend treating patients with refractory VF/VT at the scene of cardiac arrest for 30 to 45 min until they have ROSC or are declared dead (5).
Building on recent studies showing improved functionally favorable survival by rapid reversal of coronary artery ischemia after successful resuscitation following VT/VF OHCA (6–11), we implemented a new, systems-based approach that included early EMS transport of patients with refractory VF/VT to an ST-segment elevation myocardial infarction–receiving hospital with ongoing mechanical cardiopulmonary resuscitation (CPR). On arrival at the cardiac catheterization laboratory (CCL), early circulatory support with extracorporeal life support (ECLS), immediate angiography after ECLS, and percutaneous coronary intervention (PCI) were performed, as indicated. The present article describes the prevalence of coronary artery disease (CAD), incidence of acute coronary occlusion/stenosis, and the influence of reperfusion therapy on functionally favorable survival rates to hospital discharge in patients with refractory OH VF/VT cardiac arrest.
The University of Minnesota refractory VF/VT protocol has been described elsewhere (12). Briefly, early EMS transport criteria were as follows: 1) VF/VT OHCA as the first presenting rhythm; 2) 18 to 75 years of age; 3) three EMS-delivered direct current shocks and 300 mg of intravenous/intraosseous amiodarone without achieving ROSC; 4) body morphology able to accommodate a Lund University Cardiac Arrest System (LUCAS) automated CPR device; and 5) estimated transfer time from the scene to the CCL of <30 min (Figure 1).
Early EMS transport exclusion criteria included: 1) cardiac arrest of noncardiac etiology (e.g., blunt or penetrating traumatic arrest, burn-related, exsanguination, hanging, known overdose); 2) contraindications to mechanical CPR; 3) known pregnancy; 4) nursing home residents; 5) valid do not resuscitate/do not intubate orders; and 6) known terminal illness (e.g., cancer; end-stage liver, kidney, or heart disease).
North Memorial and St. Paul Fire EMS systems participated in the protocol. These 2 agencies service a population of 570,000 people in an area of approximately 1,100 square miles within a 30-min driving radius from the University of Minnesota. Patients meeting criteria were mobilized with ongoing CPR by using a LUCAS 2 automated CPR device (Physio-Control, Inc., Redmond, Washington) that compressed the chest 100 times/min. All patients had an advanced airway device placed. An inspiratory impedance threshold device (ResQPOD, Zoll Medical, Roseville, Minnesota) was used in all cases (2,13,14). Ventilation was provided in accordance with standard ACLS protocols (5,15). Patient treatment continued in the ambulance, and ACLS was performed until the patients arrived in the CCL (5).
A team of interventional cardiologists provided rotating around-the-clock call for response within 20 min of activation. Every patient requiring CPR on arrival was placed on the CCL table with the LUCAS device operating. The initial arterial access and ECLS initiation details have been described previously (12). At the initial arterial puncture, arterial blood gas and lactic acid samples were sent for processing.
ECLS was initiated in all patients meeting early transport criteria who had no Resuscitation Discontinuation Criteria and had not obtained ROSC by that time (Figure 1). ECLS was performed with a pre-primed CardioHelp circuit consisting of a centrifugal pump (Maquet Rotaflow, Maquet Cardiovascular, LLC, Wayne, New Jersey). Interventional cardiologists placed all devices.
Once hemodynamic/perfusion stability was obtained with either achievement of ROSC or initiation of ECLS, coronary angiography was performed and revascularization accomplished based on the clinical judgment of the interventional cardiologist. All coronary interventions were performed with intravenous heparin and an activated clotting time target of 250 to 300 s. At the initiation of ECLS, all patients received a bolus of unfractionated heparin (100 U/kg). At the end of the case, all patients treated with PCI were given aspirin and ticagrelor through a nasogastric tube. If the nasogastric tube could not be placed, cangrelor was infused intravenously until ticagrelor could be given. Standard doses were used for each of the medications (16–19).
Every patient arriving at the CCL was already hypothermic, with an average core temperature (blood temperature measured by the ECLS console) of 34°C (34.3 ± 0.7°C). The patient was maintained at 34°C for 24 h, except in the case of bleeding complications, when it was increased to 36°C (20–22). The criterion for hospital admission after ECLS initiation was achievement of sustained organized electrical rhythm with or without mechanical cardiac contraction. A multidisciplinary team of cardiology critical care, neurocritical care, and cardiothoracic surgery physicians managed patients admitted to the hospital in the cardiac intensive care unit.
Survival to hospital discharge with Cerebral Performance Category (CPC) 1 or 2 was the primary outcome. Secondary outcomes were 3-month survival with CPC 1 or 2 and protocol-based complications.
The Institutional Review Board of the University of Minnesota approved the anonymous data analysis and chart review extraction for this publication. Informed consent was waived. Historical comparison group patients were identified by using an anonymized cardiac arrest database (Cardiac Arrest Registry to Enhance Survival) (1,23).
To provide a historical group for the primary outcome comparison, resuscitation outcome data were analyzed from an immediately prior time period, starting January 1, 2014, and ending November 1, 2015, from the same 2 participating EMS systems with the following criteria: 1) OHCA of presumed cardiac etiology; 2) first presenting rhythm of VF/VT; 3) 18 to 75 years of age; and 4) received intravenous/intraosseous amiodarone 300 mg. Nursing home residents and patients with known terminal illness (e.g., cancer; end-stage liver, kidney, or heart disease) were excluded. Baseline demographic, cardiac arrest, and medical history characteristics are shown in Table 1.
A single, nonadjusted comparison was performed between our cohort and the historical comparison group with a Fisher exact test. Data are presented as mean ± SD.
From December 1, 2015, to December 1, 2016, a total of 72 patients with OHCA were transported by EMS. Sixty-two (86%) of the 72 patients met criteria for early EMS transport and represent the study population. Seven of these 62 (11%) patients met CCL Resuscitation Discontinuation Criteria on arrival and were declared dead. Fifty-five (89%) of 62 patients received continued CCL resuscitation and coronary angiography. Five (9%) of 55 patients had sustained ROSC, and 50 (91%) of 55 patients had ECLS initiated before coronary angiography. Forty-six (84%) of these 55 patients had revascularization with PCI. Of the 50 ECLS patients, 8 were declared dead in the CCL after 90 min because of failure to achieve a sustained organized electrical rhythm. Forty-seven (76%) of the 62 EMS-transported patients were admitted to the hospital (Figure 2).
Patient demographic characteristics, medical history, and resuscitation characteristics are shown in Table 1. The majority were white men with a mean age of 58 ± 10 years. The incidence of known CAD in the medical history was low (5 of 55 [9%]), and no patient or family member reported ischemic symptoms in the weeks, days, or immediate period before the cardiac arrest. Thirty-four (55%) of 62 cardiac arrests occurred in the home, 52 (84%) of 62 received bystander CPR, and all patients presented to EMS with VF/VT. The average time from the 911 call to CCL arrival was 58 ± 17 min. The mean time required to initiate ECLS was 6.1 ± 1.8 min from CCL arrival. The time from ECLS to balloon inflation was 6.0 ± 3.0 min.
Fifty-five patients received coronary angiography. Acute thrombotic lesions were present in 35 (64%) of the 55 patients, whereas 18 (33%) of the 55 had chronic total occlusions. Forty-six (84%) of 55 had significant CAD (>70% stenosis), and PCI was performed in all 46 (100%) patients. Single-vessel disease was present in 14 (30%) of 46 patients, and ≥2-vessel disease was present in 32 (70%) of 46 patients. Stents were implanted in 45 (98%) of 46 patients, with a mean of 2.7 ± 2.0 implanted stents/patient. The mean SYNTAX score was 29.4 ± 14 (Table 2) (24).
Forty-seven (76%) of 62 patients were admitted to the hospital (cardiac intensive care unit). Twenty-eight (45%) of the 62 patients were discharged alive, and 26 (42%) of 62 were discharged with favorable neurological function (CPC 1 or 2). Two (3%) of the 62 patients were discharged with unfavorable neurological function (CPC 3 and 4, respectively). At 3 months, 26 (42%) of the 62 patients were alive, and all had normal neurological function (CPC 1). At 3 months, 2 (3%) of the 62 patients were alive with CPC 3 and 4.
Left ventricular function was severely compromised in all admitted patients for the first 48 h, but significant recovery was observed within 5 days. An intra-aortic balloon pump was inserted in 25 (45%) of 55 patients. The mean left ventricular ejection fraction in survivors was 18 ± 19% at 24 h, 34 ± 19% at 48 h, 43 ± 16% at 5 days, and 48 ± 11% at hospital discharge. After CCL treatment, ECLS was continued for 3.0 ± 2.0 days.
We observed the following vascular complications from ECLS placement. Four patients had significant retroperitoneal bleeding requiring transfusion of multiple units of blood product. Three patients developed an ischemic leg after thrombosis of the distal perfusion catheters. Two of those patients had the cannulas removed, and there were no long-term complications. One patient developed an ischemic leg while still requiring cardiopulmonary mechanical support and was placed on central ECLS for an additional week. The patient was discharged from the hospital with CPC 1.
Of the 47 patients admitted to the hospital, 19 were declared dead within 5 ± 2 days. The 28 survivors were discharged from the hospital after 14 ± 18 days.
The following factors were associated with survival to hospital discharge: 1) earlier arrival of first responders after the 911 call; 2) lower lactic acid levels; 3) intermittent or sustained ROSC before arrival at the CCL; 4) higher end-tidal carbon dioxide upon CCL arrival; and 5) the presence of CAD as a reversible cause (Table 3).
The historical comparison group (170 patients) had an average age of 56 ± 7 years, 73% were men, and 78% were white. Cardiac arrest occurred at home 48% of the time, and 75% had bystander CPR (Table 1). Consistent with the EMS protocol at that time, resuscitation efforts were continued at the scene for 45 to 60 min, until ROSC was achieved or death was declared. In the historical comparison group, 26 (15.3%) of the 170 patients survived to hospital discharge with CPC 1 or 2 versus 26 (42%) of 62 patients with the current refractory VF/VT protocol (odds ratio: 4.0; 95% confidence interval: 2.08 to 7.7; p < 0.0001). This historical comparison group had a ROSC rate (37.0% vs. 35.9%) and rate of functionally favorable survival (15.3% vs. 18.8%) comparable to the study patients in ALPS (Amiodarone, Lidocaine, or Placebo Study), which also evaluated patients with refractory VF/VT OHCA (Central Illustration) (25).
The proportion of patients achieving each stage of the resuscitation continuum from emergency department arrival to hospital discharge in both groups is compared in Figure 3, highlighting the differences provided by ECLS capability.
We report, for the first time, the incidence of CAD in consecutive patients with refractory OH VF/VT cardiac arrest undergoing emergent coronary angiography. Previously, achievement of ROSC after cardiac arrest was a requirement before consideration of emergent angiography. Accordingly, previous reports of a variable incidence of CAD based on angiographic findings have been available only in patients resuscitated from OHCA (6). Advances in improved CPR hemodynamics and emergent ECLS now make it possible to evaluate and treat refractory OHCA.
We found a high prevalence of complex CAD, acute thrombotic lesions, and chronic total occlusions in the refractory OH VF/VT cardiac arrest patient group. The mean SYNTAX score of 29.4 ± 13.9 reported in the present study reflects the complexity and severity of the CAD observed (24). This finding, combined with the relatively high survival rates in our patients undergoing revascularization, substantiates a role of acute and/or chronic ischemia in the persistence of VF/VT cardiac arrest refractory to standard treatment. The severity of underlying coronary pathology is presumably causative or at least significantly contributory in the majority of patients. As such, continued treatment of this patient group with noninvasive strategies, such as ACLS protocols and medications, are unlikely to significantly improve survival without addressing reversal of the underlying pathophysiology.
A relevant finding of our study is that this patient group seems to be treatable. The availability of mechanical CPR, allowing safe and continued treatment during EMS transport while providing optimized blood flow during CPR, was central to our protocol. The application of ECLS also makes treatment of these patients feasible. With ECLS, immediate ROSC is not an outcome necessary for admission to the hospital and subsequent survival. ECLS provides mechanical/artificial ROSC for a patient group otherwise unable to initially achieve it; it also offers the opportunity to diagnose and treat reversible coronary artery occlusion present in the majority of patients. In addition, ECLS provides a necessary bridge to recovery from severe cardiogenic shock and stunned myocardium. We observed a remarkably predictable recovery from severe left ventricular dysfunction over a 2- to 5-day period if continued ECLS hemodynamic support was provided.
Given the complexity and severity of documented CAD, it is notable that the incidence of previously known CAD was low, and no patient or family member reported ischemic symptoms in the weeks, days, or immediate period before their event. Thus, cardiac arrest was the first clinical manifestation of well-established, severe CAD in the majority of patients.
The survival rates seen in our study are preliminarily encouraging. Our EMS–hospital emergency care system, focused on early EMS transport and reperfusion therapy, seemed to improve the rate of functionally favorable survival to hospital discharge (42%) versus that of the historical comparison group (15.3%). Whether this apparent survival advantage can be sustained or generalized to other centers remains to be determined. Irrespective, this experience supports the contention that successful treatment of this patient population can be achieved (Central Illustration).
An important concern raised with aggressive resuscitation protocols is the possibility that many survivors would be left with severe neurological impairments. However, our initial experience is reassuring in that 26 (93%) of 28 survivors were alive with normal function (CPC 1) at 3 months. These patients were often in the prime of their lives and have a reasonable expectation for continued quality and duration of life. Despite being refractory to standard treatment, the initial rate of unfavorable functional survival (2 of 62 [3%]) seems comparable with unfavorable outcomes reported for all patients with cardiac arrest treated with current therapy (4). Nonetheless, the risk/benefit ratio of this approach warrants careful and continued assessment.
Our study is, to our knowledge, the largest that has been reported in the United States. The SAVE-J (Study of Advanced Cardiac Life Support for Ventricular Fibrillation with Extracorporeal Circulation in Japan) trial, performed in Japan, was a prospective observational study comparing 454 patients with VF/VT arrest admitted to 46 hospitals over 3 years (26). Of these, 234 patients were provided ECLS. Patients received ECLS if they were admitted to an ECLS-capable center, whereas patients admitted to non-ECLS centers received standard therapy. Overall 1-month survival was improved in the ECLS group versus the non-ECLS group (29% vs. 6%). Johnson et al. (27) reported 26 cases of refractory cardiac arrest over a 7-year period with only 42% of those patients presenting with VF/VT. The University of Minnesota and SAVE-J cohorts were similar with regard to age, sex, location, and other demographic characteristics, and they both exclusively included patients treated for OHCA due to refractory VF/VT. Of note, both cohorts had identical proportions of patients who presented with acute coronary syndromes (64%). Differences that favored survival in the University of Minnesota cohort compared with the SAVE-J cohort were: 1) higher bystander CPR rates (84% vs. 48.8% for SAVE-J); and 2) higher proportion of patients who had witnesses to their cardiac arrest (80% vs. 71% for SAVE-J). The major difference that favored survival in the SAVE-J cohort compared with the University of Minnesota cohort was the much shorter time from the 911 call to hospital arrival (29 min vs. 58 min). Finally, in the CHEER (Mechanical CPR, Hypothermia, ECMO and Early Reperfusion) trial, Stub et al. (28) enrolled 11 patients with OHCA, and all presented with VF. Two had ROSC on arrival. Of the 9 patients placed on ECLS, 3 survived, and 5 of 11 survived to hospital discharge (45% survival rate), very similar to our reported outcomes.
The generalizability of our experience is unknown. Functionally favorable survival rates may increase or decrease with broader experience. As with any clinical protocol, selection bias cannot be excluded. Nonetheless, no patient meeting criteria for the protocol was missed from inclusion during the reported period. EMS providers over-transported 10 (14%) of 72 patients who did not meet EMS transport criteria. False-positive protocol activations were inevitable given this time-sensitive intervention and are currently being addressed with ongoing EMS quality improvement processes. A cost analysis was beyond the scope of this initial experience. However, an analysis of quality-adjusted life-years will be important and informative for this approach as it becomes more widely applied.
Complex but treatable CAD was prevalent in patients with refractory OH VF/VT cardiac arrest who also met criteria for continuing resuscitation in the CCL. A systems approach using ECLS and reperfusion seemed to improve functionally favorable survival.
COMPETENCY IN MEDICAL KNOWLEDGE: Complex but treatable CAD seems to be highly prevalent in patients experiencing refractory OH VF/VT cardiac arrest. A system of care including rapid transport of patients with ongoing refractory VF/VT to the hospital, where ECLS and reperfusion can be provided, improved functionally favorable survival. This approach substantiates a role of acute and/or chronic ischemia in the persistence of VF/VT cardiac arrest refractory to standard treatment.
TRANSLATIONAL OUTLOOK: Prospective clinical trials are necessary to delineate the aspects of this protocol critical for improved survival. Meanwhile, as the clinical availability of ECLS increases, use of this technology to treat patients with refractory cardiac arrest is likely to increase.
This work was supported by a philanthropic gift of the Bakken Family and the Robert K. Eddy Endowment for Resuscitation Medicine. Dr. Benditt has served as a consultant for Medtronic and Zoll; and has equity in Medtronic and Abbott. Dr. Lurie has a patent for impedance threshold and active compression-decompression devices with royalties paid; and is a consultant for Zoll. Dr. Aufderheide has served as the primary investigator for studies sponsored by JDP Therapeutics and Hospital Quality Foundation. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- advanced cardiac life support
- coronary artery disease
- cardiac catheterization laboratory
- Cerebral Performance Category
- cardiopulmonary resuscitation
- extracorporeal life support
- emergency medical services
- Lund University Cardiac Arrest System
- out-of-hospital cardiac arrest
- percutaneous coronary intervention
- return of spontaneous circulation
- ventricular fibrillation
- ventricular tachycardia
- Received May 14, 2017.
- Revision received June 21, 2017.
- Accepted June 22, 2017.
- 2017 American College of Cardiology Foundation
- Adabag S.,
- Hodgson L.,
- Garcia S.,
- et al.
- Aufderheide T.P.,
- Frascone R.J.,
- Wayne M.A.,
- et al.
- Link M.S.,
- Berkow L.C.,
- Kudenchuk P.J.,
- et al.
- Garcia S.,
- Drexel T.,
- Bekwelem W.,
- et al.
- Garot P.,
- Lefevre T.,
- Eltchaninoff H.,
- et al.
- Dumas F.,
- Bougouin W.,
- Geri G.,
- et al.
- Dumas F.,
- Cariou A.,
- Manzo-Silberman S.,
- et al.
- Yannopoulos D.,
- Bartos J.A.,
- Martin C.,
- et al.
- Sugiyama A.,
- Duval S.,
- Nakamura Y.,
- Yoshihara K.,
- Yannopoulos D.
- Yannopoulos D.,
- Aufderheide T.P.,
- Abella B.S.,
- et al.
- Aufderheide T.P.,
- Sigurdsson G.,
- Pirrallo R.G.,
- et al.
- Held C.,
- Asenblad N.,
- Bassand J.P.,
- et al.
- James S.K.,
- Storey R.F.,
- Khurmi N.S.,
- et al.,
- PLATO Study Group
- Steg P.G.,
- Harrington R.A.,
- Emanuelsson H.,
- et al.,
- PLATO Study Group
- Iqbal J.,
- Vergouwe Y.,
- Bourantas C.V.,
- et al.
- Kudenchuk P.J.,
- Daya M.,
- Dorian P.,
- Resuscitation Outcomes Consortium Investigators