Accepted Manuscript Title: A retrospective Please check the presentation of article title footnotes, and correct if necessary.–>study of pulseless electrical activity, bedside ultrasound identifies interventions during resuscitation associated with improved survival to hospital admission. A REASON Study Authors: Romolo Gaspari, Anthony Weekes, Srikar Adhikari, Vicki Noble, Jason T. Nomura, Daniel Theodoro, Michael Woo, Paul Atkinson, David Blehar, Samuel Brown, Terrell Caffery, Emily Douglass, Jacqueline Fraser, Christine Haines, Samuel Lam, Michael Lanspa, Margaret Lewis, Otto Liebmann, Alexander Limkakeng, Fernando Lopez, Elke Platz, Michelle Mendoza, Hal Minnigan, Christopher Moore, Joseph Novik, Louise Rang, Will Scruggs, Christopher Raio PII: DOI: Reference:

S0300-9572(17)30607-X http://dx.doi.org/10.1016/j.resuscitation.2017.09.008 RESUS 7311

To appear in:

Resuscitation

Received date: Revised date: Accepted date:

26-5-2017 7-9-2017 11-9-2017

Please cite this article as: {http://dx.doi.org/ This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

TITLE – A retrospective study of Pulseless Electrical Activity, Bedside ultrasound identifies interventions during resuscitation associated with improved survival to hospital admission. A REASON Study. Authors Romolo Gaspari, MD, Ph.D*.

University of Massachusetts Medical School, Worcester, MA

Anthony Weekes, MD Carolinas Medical Center, Charlotte, NC Srikar Adhikari, MD Vicki Noble, MD

University of Arizona, Tucson, AZ Massachusetts General Hospital, Boston, MA

Jason T Nomura, MD

Christiana Care Health Systems, Newark, DE

Daniel Theodoro, MD MSCI

Washington University School of Medicine, St Louis, MO

Michael Woo, MD

University of Ottawa, Ottawa, ON, Canada

Paul Atkinson, MD

Dalhousie University, Saint John, NB, Canada

David Blehar, MD

University of Massachusetts Medical School, Worcester, MA

Samuel Brown, MD

University of Utah, Salt Lake City, UT

Terrell Caffery, MD

LSU Health Sciences Center, Baton Rouge, LA

Emily Douglass, BA

Massachusetts General Hospital, Boston, MA

Jacqueline Fraser, RN

Saint John Regional Hospital, Saint John, NB, Canada

Christine Haines, MD Samuel Lam, MD

North Shore University Hospital, Manhasset, NY Advocate Christ Medical Center, Chicago, IL

Michael Lanspa, MD

University of Utah, Salt Lake City, UT

Margaret Lewis, MD

Carolinas Medical Center, Charlotte, NC

Otto Liebmann, MD

Brown University, Providence, RI

Alexander Limkakeng , MD

Duke University School of Medicine, Durham, NC

Fernando Lopez, MD

Duke University School of Medicine, Durham, NC

Elke Platz, MD

Brigham and Women's Hospital, Boston, MA

Michelle Mendoza, MD University of Massachusetts Medical School, Worcester, MA Hal Minnigan, MD, PhD Indiana University, Tiptin, IN Christopher Moore, MD Yale University School of Medicine, New Haven, CT Joseph Novik, MD

NYU Bellevue Hospital, New York, NY

Louise Rang, MD

Kingston General Hospital, Kingston, ON, Canada

Will Scruggs, MD

Castle Hospital, Kailua, HI

Christopher Raio , MD North Shore University Hospital, Manhasset, NY The authors represent the REASON research network. (Real-time Evaluation and Assessment with Sonography Outcomes Network). *Corresponding Author Romolo J Gaspari, MSc, MD, PhD Department of Emergency Medicine UMASS Memorial Medical Center

Worcester MA, 01655 Phone – 508-334-7943 Fax – 508-421-1400 Email – [email protected] Meeting – This abstract and data was presented at the national SAEM conference in 2015 in San Diego CA. Grant – There was no funding for this research. Conflict of Interest Word Count – 1840

RG, AW, SA, VN, JN, and DT conceived the study and designed the trial, supervised the conduct of the trial and data collection. All authors undertook recruitment of participating centers and patients and managed the data, including quality control. RG and DT provided statistical advice on study design and analyzed the data; RG chaired the data oversight committee. RG drafted the manuscript, and all authors contributed substantially to its revision. RG takes responsibility for the paper as a whole.

Keywords: Cardiac Arrest; Pulseless Electrical Activity; Ultrasound;Point-of-Care Ultrasound; Bedside Ultrasound

INTRODUCTION Background

No large studies explore how point of care ultrasound (POCUS) should be utilized during advanced cardiac life support (ACLS). Over the last decade, however, several studies concluded that POCUS might have a prognostic role. 1-3 Defining a role for POCUS during cardiac arrest is important because intra-arrest ultrasound features may identify a patient population that would benefit from specific interventions that deviate from standard ACLS protocols 2. An example is when a large pericardial effusion is visualized on ultrasound in the setting of pulseless electrical activity (PEA) arrest resulting in a drainage procedure rather than another dose of epinephrine. PEA may be defined as a syndrome of electrical cardiac activity in an unconscious patient without a palpable pulse not attributed to a ventricular tachyarrhythmia. PEA may span a continuum from those with electrical activity but no mechanical activity (the concept of true electro-mechanical dissociation) to those with electromechanical activity accompanied by hypotension so profound that palpable pulses go undetected (referred to as “pseudo-pulseless electrical activity).4,5 While some literature suggests electrocardiogram tracings may assist in discerning these distinct pathophysiologic states; existing ACLS protocols do not discriminate between these two conditions.6,7 The application of POCUS in cases of PEA cardiac arrest was first described in 1988.5 Using ultrasound to differentiate sonographic features of distinct categories of patients in PEA, however, and associating those sonographic features with clinical outcomes has not been described. POCUS may identify a subset population of patients in PEA who respond differently to therapeutic interventions during resuscitation following PEA cardiac arrest. Combining data from a previously published prospective observational study on ultrasound use during cardiac arrest with a review of the ultrasound imaging from that study, we sought to describe the different survival effect following therapeutic interventions on sub-populations of PEA as defined by sonographic characteristics.

METHODS Study Setting and Design This study analyzes data from a 20 site multi-center, prospective, observational trial involving sites across the United States and Canada focusing on ultrasound and cardiac arrest (Clinical Trials Registry NCT01446471) performed by the Real-time Evaluation and Assessment Sonography Outcomes Network (REASON). Sites involved in the study included a single central coordinating site, six geographic regional centers, and 13 local sites reporting to the regional centers. Study site characteristics have been previously published.8 Data were uploaded into a centralized database (REDCap) with Institutional Review Board approval for each site obtained from the respective institutions. Patient Population The larger database consisted of patients presenting to the emergency department (ED) in cardiac arrest, receiving resuscitation per standard American Heart Association ACLS protocol, found to be in either asystole or PEA, and had an ultrasound performed during the resuscitation. For the current study we analyzed patients from the previous study who presented with PEA as the initial rhythm upon arrival to the emergency department. We excluded patients with no cardiac activity on the initial ultrasound. Interventions were reviewed and categorized as standard ACLS interventions or interventions outside of the ACLS protocol. Only those interventions performed during the resuscitation and before documented return of spontaneous circulation (ROSC) were included in the analysis. Ultrasound Imaging Ultrasound imaging performed in this study was limited to transthoracic echocardiography with imaging performed during pauses for pulse checks or rhythm checks. The study protocol was designed to prevent any delay or pause in CPR. The physician performing the resuscitation interpreted the ultrasound in real-time during the resuscitation and was not blinded to the interventions or ultrasound results. The initial interpretation was recorded

as positive or negative for any cardiac activity. A secondary image review was performed of recorded ultrasound images blinded to any clinical or therapeutic information. Images were categorized as having any cardiac activity or not, and further categorized as having organized cardiac activity or non-organized movements. Organized cardiac activity was defined as movement of the myocardium with change in size of the ventricular cavity and synchronized movement of the ventricular wall. Non-organized cardiac activity was defined as movement of the myocardium without change in size of the ventricular cavity, e.g. agonal or twitching activity.5 Cardiac activity was determined as organized or disorganized using visual estimation. All images were reviewed independently, and blinded to other interpretations by a pair of investigators to assess agreement. Agreement between investigators was assessed using a kappa analysis. Outcomes The primary outcome was survival to hospital admission. Secondary outcomes included ROSC and survival to hospital discharge. Utstein nomenclature was used for study data and included recommended data points with the exception of neurologic outcomes9. Statistical Analysis Data were obtained and uploaded into a central database from study sheets, initial patient encounter, patient records, and EMS records. Data elements included in the study database were chosen based on previously published studies10-13. Data collection methods were discussed in our prior publication.8 Univariate analysis was performed with the use of MannWhitney U test for continuous variables and Fisher’s exact test for dichotomous variables. Univariate analysis was first performed independently for two outcomes of 1) survival to hospital admission and 2) ROSC. Results are provided as odds ratios (95% CI) with p values. Similar to prior research, variables with p < 0.2 from univariate analysis were included in initial multivariate modeling14. RESULTS

In our cohort of 225 patients in PEA cardiac arrest with cardiac activity on ultrasound, the overall survival rate to ROSC, hospital admission and hospital discharge was low at 55.5%, 32.0% and 1.8% respectively. Of the 225 patients with cardiac activity, 33.3% (n=75) had organized activity, 42.2% (n=95) had disorganized activity and 24.4% (n=55) had insufficient imaging to determine the type of activity. In these cases (n=55) with short amount of time to obtain images during resuscitation, cardiac activity was visualized, but there was insufficient image resolution to categorize the type of activity. Agreement between image reviewers for categorizing activity as organized or disorganized was good (kappa 0.77). Baseline characteristics of PEA cardiac arrest patients with organized and disorganized cardiac activity on ultrasound are summarized in Table 1. Patients with organized cardiac activity demonstrated increased survival rates following standard ACLS interventions (FIGURE 1). Patients with organized cardiac activity treated with bolus epinephrine demonstrated a ROSC and survival to hospital admission rate of 54.7% and 37.7%. Patients with disorganized activity treated with standard ACLS interventions demonstrated lower ROSC (37.2%) and survival to hospital admission (17.9%), p<0.005). There was no difference in survival to hospital discharge. The majority of patients in PEA with any cardiac activity on ultrasound (56.0%) were treated following established ACLS protocols with bolus epinephrine as the predominant therapeutic agent. A significant minority (99 of 225, 44.0%) of patients were treated with medical interventions that are not specified in ACLS protocol for PEA (Table 2). The most common intervention was Calcium Chloride followed by continuous intravenous adrenergic agents. Continuous intravenous adrenergic agents were initiated during the resuscitation after the ultrasound demonstrated cardiac activity in 39 patients, with dopamine (17 patients) or norepinephrine (15 patients) being the most common adrenergic agents. Other adrenergic agents included continuous epinephrine, dobutamine and neosynephrine. Patients in PEA arrest with organized cardiac activity visualized on ultrasound and started on continuous adrenergic agents during the resuscitation demonstrated ROSC, survival to hospital admission and survival

to hospital discharge of 90.9%, 45.5%, and 4.5% respectively. Patients given continuous intravenous adrenergic agents also received bolus doses of epinephrine, but patients who survived to hospital admission received less epinephrine, 3.0 mg (95%CI-1.2 to 4.8) versus 5.0mg (95%CI – 3.5 to 6.5). Treatment with continuous adrenergic intravenous agents was not associated with increased survival in patients with disorganized cardiac activity. Patients in PEA arrest with dis-organized cardiac activity visualized on ultrasound and started on continuous adrenergic agents during the resuscitation demonstrated ROSC, survival to hospital admission and survival to hospital discharge of 47.1%, 0.0%, and 0.0% respectively. This is not statistically different then patients with disorganized cardiac activity treated with standard ACLS medications (37.2%, 17.9%, 1.3% respectively). Multivariate analysis identified 2 variables that were associated with survival to hospital admission in patients with cardiac activity receiving IV adrenergic agents, age and bystander CPR. These variables were included in the model for survival to ROSC for consistency of the model. Patients with organized cardiac activity receiving continuous intravenous adrenergic agents were 8.2 (95% CI 1.8-32.1) times more likely to survive to ROSC compared to patients with organized activity that did not receive intravenous adrenergic agents. Patients with disorganized activity who received IV adrenergic agents showed little to no improvement in survival (OR 1.5, 95% CI 0.5 to 4.3). Survival to hospital admission was not significantly different for patients without organized cardiac activity who received IV adrenergic agents (1.37, 95% CI 0.5 – 3.8) compared to patients with disorganized activity who did not receive IV adrenergic agents (0.00, 95% CI 0.0 – 1.0), p=0.078. LIMITATIONS Our study has several limitations. First, these are observational findings from a non-randomized study and association is not the same as causality. Second, this is a subset analysis of patients from a larger

prospective study, so it may not be representative of the population of patients undergoing resuscitation from PEA cardiac arrest. The bedside sonographers were not asked in the original study to comment on organized vs disorganized cardiac activity. Interpretation by reviewers for this study was not inferred, but represents an interpretation of a recorded image. We acknowledge that the agreement between image reviewers for categorizing cardiac activity as organized vs disorganized was not perfect but good (kappa 0.77). Finally, we did not have sufficient numbers to comment on survival to hospital discharge and the confidence intervals of the 95% CI are wide due to the small sample size. DISCUSSION This study suggests that patients with PEA and organized cardiac activity on bedside ultrasound respond differently to medical ACLS interventions compared to cases with agonal cardiac activity or cardiac twitching. Ultrasound has been proposed as an adjunct to ACLS, but the exact role of ultrasound in resuscitation following cardiac arrest continues to be debated. The ability of ultrasound to determine a subset of patients in PEA for specific therapies would fundamentally change ACLS, as the current ACLS PEA algorithm does not accommodate sub-pathways of treatment for PEA. Bedside ultrasound image quality is limited by the brief time window and ongoing CPR. Future advances in ultrasound equipment may provide more detailed information to direct care during CPR.” Bedside ultrasound may be useful in

discerning patients whose myocardial contractility remains severely impaired but intact from those cases with only agonal activity that will be addressed only by future therapeutic advances once the underlying pathophysiology has been identified and reversed. We speculate that patients with organized cardiac activity are physiologically different from patients with agonal cardiac activity. The first group may represent patients who are so profoundly hypotensive that peripheral pulses are not palpable, but the basic physiology of the cardiac contraction remains intact. Alternatively, patients with agonal cardiac twitching may represent a population of patients with a fundamentally different cellular physiology who respond differently to pharmacologic interventions.

Our results show that differences in contractility are visible sonographically in a timeframe that allows medical decision-making in a patient in PEA cardiac arrest. There is an association between longer resuscitations and organized cardiac activity, but it is unclear if the characteristics of cardiac activity prompted increased resuscitation efforts by the health care team, or if the organized activity resulted in increased survival that in turn caused longer resuscitation. It is not clear why physicians chose the non-ACLS medication. In the case of continuous intravenous adrenergic agents it is possible that visible cardiac activity prompted the physician to start a medication that would support the visible findings on ultrasound. It is also possible that other unidentified factors such as a change in the cardiac rhythm on the monitor may have contributed to the decision. It is also unclear if initiation of any of the non-ACLS medications contributed to an increased survival, or if the findings on ultrasound identify a population more likely to recover from their initial insult regardless of medical intervention. Our data supports the assertion that patients with organized cardiac activity responded differently to both standard ACLS medications and continuous intravenous vasopressors. In this study patients with organized cardiac activity suggest an association between continuous intravenous adrenergic agents and survival, but survival was also negatively associated with use of bolus epinephrine. The administration of bolus doses of epinephrine is not without potential deleterious effects.15 The two randomized trials in humans assessing the effect of epinephrine demonstrated no survival benefit beyond ROSC16,17. In a randomized controlled trial (RCT) of epinephrine from 2011, outof-hospital cardiac arrest patients receiving bolus epinephrine demonstrate survival to hospital admission of 23.5% compared to 11% in patients receiving placebo (p=ns)17. Another RCT on epinephrine in out-of-hospital cardiac arrest from 2012 found similar results with survival to hospital admission of 22% following epinephrine16. Patients given continuous infusions of adrenergic agents in this study demonstrate survival rate to hospital admission of 45.5% if they had organized cardiac activity. This study was not designed to determine causation so future prospective studies need to be

performed. Our findings support the concept of introducing ultrasound imaging and clinical decision-making early during cardiac arrest care. It may even be possible to integrate ultrasound into the pre-hospital setting. Our results and other publications suggest that possibly even more patients could be identified for alternative therapeutic interventions in the pre-hospital arena18. Current pre-hospital training does not cover ultrasound imaging, but introducing ultrasound imaging in the this setting would support non-ACLS interventions in the environment that has the highest potential to improve outcomes. Future studies are needed to address this issue. In conclusion, patients presenting in PEA cardiac arrest demonstrate findings on bedside ultrasound that are associated with differences in how they respond to therapy. Patients with organized cardiac activity respond better to standard ACLS interventions. Patients with organized cardiac activity may respond differently to medications such as continuous intravenous vasopressors. Further prospective research on ultrasound guided interventions on patients in PEA with organized cardiac activity is needed. Acknowledgements: We would like to thank Virginia Mangolds who provided organizational support and Glenn Jones who provided statistical support for this study. Conflict of Interest Dr. Romolo Gaspari has no conflict of interest.

REFERENCES 1. Niendorff DF, Rassias AJ, Palac R, Beach ML, Costa S, Greenberg M. Rapid cardiac ultrasound of inpatients suffering PEA arrest performed by nonexpert sonographers. Resuscitation 2005;67:81-7. 2. Hernandez C, Shuler K, Hannan H, Sonyika C, Likourezos A, Marshall J. C.A.U.S.E.: Cardiac arrest ultra-sound exam--a better approach to managing patients in primary nonarrhythmogenic cardiac arrest. Resuscitation 2008;76:198-206. 3. Chardoli M, Heidari F, Rabiee H, Sharif-Alhoseini M, Shokoohi H, Rahimi-Movaghar V. Echocardiography integrated ACLS protocol versus conventional cardiopulmonary resuscitation in patients with pulseless electrical activity cardiac arrest. Chin J Traumatol 2012;15:284-7. 4. Paradis NA, Martin GB, Goetting MG, Rivers EP, Feingold M, Nowak RM. Aortic pressure during human cardiac arrest. Identification of pseudo-electromechanical dissociation. Chest 1992;101:123-8. 5. Bocka JJ, Overton DT, Hauser A. Electromechanical dissociation in human beings: An echocardiographic evaluation. Annals of emergency medicine;17:450-2. 6. Mehta C, Brady W. Pulseless electrical activity in cardiac arrest: electrocardiographic presentations and management considerations based on the electrocardiogram. The American journal of emergency medicine 2012;30:236-9. 7. Jentzer JC, Clements CM, Wright RS, White RD, Jaffe AS. Improving Survival From Cardiac Arrest: A Review of Contemporary Practice and Challenges. Annals of emergency medicine 2016;68:678-89. 8. RJ G, A W, S A, et al. Emergency department point-of-care ultrasound in out-of-hospital and in-ED cardiac arrest. Resuscitation 2016;109:33-9. 9. Jacobs I, Nadkarni V, Bahr J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update and simplification of the Utstein templates for resuscitation registries. A statement for healthcare professionals from a task force of the international liaison committee on resuscitation (American Heart Association, European Resuscitation Council, Australian Resuscitation Council, New Zealand Resuscitation Council, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa). Resuscitation 2004;63:233-49. 10. Sasson C, Rogers MA, Dahl J, Kellermann AL. Predictors of survival from out-of-hospital cardiac arrest: a systematic review and meta-analysis. Circulation Cardiovascular quality and outcomes 2010;3:63-81. 11. Cha WC, Lee EJ, Hwang SS. The duration of cardiopulmonary resuscitation in emergency departments after out-of-hospital cardiac arrest is associated with the outcome: A nationwide observational study. Resuscitation 2015. 12. Mader TJ, Nathanson BH, Millay S, et al. Out-of-hospital cardiac arrest outcomes stratified by rhythm analysis. Resuscitation 2012;83:1358-62. 13. Andersen LW, Bivens MJ, Giberson T, et al. The relationship between age and outcome in out-of-hospital cardiac arrest patients. Resuscitation 2015;94:49-54. 14. Hasselqvist-Ax I, Riva G, Herlitz J, et al. Early cardiopulmonary resuscitation in out-ofhospital cardiac arrest. The New England journal of medicine 2015;372:2307-15. 15. Callaway CW. Epinephrine for cardiac arrest. Curr Opin Cardiol 2013;28:36-42. 16. Olasveengen TM, Wik L, Sunde K, Steen PA. Outcome when adrenaline (epinephrine) was actually given vs. not given - post hoc analysis of a randomized clinical trial. Resuscitation 2012;83:327-32.

17. Jacobs IG, Finn JC, Jelinek GA, Oxer HF, Thompson PL. Effect of adrenaline on survival in out-of-hospital cardiac arrest: A randomised double-blind placebo-controlled trial. Resuscitation 2011;82:1138-43. 18. Breitkreutz R, Price S, Steiger HV, et al. Focused echocardiographic evaluation in life support and peri-resuscitation of emergency patients: a prospective trial. Resuscitation 2010;81:1527-33. Figure Caption

Figure Legend Figure 1 – Survival Following ACLS Figure Legend – Data is presented as average with 95% confidence interval or number of patients in the category. Survival data includes patients with organized activity and disorganized cardiac activity. Patients where the activity was unable to be categorized were not included in the figure. Figr-1

Table 1 – Baseline Characteristics of patients in PEA with cardiac activity on ultrasound Organized Cardiac

Disorganized Cardiac

Activity

Activity

Length of Resuscitation min (mean)

38.1 (24.6)

26.2 (18.4)

P=0.002

Age (years)

65.5 (16.7)

68.5 (14.7)

P=ns

Gender (% male)

58.1%

62.4%

P=ns

Downtime (min)

7.3 (12.4)

5.1 (9.0)

P=ns

Location of Arrest (% in ED)

17.6%

23.6%

P=ns

Bystander CPR (% present)

33.8%

40.9%

P=ns

1 mg Epinephrine doses

3.5 (2.2)

3.7 (1.8)

P=ns

Data presented as mean (standard deviation) or percent of total. P value significance was calculated at 0.05.

Table 2 - Non-ACLS medical interventions during PEA Agent

Number of patients

Calcium Chloride

57

Continuous intravenous adrenergic agents

39

Magnesium Chloride

4

Thrombolytics

6

Intralipid

1

Note – some patients received more then one non-ACLS medication