Chapter 46 Michael R. Sayre Each year, EMS professionals assess about 360,000 patients who are in cardiac arrest [1]. Resuscitation is attempted for about 60% of those assessed. Over the past decade, the median survival rate to hospital discharge for all patients with attempted resuscitation has slowly risen from about 5% to about 10%, with individual systems reporting rates between 3% and 16% [1–4]. Despite significant efforts devoted to improving survival from out-of-hospital cardiac arrest (OHCA) since 2000, a large gap remains between research on the disease and its effect on public health [5]. This chapter will review some of the issues that investigators should consider when developing OHCA clinical research. Since its introduction in 1991, the Utstein style has become the gold standard for reporting data from OHCA [6]. This system of uniform terms and definitions for resuscitation reporting, last updated in 2014, allows useful comparison of outcomes among systems of out-of-hospital care across a plethora of countries [7]. Aspects of the Utstein style are controversial. For example, there is wide variation in classification of the etiology of cases. When hospital and autopsy information is available, investigators will discover that many patients have a diverse set of causes in addition to heart disease compared to a determination based only on the EMS records [8]. A challenge is differentiation of cardiac arrest from simple syncope. Cardiac arrest is the sudden onset of lack of blood circulation leading to unresponsiveness. Although patients are commonly apneic on EMS arrival, an agonal respiratory pattern is often present early in the evolution of an arrest [9]. A cardiac arrest event persists longer than several seconds and generally does not resolve spontaneously. Cases lasting only a few seconds most likely represent syncopal episodes rather than true cardiac arrest. Although rarely the patient is revived with only a brief period of cardiopulmonary resuscitation (CPR), therapy in addition to chest compressions and ventilation is typically needed to restore spontaneous circulation [10]. From a practical point of view, the simplest approach is to include cases in which EMS provided chest compressions or a shock with a defibrillator. There is wide variation in the reported incidence of EMS-treated cardiac arrest, ranging from 48 to 70 cases per 100,000 population per year [2]. This difference has been primarily attributed to variations in case ascertainment but real differences in incidence rates likely exist among communities, such as variations in prevalence of heart disease among populations and a general decline in the incidence of ventricular fibrillation (VF) outside the hospital [11]. Those patients obviously dead at the scene should ideally be tracked even though they have no chance of successful resuscitation, in order to verify case ascertainment approaches. These include patients with rigor mortis, decapitation, or dependent lividity on the arrival of EMS responders (Table 46.1) However, these cases are not included in the denominator of survival statistical analyses. Table 46.1 Possible exclusion criteria When the Utstein style was proposed, only arrests of cardiac etiology were included to maximize comparability between studies. Determination of cardiac etiology is often accomplished retrospectively through history obtained from family members, secondary survey physical examination, hospital chart review, autopsy reports, and EMS run records. Accurate determination of cardiac etiology is unlikely in some cases with only EMS records [8]. As a result, the Resuscitation Outcomes Consortium has chosen to track all cases without obvious trauma regardless of etiology. With the declining incidence of VF, the proportion of OHCAs that are of non-cardiac etiology appears to be increasing. That may be sufficient reason to pay more attention to the care of these complex patients [12]. Some researchers point out that a patient who has a pulse following a shock by an automated external defibrillator (AED) used by a member of the public prior to EMS arrival and who does not receive any EMS CPR does not directly benefit from EMS cardiac arrest care; but most investigators include those patients. Patients with advance directives indicating their desires to avoid any resuscitative efforts should also be excluded from research studies. Often these patients may initially receive treatment only to have resuscitation preemptively terminated once EMS learns about the advanced directive. The pediatric population presents additional challenges for OHCA investigator. The incidence of cardiac arrest in this population is low; data suggest that fewer than 1,000 children experience out-of-hospital VF events per year in the entire United States [13]. Additional ethical and consent issues are raised when dealing with children younger than age 18, and accurate determination of age is not always possible during sudden cardiac arrest. The latest edition of the Utstein reporting guidelines includes a section on pediatric cardiac arrest [7]. The reader’s ability to compare results across different studies relies upon an accurate description of the patients included in the study, such as their age, sex, geographic locale, and EMS response. Factors such as previous cardiac disease, other comorbidities, tobacco use, and family history may also be useful, although they may be more difficult to obtain [14]. A standard approach to describing patients who were included and excluded can often by achieved by using the Consolidated Standards of Reporting Trials (CONSORT) patient flow diagram [15]. A description of the location (e.g. home or EMS vehicle) and witness status of the arrest should be included (Table 46.2). The description of the EMS system in which the study was conducted and the associated population density (rural, suburban, or urban) are relevant because these characteristics may affect the results of the study and will assist the reader with determining its generalizability. For example, a study of a device for supporting circulation likely will be doomed to failure if the trained responders cannot be at the patient’s side until 25 minutes after the patient’s collapse [16]. Ideally, the number of potential first responders and paramedics should be reported so that the frequency of exposure of individual EMS professionals to cardiac arrest events can be estimated [17]. Table 46.2 Arrest characteristics
Cardiac arrest-related research methodology
Introduction
Consistent definitions
Exclusion criteria
Etiology independent
Etiology related
Rigor mortis
Traumatic cause
Decapitation
Drowning
Dependent lividity
Accidental hypothermia
Do not resuscitate order
Asphyxia
Age <18 years
Toxic ingestion or overdose
Prisoner
Electrocution
Pregnant
Sepsis
Population description
Location
Witness
Home/residence
Spouse
Extended care facility
Family member
Medical/dental clinic
Bystander
Public/commercial building
Off-duty medical personnel
Workplace
On-duty fire/police personnel
Educational institution
EMS
Outdoors
Data collection and reporting