Key Clinical Questions
What are the fundamental goals of cardiopulmonary resuscitation?
Which components of resuscitation are considered vital to success?
How can the common pitfalls of resuscitation be surmounted?
What treatments should be instituted immediately upon successful resuscitation?
How should outcomes in resuscitation shape the discussion of advanced directives?
Cardiopulmonary resuscitation is a time-dependent, team-based effort to reverse physiologic events that may culminate in a patient’s imminent death. Biblical and ancient Egyptian hieroglyphic texts allude to mouth-to-mouth ventilation in divine contexts, but other texts indicate Jewish midwives used mouth-to-mouth resuscitation as early as 3300 years ago to revive stillborn children.
In the United States an estimated 375,000 to 750,000 hospitalized patients suffer in-hospital cardiac arrest (IHCA) requiring advanced cardiac life support (ACLS) annually. The incidence of IHCA is estimated to be as high as 1% to2% of all patients admitted to academic hospitals with a prevalence of approximately 65 people per 100,000 nationally.
In-hospital cardiac arrest encompasses a spectrum of disorders from insufficient cardiac output to generate appreciable cerebral perfusion such as arrhythmia or shock to complete cessation of cardiac activity. Vital sign anomalies may often herald impending inpatient cardiac arrest by minutes to hours, but many cardiac arrests occur suddenly and without warning. Acute pulmonary arrest (very common in pediatric populations, often due to airway obstruction; but much less common in adults) may precede IHCA and may occur from sedative or opiate analgesic overdose.
This chapter focuses on (1) the techniques that are essential to successful cardiopulmonary resuscitation especially with attention to good neurologic recovery (as defined by the cerebral performance category of zero or one), and (2) decision making based on patient resuscitation status.
Since standardization of closed chest cardiac massage (CCCM)—that is, chest compressions—was first described systematically in the medical literature in 1960, CCCM has remained the only reliable means of reviving a patient in cardiopulmonary collapse. It is an effective and powerful intervention that, when unnecessarily delayed, may lead to poor patient outcomes. In one study, survival dropped from 34% to 14% if CCCM was delayed even as little as 60 to120 seconds from the time the patient collapsed. Therefore, clinicians must recognize and respond to cardiac arrest immediately for resuscitation measures to be effective.
Advanced cardiac life support combines basic life support (BLS) measures with specific interventions, such as medication, defibrillation, transthoracic pacing, and advanced airway management.
While often considered adequate for institution credentialing purposes, completion of American Heart Association (AHA) courses fails to result in long-term meaningful skill performance. Health care providers’ capabilities to demonstrate appropriate technique for CCCM and capabilities to successfully navigate the steps of cardiopulmonary resuscitation begin to degrade just weeks following course completion. Therefore, for the whole medical team to respond concisely and in a coordinated fashion, clinicians must have extensive medical knowledge, training, drilling practice, continued education, and feedback.
Many providers are reluctant to initiate CCCM without complete assurance that the patient is truly in cardiopulmonary arrest (confirmed by vital signs or electrocardiographic rhythm), often leading to unnecessary delays in initiation of potentially lifesaving treatment. Furthermore, fundamental pulse assessment, even in nonemergency situations, cannot reliably and accurately predict the presence or absence of a pulse. One study tasked providers to determine whether or not patients had palpable pulses during elective cardiopulmonary bypass surgery. Ultimately providers took around 20 seconds to assess the pulse and were less than 70% accurate.
Time spent gathering cardiac monitoring, attaching leads, and setting up equipment can further delay promptly needed interventions to prevent death. In fact, clinicians may need to initiate CCCM prior to confirming cardiopulmonary arrest. Prompt initiation of CCCM for any patient who appears to be in extremis (ie, unarousable or clinically unstable with suspicion of cardiopulmonary arrest) should occur until confirmatory evaluation, often by a multispecialty resuscitation team, offers a high degree of confidence that CCCM can be discontinued. Providers should share a culture of support that accentuates that the greater harm to patients is in failing to initiate CCCM in contrast to the potential harms of CCCM (rib fracture, pneumothorax, organ perforation).
Cardiopulmonary arrest heralds death and may be an expected outcome in many hospitalized patients. However, rarely is cardiopulmonary arrest the first manifestation of physiologic events that ultimately culminate in collapse: patients frequently have alteration in mental status or significant vital sign changes (pyrexia, hypotension, bradycardia, decrease in oxygen saturation, change in respiratory rate), often hours before developing cardiac arrest. Intervention during this prearrest period may prevent cardiac arrest altogether. Alternatively, health care personnel may identify patients who are at the end of life and may thus benefit from a meaningful discussion about limiting resuscitative measures, including offering “Do Not Resuscitate” or “Allow Natural Death” orders. Many patients are not well informed about the resuscitative process and may have inflated images of routine successful resuscitation shaped from popular culture embodied by television and film. Clinicians often perform cardiopulmonary resuscitation on patients without informed consent—a discussion of the relevant risks, benefits, and alternatives to therapy along with the clinicians’ recommendations. Thus the prearrest period may offer an unparalleled opportunity to give patients an active role in deciding whether resuscitation is desired (see Chapter 215 [Communication Skills for End of Life Care]).
While no one specific condition results in cardiopulmonary collapse, many health-care-associated interventions predispose patients to arrest and often require minimal intervention early on to alter the course of catastrophe (Table 137-1). Intervention during impending cardiac arrest requires a detailed history of recent interventions ranging from invasive procedures to recent sedation or anesthesia.
TABLE 137-1Interventions to Specific Conditions that may Prevent Evolution to Cardiopulmonary Arrest in Hospitalized Patients ||Download (.pdf) TABLE 137-1 Interventions to Specific Conditions that may Prevent Evolution to Cardiopulmonary Arrest in Hospitalized Patients
|Cause ||Intervention |
|Hypoxia due to medication or anesthesia ||Supportive oxygen, reversal agents (naloxone for opiates, flumazenil for benzodiazepines) |
|Acidemia due to hypercapnic respiratory failure from medication or obstructive sleep apnea ||Ventilation support (noninvasive or mechanical ventilation) |
|Pulmonary embolism ||Appropriate VTE prophylaxis (pharmacologic unless significant contraindication); high index of suspicion and timely treatment |
|Cardiac arrhythmia due to acute coronary syndrome ||Appropriate early intervention including antiplatelet therapy, beta-blockers, anticoagulation and early percutaneous coronary intervention (PCI) |
|Hyperkalemia ||Calcium, sodium bicarbonate, insulin with dextrose, consideration of early hemodialysis; check for acid-base derangements |
|Hypokalemia ||Correction of magnesium (first) followed by potassium; check for acid-base derangements |
|QT prolongation ||Attention to medications known to prolong the QT interval (such as fluoroquinolones) and consideration of cardiac monitoring |
|Hypotension from severe sepsis ||Early massive volume resuscitation with consideration of inotropes |
|Anticipated end-of-life care ||Discussion of appropriate “Do Not Resuscitate” or “Allow Natural Death” orders and palliative care in appropriate patients |
Responses to inpatient emergencies require multiple individuals who take on specific roles and integrate as a team. For care to function effectively and seamlessly during health care emergencies, each clinician must assume a narrowly focused essential function or task (such as assessing a patient’s airway, recording data in a flowsheet, or ensuring chest compressions are adequate) and perform the task with high quality to facilitate the best possible patient outcome engendered by the team as a whole.
RAPID RESPONSE TEAMS AND THE PREARREST PERIOD
Recognizing that early intervention in impending cardiopulmonary arrest may prevent the arrest altogether, many hospitals have implemented rapid response teams (RRTs), consisting of any combination of critical care nurses, respiratory therapists, pharmacists, and/or physicians to attend to patients who exhibit one or more parameters of clinical instability but are not yet in extremis. Rapid response teams facilitate earlier communication with and transfer of care to intensive care units under the care of critical care teams and (when available) intensivists, which appear to reduce mortality in some centers, and need for crisis activation of cardiopulmonary arrest teams (ie, code teams). Additionally, RRTs have seen a marked expansion and elaboration in many disease states, such as improved identification of patients with sepsis and rapid implementation of early goal-directed therapy in patients with sepsis; and stroke teams in patients with neurological crises. Consistently, RRTs prompt discussion with patients and families about advanced directives (do not resuscitate orders or limitations of care), and reduce escalation of care in patients who do not desire such aggressive interventions and when the medical condition is expected to be immediately terminal.
Hospitalists must foster a culture of safety where any provider (or patient or family member) may initiate an RRT for any reason without fear of reprisal or judgment. Hospitalists should always thank other clinicians for calling RRTs and keeping the patients’ safety of the utmost concern.
Respiratory arrest from medications (anesthesia, benzodiazepines, or opiates) may lead to cardiac arrest through hypoxia and changes in the pH due to combined metabolic and respiratory acidosis. Respiratory arrest is often masked for some time due to the ubiquity of oxygen administration in hospitalized patients, which may lead to a prolonged period of hemoglobin oxygenation while ventilation may have already decreased or stopped. Overreliance on pulse oximetry as a sole source of interpreting ventilation effort may delay response to respiratory arrest until the patient is hypoxic and has developed profound acidemia. Systemic hypoxia causes pulmonary artery constriction, right ventricular failure, and systemic hypotension from poor right heart output coupled with loss of vascular tone from hypoxia (circulatory shock).
Cardiac arrest may occur from multiple distinct mechanisms. True cardiac arrest (cardiac standstill) occurs either as a primary mechanism (from arrhythmias like ventricular fibrillation that prevent normal cardiac function) or as a secondary mechanism (from asystole or from an extended period of failed resuscitation and cardiac myocyte death). Most cardiopulmonary arrest episodes do not occur due to true cardiac standstill but rather from marked impairment in cardiac output resulting in systemic arterial hypotension, tissue hypoxia, and organ failure. Precardiac, intracardiac, or postcardiac mechanisms may independently or in combination result in cardiopulmonary arrest (Table 137-2).
TABLE 137-2Cardiac Arrest Etiology by Anatomic Location ||Download (.pdf) TABLE 137-2 Cardiac Arrest Etiology by Anatomic Location
|Precardiac ||Intracardiac/intrapulmonary ||Postcardiac |
Shock (septic, distributive)
Cardiac arrhythmia (ventricular or atrial)
Left ventricular rupture
SUBTYPES OF CARDIAC ARREST
Once appropriate resuscitation equipment has arrived, clinicians should immediately begin to differentiate whether the cardiac arrest is due to a “shockable” or “nonshockable” cardiac rhythm.
Transthoracic electrical shocks can terminate some pathological cardiac rhythms that inhibit normal cardiac function. These can include ventricular fibrillation, ventricular tachycardia, AV nodal reentrant tachycardia, atrial fibrillation, and atrial flutter. While ventricular fibrillation has a very characteristic pattern, the other rhythms may be difficult to differentiate during an emergency and in the absence of 12-lead electrocardiography. In the setting of an unconscious patient in severe distress, who is obtunded or clinically severely unstable, all of these rhythms are considered pathologic and warrant immediate electrical shock.
Despite recommendations by the International Liaison Committee of Resuscitation (ILCOR) (the subsection of the American Heart Association responsible for publication of the ACLS guidelines) that differentiation of the exact cardiac arrhythmia may dictate very different types of cardiac intervention, ranging from dose (in joules) of electrical therapy to medication selection, confirming an exact rhythm diagnosis may not be practical. Thus, it is reasonable to treat all of these rhythms similarly in a cardiopulmonary arrest in the event of clinical uncertainty. Fundamentally similar to administration of CCCM, delays in electrical therapy may significantly negatively impact patient outcomes with even minimal delays. If a patient is not critically ill, then time allows for conscientious assessment of cardiac rhythm via 12-lead electrocardiogram (ECG) with appropriately targeted therapies for the underlying arrhythmia. (see Chapter 132 [Supraventricular Tachyarrhythmias] and Chapter 124 [Ventricular Arrythmias]).
Precise differentiation between ventricular fibrillation, ventricular tachycardia, AV nodal reentrant tachycardia, atrial fibrillation, and atrial flutter may not be practical when a patient is in severe distress, obtunded, or clinically severely unstable. Thus in the event of clinical uncertainty it is reasonable to treat all of these rhythms similarly during a cardiopulmonary arrest.
Ventricular fibrillation results from disorganized myocardial electrical activity, and the heart is unable to generate a contraction to produce cardiac output. Hospitalists should be able to identify ventricular fibrillation confidently on rhythm strip (Figure 137-1).
Rhythm strip of a patient with ventricular fibrillation.
The characteristic physiologic phases of ventricular fibrillation arrest underscore the importance of rapid electrical therapy. During the first few minutes of ventricular fibrillation (reflecting the combination of the “acute” and “electrical” phases of arrest, lasting up to 5-6 minutes), the myocardium is highly responsive to counter shock. This explains in part why successful defibrillation is so common on commercial airlines and in casinos where employees are trained to rapidly attach and initiate automated external defibrillators (AEDs). The acute and electrical phases can be extended when CCCM is initiated promptly, thus underscoring how critical CCCM is as an immediate therapy while definitive defibrillation equipment is located, attached, and initiated.
In the absence of CCCM, patients will degenerate into the “circulatory” phase where electrical therapies are less effective due to progressive tissue hypoxia and myocyte death. During this phase, CCCM may need to be performed for several minutes antecedent to successful defibrillation. However, during the initial moments of a pulseless arrest, immediate rhythm identification and defibrillation of shockable rhythm takes precedence over CCCM.
Unchecked, patients will eventually enter the “metabolic” phase of ventricular fibrillation starting around the tenth minute of cardiac arrest. In the absence of effective CCCM, irreversible brain damage occurs. While there remains a slim hope of successful cardiac resuscitation at this point, survival to hospital discharge rapidly becomes improbable.
Ventricular tachycardia resulting in cardiopulmonary arrest fundamentally is identical to ventricular fibrillation in treatment: CCCM and early electrical shock are indicated.
Perhaps the most overwhelming change to resuscitation in recent years is the acknowledgment of severe ventricular stunning following electrical shock. For several minutes following defibrillation—and extending for a variable duration thereafter—the heart is mechanically dysfunctional and unable to generate an adequate cardiac output for organ perfusion or brain function. Consequently, it is absolutely critical to reinitiate CCCM for 1 to 2 minutes after defibrillation whether or not the shock is successful at aborting the ventricular arrhythmia.
Nonshockable unstable or pulseless rhythms (characterized by bradycardia, asystole, and pulseless electrical activity [PEA]) constitute the majority of inpatient cardiac arrests. Deterioration in clinical status signified by deviations in mental status or marked changes in vital signs often foreshadows these types of cardiopulmonary arrests, and they may be preventable.
Bradycardia may be due to a primary arrhythmia (such as sick sinus syndrome or arterioventricular [AV] block), or may be due to a secondary cause such as medications (particular AV nodal blocking agents) or excessive vagal tone (due pain or nausea). Bradycardia severe enough to cause hemodynamic instability warrants immediate correction and treatment of the underlying cause. Atropine is a vagolytic that can potently reverse excessive vagally mediated bradycardia. However, with unpredictable effects and a narrow therapeutic window (too high or too low a dose of atropine can potentially paradoxically worsen bradycardia), its use is confined to select patients and only for short-term use. Chronotropic agents such as dopamine can be administered if time allows setup of an intravenous drip.
Bradycardia may respond to transcutaneous pacing, but this must be instituted rapidly. In conscious bradycardic patients transcutaneous pacing may prove to be exceptionally uncomfortable but should be used to bridge to transvenous pacing. Patients may require analgesia or sedation during transcutaneous pacing while awake.
Asystole as a primary cause of cardiac arrest is uncommon. Asystole typically is the end result of another pathophysiologic process, such as sustained hypoxia or coronary thrombosis. As such, asystole is a fairly late finding. Fine ventricular fibrillation may appear electrocardiographically similar to asystole. Clinicians should always confirm suspected asystole by checking multiple defibrillator leads and increasing the electrical gain. Doing such should clarify if the rhythm is actually asystole (vs masked fine ventricular fibrillation). Whereas defibrillation is likely to benefit a patient in ventricular fibrillation (and is necessary to terminate the rhythm), shocking a patient in asystole may result in depleting the heart of any remaining adenosine triphosphate (ATP) and with it any chance of successful resuscitation. In general, if clinicians are not certain whether asystole or ventricular fibrillation is the underlying rhythm, defibrillation is favored due to the overwhelming benefit patients with ventricular fibrillation receive from defibrillation compared to the minimal excess risk posed to those already in asystole.
Pulseless electrical activity represents a complex spectrum of disorders where patients appear to have an electrocardiographic rhythm that would be anticipated sufficient to generate a cardiac output, but clinical examination reveals no evidence of a palpable pulse. By definition, PEA is not a rhythm derangement, and therefore will not respond to any form of electrical shock. Pulseless electrical activity is a problem with either too little cardiac preload (vasodilation, pulmonary embolism, or profound volume depletion), ineffective cardiac output (due to cardiac failure or stunning), or extrinsic compression of the heart muscle (tension pneumothorax, severe airway obstruction, or pericardial tamponade). While seeking a cause, clinicians must pursue concomitant treatment with CCCM in spite of the apparently normal-appearing cardiac rhythm. Clinicians should use epinephrine and intravenous fluids along with the goal to furnish targeted treatment of the apparent cause of PEA (intubation for hypoxia or respiratory distress, needle decompression or chest tube insertion for tension pneumothorax, pericardiocentesis for tamponade, intravenous calcium for hyperkalemia, etc).
While the approach to cardiac arrest has changed considerably over the past 5 decades, survival has improved little since initial reports on CCCM in 1960. ILCOR has published basic and advanced cardiac life support guidelines every 5 years, becoming the de facto standard of care in the United States and internationally. Despite their evidence base, criticism exists that many find the guidelines to be too complex and difficult to remember even just weeks following life support course completion. Also discordantly, the single most effective stratagem in resuscitation—effective chest compressions—frequently is not taught well or performed well during or following courses, with a time-dependent loss of skill following course completion.
While many clinicians learned that resuscitation begins with the “A-B-Cs,” evidence now suggests that establishing an airway and initiating rescue breathing (accomplished in most hospitals via bag-valve-mask [BVM] ventilations) are not nearly as important as CCCM during the early phases of most adult inpatient cardiac arrests (Figure 137-2). Guidelines therefore now recommend focusing on “C-A-Bs,” emphasizing that restoring circulation with compressions and early defibrillation are of critical importance. Certainly, in primary respiratory arrest (such as from medications) and in children (in whom respiratory arrest is much more common than cardiac arrest), management of airway and respiration must occur rapidly (Figure 137-2).
Flow diagram of assessment and treatment during cardiac or pulmonary arrest.
CHEST COMPRESSIONS (CLOSED CHEST CARDIAC MASSAGE)
When an apparently unconscious patient cannot be aroused, clinicians should assume that the patient might be in cardiopulmonary arrest and should institute chest compressions without delay. The mantra for quality chest compressions is to push hard, pump fast, and allow good chest recoil.
Many providers are concerned with pushing down on the sternum too hard; however, evidence does not support the assertion that pushing too hard occurs. Current recommendations suggest a compression depth of at least 2 inches, but that measurement is very difficult to extrapolate in clinical terms during resuscitation. The best advice is for rescuers to push as deeply as possible with arms locked and with the rescuer’s shoulders directly over the patient’s sternum with the rescuer using his or her body weight and waist flexion to deliver the compression.
Because most patients go into cardiopulmonary arrest in hospital beds, achieving proper positioning may be difficult, particularly when a patient is obese or the rescuer’s arm length is short; a stool or other lift may prove critical for proper hand positioning. Compressing the chest solely with the force of the rescuer’s arms may be highly kinetic in appearance but will offer virtually no benefit to the patient.
Compressions must be done on a hard surface, something a hospital mattress does not offer. As soon as possible, a backboard should be inserted behind the patient. Real-time feedback devices, such as accelerometers, may offer the best opportunity for ensuring adequate compression depth; however, these devices have technological limitations. Frequently the devices interpret total patient motion as compression depth when in fact a substantial amount of the compression is expended compressing the mattress and not the patient’s chest. Compression depth indirectly correlates with patient survival.
Since ideal chest compressions result in only one-third normal cardiac output, about 10% of normal cerebral blood flow, and <5% of normal cardiac blood flow, compression rate has a substantial effect on tissue perfusion. Target compression rate of at least 100 compressions per minute should be instituted, but interruptions in chest compressions during change of rescuers, intubation, or rhythm analysis all result in a markedly lower total number of compressions over time. Studies consistently show that compressions are almost uniformly lower than 100 per minute, underscoring the need for practice, simulation, and feedback for all rescuers on a regular basis following completion of chest compression training.
Good compressions require a high degree of physical ability, but frequent rescuer rotation must be balanced with the need for uninterrupted compressions. While automated solutions (such as mechanical compression devices) may eventually replace most rescuer-performed compressions in inpatients, current devices are unwieldy or have failed to show noninferiority to manual chest compressions.
Some aspects of resuscitation are incompatible with ongoing chest compressions (such as rhythm analysis and, at times, intubation). In these situations, rescuers must limit the duration of the interruption as return of spontaneous circulation and neurologically intact survival are directly tied to chest compression rate. Furthermore, current evidence suggests it is essential to resume chest compressions immediately following a defibrillation attempt independent of the rhythm due to left ventricular stunning associated with highly impaired cardiac output.
Lastly, the beneficial effects of chest compressions are lost within seconds of discontinuation. Chest compressions’ benefits appear to be additive to one another over time with subsequent compressions improving circulation and perfusion pressures; if compressions are widely spaced or stopped for any length of time, the benefits of the previous compressions’ vascular effects are lost. Defibrillation success also depends on short “hands off” intervals (time when no compressions occur) as the chances of successful defibrillation diminish within seconds. This latter finding suggests that there is very little latitude for prolonged rhythm analysis during a cardiac arrest.
At the completion of a chest compression, rescuers must extend at the waist allowing the patient’s chest to rise back to its rest position. The mechanisms by which compressions exert their physiologic effect appear to be a combination of increased intrathoracic pressure leading to compression of the great vessels and direct pumping of the heart through reduction of the anterior-posterior diameter of the chest. The recoil phase is effectively “diastole,” and incomplete recoil of the chest thus results in impaired blood return to the great vessels and heart resulting in further impairment in circulation in an already desperate perfusion environment. If recoil is consistently poor, rescuers will be incapable of surmounting this critical phase of circulation with adequate or consistent depth of compressions. Compression rate much more than 140 per minute will reduce effective recoil and return of spontaneous circulation.
Defibrillation is used to depolarize all myocytes simultaneously in order to achieve a uniform repolarization period whereby the sinoatrial node theoretically resumes the pacemaking role of the heart, thus restoring normal cardiac function. The standard dose for defibrillation is 120 to 200 J for biphasic defibrillators (and 360 J for monophasic defibrillators).
Caregivers will be unlikely to accurately determine the workings of a defibrillator in a crisis without copious practice beforehand. Automated external defibrillators are ubiquitous even in nonhospital settings, but even their setup may prove to be puzzling during a crisis if providers have not practiced using them.
Most manual defibrillators have self-adhesive defibrillator pads that are applied to the sternum and back (or alternately to the sternum and left midaxillary line, about the location of the cardiac apex) to deliver shocks. The pads concomitantly offer a “quick look” mode, displaying the patient’s cardiac rhythm independent of cardiac leads. This means rescuers can simultaneously identify a patient’s cardiac rhythm while charging the pads, and then deliver a shock.
Automated external defibrillators utilize self-adhesive pads as well but require a period of up to 20 seconds for computer rhythm analysis, during which time rescuers are not performing CCCM. All hospitalized patients in cardiac arrest should have self-adhesive defibrillator pads attached as these offer continuous monitoring and allow rescuers to deliver shocks or to pace the patient. Newer models also offer real-time feedback to compression depth, recoil quality, and compression rate.
Most defibrillators in clinical use employ a biphasic waveform, in which polarity reverses during the shock. Biphasic devices appear to confer better neurological outcomes. Defibrillators can deliver shocks that are synchronized to patients’ cardiac rhythms to prevent administration of a shock during the critical repolarization phase represented by the T-wave (resulting in the “R on T” phenomenon and concomitant risk for precipitation refractory arrhythmias). Since ventricular fibrillation is a disorganized cardiac rhythm where no discrete T-waves exist, defibrillators set to “synchronization mode” may not delineate a safe period to deliver a shock and therefore may not fire at all. Most defibrillators will not have synchronization enabled when turned on without a clinician specifically enabling it. Nevertheless, clinicians must have good familiarity with the location of the synchronization button along with all functions on their hospitals’ defibrillators so that when a defibrillator fails to fire, clinicians know where and how to deactivate synchronization prior to another attempt at defibrillation.
Rescuers should attempt to defibrillate a patient as soon as the code cart arrives and the defibrillator is fully set up and ready to deliver a shock. Compressions should continue unabated until the defibrillator is fully prepared, otherwise unnecessary hands-off intervals will result in poorer patient outcomes. Early defibrillation is critical as soon as a shockable rhythm is diagnosed or suspected clinically since the window for successful defibrillation decreases as patients progress from the electrical phase of ventricular fibrillation into the circulatory and metabolic phases. Chest compressions extend this window for a limited period of time, maintaining patients’ responses to defibrillation.
Positive pressure ventilation via BVM provides oxygenation and ventilation. The design of the BVM is such that a one-handed squeeze provides the appropriate tidal volume for most adults in cardiopulmonary collapse: roughly 750 mL. Often, however, rescuers will use two hands to squeeze the bag, resulting in larger tidal volumes that may exceed 1000 mL. Data suggest that rescuers often deliver BVM ventilations at rates well beyond the recommended 1 ventilation every 5 seconds, with at least one report where ventilation rate exceeded the chest compression rate.
The ideal technique for bag-valve-mask use involves three hands: two to properly seal the mask over the patient’s mouth and nose while tilting the head back, and one hand (from a second rescuer) to squeeze the bag. The BVM is designed for a one-handed squeeze, which provides the appropriate tidal volume (750 mL) for most adults in cardiopulmonary collapse. Two-hand BVM squeeze may lead to hyperventilation and auto-positive end-expiratory pressure. The rate of BVM ventilations should be 1 ventilation every 5 seconds.
Well-intentioned rescuers often believe that hyperventilation will result in improved oxygenation, improved carbon dioxide levels, and improved acid-base balance, but in fact hyperventilation sets off a cascade of pathophysiologic changes that culminate in very high intrathoracic pressures, decreased cerebral and coronary circulation, and decreased survival. Since tissue perfusion (and by convention intact circulation) is a prerequisite to oxygenation, hyperventilating a patient without attending to proper chest compressions fails to bolster tissue oxygen levels. While carbon dioxide levels rise rapidly in circulatory collapse, carbon dioxide can only be off-gassed via ventilation if venous blood flow is able to enter the thorax and thereby the lungs. The single most effective treatment for the combined respiratory and metabolic acidosis uniform in cardiopulmonary arrest is resumption of physiologically normal circulation. Therefore, the proper route to achieve rescuers’ intents is via high-quality chest compressions with supplemental oxygen administered via BVM ventilations (if rescuers are competent in its use) or via passive “blow-by” oxygen administered from a nonrebreather mask (if rescuers’ skills are in doubt).
During cardiopulmonary arrest, the upper airway musculature may become lax, resulting in the tongue and jaw occluding the airway. Proper head positioning during ventilations will help decrease the risk of airway obstruction, but frequently other measures are required. Placement of either an oropharyngeal airway (in unconscious patients) or nasopharyngeal airway (in conscious patients) requires little training and may help stabilize the upper airway sufficiently to provide effective BVM ventilations. Nevertheless, rescuers may need to obtain an advanced airway (via laryngeal mask, endotracheal tube, or tracheostomy) in order to properly ventilate the patient. Only medical personnel with considerable training and experience in advanced airways should attempt to place an invasive airway. Experience is necessary in part to limit the duration of time needed to achieve the airway (during which CCCM frequently is halted) and to minimize the risk of complications (laryngeal spasm, esophageal intubation, induction of vomiting).
Almost all patients who suffer cardiopulmonary arrest will require an invasive airway to allow for ventilation in the postresuscitation period, but controversy exists as to the ideal timing to attempt airway stabilization. The airway is the key focus in patients in pure respiratory arrest, but increasingly it appears that CCCM and initial defibrillation attempts should supersede airway priority in patients in cardiac arrest. In fact, recently ILCOR removed ventilation as a component of layperson adult resuscitation (out-of-hospital cardiac arrest) to further emphasize the importance of chest compressions as the single most important intervention during cardiopulmonary collapse. Further data suggest that a delayed approach to intubation (waiting >5 minutes into a cardiac arrest) results in equivalent survival to hospital discharge as earlier intubation.
Vasoactive medications have traditionally played a large role in treatment of cardiac arrest care. Two vasoactive medications, epinephrine and vasopressin, are guideline recommended for use during cardiac arrest. Epinephrine is a sympathomimetic agent with beta-1, beta-2, and alpha-1 agonist properties. Beta-1 stimulation increases heart rate (chronotropy) and also increases contractility (inotropy), allowing for increased cardiac output secondary to higher heart rate and stroke volume. Beta-2 stimulation causes arterial vasodilation, yet its effect is mostly countered by the alpha-1 stimulation, which causes arterial vasoconstriction. Epinephrine continues to be the primary adrenergic agent used during resuscitation care. The alpha-adrenergic stimulating properties of epinephrine have been found to increase coronary perfusion pressure, but its beta-adrenergic effects are potentially detrimental, having been found to increase postresuscitation myocardial dysfunction and possibly worsen reperfusion injury. High-dose epinephrine (1 mg followed by 3 mg and then 5 mg of epinephrine depending on response) is definitely injurious, resulting in higher rates of successful cardiac resuscitation but disproportionately higher rates of poor neurologic outcomes.
One milligram of epinephrine should be administered intravenously after the rescuers deliver the first defibrillation and then every 3 to 5 minutes thereafter. Interruption of chest compressions to place a central line is not acceptable, especially if intraosseous (IO) access is available or the femoral vein is accessible. If no intraosseous or intravenous access is available, double-dose epinephrine may be administered down an endotracheal tube. Other medications that may be administered via endotracheal tube if indicated include naloxone, atropine, vasopressin, epinephrine, and lidocaine (NAVEL mneumonic).
Forty units of intravenous vasopressin may be used instead of epinephrine initially as a one-time dose. Clinical studies on vasopressin generally fail to demonstrate significantly improved outcomes over epinephrine, however. Neither vasopressin nor epinephrine have demonstrated improved neurological outcomes or survival to hospital discharge.
Atropine is an anticholinergic medication that increases heart rate via blockade of acetylcholine-mediated bradycardia. Atropine use should be confined to symptomatic bradycardia, as this medication has no role in other forms of cardiopulmonary arrest. Atropine has a narrow therapeutic window requiring at least 0.5 mg per dose, and not exceeding 3 mg cumulative dose, and may paradoxically worsen bradycardia. The 2010 ACLS guidelines recommend use of antiarrhythmic therapy for patients suffering cardiac arrest due to ventricular fibrillation or pulseless ventricular tachycardia. Amiodarone (300 mg delivered as an intravenous bolus) has proven to be superior to lidocaine based largely on out-of-hospital data showing improved rates of return of spontaneous circulation. No studies demonstrate that neurologic outcomes or survival to hospital discharge are better in patients receiving antiarrhythmic therapy, however.
Other medications, including sodium bicarbonate, calcium, and magnesium, have very specific roles in resuscitation, but their use is beyond the scope of this chapter. None of these medications is indicated routinely, and all have been associated with worse overall outcomes in nonselected patients in cardiopulmonary arrest.
Recent data suggest that a structured combination of intravenous vasopressin plus steroid (methylprednisolone) plus epinephrine (VSE therapy) improves good neurologic outcomes (Cerebral Performance Category 0-1) and is particularly effective at reducing postresuscitation shock. Wide scale adoption of VSE therapy awaits further clinical trials or guideline recommendation, but potentially improves the possibility of robust neurological survival postarrest (number needed to treat for good neurological outcome = 11).
RECOMMENDED APPROACH TO THE PATIENT IN CARDIOPULMONARY ARREST
The patient in apparent cardiopulmonary arrest requires prompt initiation of CCCM with very close attention to proper rate, depth, and recoil of compressions. Concomitantly the patient should be placed on a nonrebreather at >15 L/min of oxygen until adequate personnel are available to initiate stabilization of the airway.
Electrical shock serves as the penultimate goal in patients suffering from a shockable arrhythmia and in those in whom certainty of the underlying rhythm is not possible. While initial studies looking at further reducing the hands-off period of compressions by continuing CCCM during defibrillation appeared to suggest the potential safety of this approach, some data suggests possible risk to providers touching patients during defibrillation. Thus, the optimal strategy to minimize the hands-off interval is to charge the defibrillator while compressions are in progress, and then perform rhythm analysis with a countershock if indicated, prior to promptly resuming compressions.
Rescuers should attempt an initial “quick look” using manual defibrillator pads after charging the pads (120-150 J with a biphasic defibrillator or 360 J with a monophasic defibrillator). If indicated, personnel should deliver the shock and then resume resuscitation efforts.
Closed chest cardiac massage should continue with minimal interruption (5-15 seconds) for rhythm analysis and shock delivery. Following the shock, CCCM should resume immediately without further rhythm analysis until at least an additional 150 compressions are delivered over the next 60 to 90 seconds due to the possibility of postshock ventricular stunning. After the first shock is delivered, rescuers need to embark on securing intravenous access and managing the patient’s airway and breathing.
Initial airway maneuvers should focus solely on oxygen delivery and alleviating any upper airway flow resistance with temporary placement of an oropharyngeal or nasopharyngeal airway. If personnel with adequate airway training, practice, and experience respond to the cardiopulmonary arrest, BVM ventilations can be initiated at a rate of 5 ventilations per minute. An invasive airway should not be placed until the patient has received adequate chest compressions and an initial shock.
Once the rhythm is reassessed following 2 minutes of CCCM, if the patient remains in a shockable rhythm, a second shock is administered (at higher dose if appropriate) and epinephrine 1 mg or vasopressin 40 IU should be administered intravenously followed by another 2 minutes of compressions to allow for adequate circulation of the medication. The rhythm is then reassessed and another shock (at higher dose if appropriate) delivered if the arrhythmia is still present once again with a focus on resumption of chest compressions as soon as the shock is completed.
Clinicians should administer amiodarone 300 mg as an intravenous bolus if the arrhythmia continues followed by an additional shock 1 minute later. By this point in the cardiopulmonary collapse, clinicians with specialty training in hospital medicine, intensive care medicine, or emergency medicine should determine the next steps, if any, at ongoing resuscitation attempts.
In the case of a nonshockable rhythm, the defibrillation attempts and antiarrhythmic therapy would not be warranted, but otherwise the resuscitation should proceed in a similar manner.
In all cases of resuscitation, clinicians should attempt to determine the underlying cause while avoiding the pitfall of inaction (Table 137-3).
TABLE 137-3Pitfalls in Cardiopulmonary Resuscitation and High-Impact Countermeasures to Abrogate Them ||Download (.pdf) TABLE 137-3 Pitfalls in Cardiopulmonary Resuscitation and High-Impact Countermeasures to Abrogate Them
|Pitfall ||Cause ||Countermeasure |
| || |
Concern about potentially causing harm (rib fractures, organ perforation, pain, etc)
Uncertainty about whether the patient is truly in cardiopulmonary arrest
| || || |
| || || |
| || |
Concern about causing torsades-de-pointes or ventricular fibrillation
Concern about misidentification of a benign rhythm for a malignant one
Lack of familiarity with equipment and its use
Transthoracic shocks are the only intervention, other than CCCM, that have meaningfully valuable impacts on patient survival, but survival begins to decrease every second a patient remains in a shockable rhythm. Early shocks lead to improved outcomes
Torsades-de-pointes and ventricular fibrillation are themselves shockable rhythms
Recognize that correct rhythm diagnosis may not be possible early on in a resuscitation
Recurrent training, practice, simulation, and feedback are essential to educate those who are expected to use resuscitation equipment in an emergency
POSTRESUSCITATION CARE AND TERMINATING THE RESUSCITATION
No criteria exist to guide clinicians when resuscitation is futile except in cases of obvious rigor mortis, evidence of decapitation, or evidence of decomposition. Complicating matters further, most resuscitation research is confined to the first minutes of resuscitation, leaving clinicians to use their best judgment as to whether resuscitative measures should continue. In general, clinicians should factor in the length of time the brain has been exposed to hypoxia and inadequate perfusion in deciding whether or not the hallmark of successful resuscitation—neurologically intact survival—is likely.
The decision to forgo further resuscitation efforts should be made by the entire resuscitative team in a manner that considers premorbid level of function, duration of the resuscitative effort, and all team members’ opinions and suggestions.
Literature supports allowing family members the decision to view the resuscitation prior to termination of efforts (with evidence suggesting that family members who witness the resuscitation have lower incidences of pathological mourning, anxiety and post-traumatic stress). Including family members in viewing resuscitation efforts requires a structured approach and an experienced team member assigned to family members to answer questions or address concerns.
In those in whom resuscitation appears successful, clinicians should consider initiation of therapeutic hypothermia. Extensive research has demonstrated that inducing hypothermia following cardiac arrest improves neurologic outcomes (as defined by cerebral performance category) when patients develop cardiopulmonary arrest due to a shockable rhythm. The data for improved neurologic recovery are not as strong in PEA or asystole.
Therapeutic hypothermia requires continuous central temperature monitoring via specialized equipment such as bladder thermography. The goal of therapeutic hypothermia is rapid coolingcore temperature via external cooling devices (ice packs or specialized cooling pads), internal cooling (utilizing chilled saline), or a combination of both. Since human physiology will attempt to usurp hypothermia through shivering, patients may require paralysis to prevent shivers. After 24 hours, paralytics and cooling devices may be discontinued with the patient being allowed to passively rewarm.
The majority of the literature concerning therapeutic hypothermia has concentrated on a goal temperature of 33°C for 12 to 24 hours postcardiac arrest. However, current data suggests that avoidance of hyperthermia (maintaining temperature less than 36°C) and fever—which may be damaging to neurons—provides similar benefit to maintaining lower temperatures. Since therapeutic hypothermia requires specialized equipment, its use is often confined to centers with experience and established institutional protocols for hypothermia. In the absence of these, clinicians should strive to maintain patient normothermia and avoid even minuscule elevation of body temperature above normal.
Patients in whom an acute coronary syndrome is posited to be the cause of cardiac arrest should undergo cardiac revascularization as soon as possible to prevent recurrent arrhythmias.
INTRAVASCULAR VERSUS INTRAOSSEOUS ACCESS
Access to the circulation is needed in patients during cardiac arrest. Traditionally this has been in the form of peripheral venous access or central venous access. Peripheral IVs are fairly straightforward and simple to obtain, even in patients in cardiac arrest. However, in patients with little available immediate IV access, including patients who are IV drug users or dialysis patients, another route may be needed. Central venous lines (internal jugular, subclavian vein, or femoral vein) are one option, but require a clinician skilled in their placement. They also carry risk of pneumothorax (internal jugular and subclavian), arterial puncture and thrombosis.
Intraosseeous needle insertion is a rapid (usually <1 minute placement time) and effective method of obtaining access in adult populations. Gazin et al showed the success rate of establishing intraosseous access to be 97% after training, including a 60 minute lecture and 1 hour practical session. The IO access proved to be effective for fluid resuscitation, ACLS drug delivery, and sedative/paralytic drug delivery. Therefore, when IV access is unobtainable or difficult, IO access provides a safe and effective method to obtain access to the circulation for fluid and drug delivery. Some hospital systems are even utilizing IO as the primary access when patients in cardiac arrest do not have adequate vascular access.
Only approximately one in seven patients survive in-hospital resuscitation to discharge, and of those who initially survive, only 19% are still living 6 months after IHCA. Fifty-one percent of survivors initially present in a shockable rhythm (ventricular fibrillation or ventricular tachycardia), but shockable rhythms are outnumbered in incidence to nonshockable rhythms (asystole or pulseless electrical activity) by nearly three to one in the inpatient setting.
Successful resuscitation preserves patients’ neurologic function that the AHA defines using the Cerebral Performance Category score (Table 137-4).
TABLE 137-4Cerebral Performance Category Score and Related Neurologic Outcome ||Download (.pdf) TABLE 137-4 Cerebral Performance Category Score and Related Neurologic Outcome
|Cerebral Performance Category ||Level of functioning |
|CPC 1 ||A return to normal cerebral function and normal living |
|CPC 2 ||Cerebral disability but sufficient function for independent activities of daily living |
|CPC 3 ||Severe disability, limited cognition, inability to carry out independent existence |
|CPC 4 ||Coma |
|CPC 5 ||Brain death |
Between 14% and 23% of survivors—whose prearrest neurological function was normal—develop moderate to severe cognitive deficits after resuscitation (Cerebral Performance Category 2-3). Fewer than 2% of survivors of IHCA will suffer prolonged coma or persistent vegetative state. While gender does not appear to predict outcomes in cardiopulmonary arrest, increased age may predict worse outcomes and lower survival probably less on the basis of age itself but rather through the accumulation of myriad comorbidities (particularly chronic kidney disease, coronary disease, and malignancy). Further multivariate data from the national registry of cardiopulmonary resuscitation (NRCPR) ultimately may help settle whether age is an independent predictor of short and long-term survival.
Debriefing with clinicians after every resuscitation event allows team members to contribute their intellectual and emotional concerns. Debriefing, even just a few minutes, should occur after every resuscitation event. All resuscitations require careful post hoc analysis for team compliance with established resuscitation protocols. One very valuable source of feedback is participation in NRCPR. This international collective of nearly 15% of all U.S. hospitals, and all of the Canadian Health Care System plus several other international sites, was created by the American Heart Association to collect standardized data on all inpatient cardiac arrests. Participating institutions receive quarterly reports on their resuscitations that may be compared to local, regional, national, and international norms established by all other participating institutions. These data may offer areas where improvement is needed (such as time to initial defibrillation from onset of cardiopulmonary arrest) and may serve as valuable reinforcement to team members of areas in which mastery has been accomplished. Furthermore, NRCPR data continue to help refine the cutting edge in resuscitation research and therefore meaningfully impact future patients’ outcomes from cardiopulmonary arrest.
Post hoc resuscitation analysis offers a partial intervention to impact future care, but continual retraining of providers offers the most proactive means to deliver high-quality care. Simulation labs offer an opportunity to objectively quantify team members’ capabilities at performing chest compressions, BVM ventilations, and other logistics of team flow that may elude analysis during real-life resuscitations. Since studies show rapid degradation in resuscitation skills (both manual and intellectual) within days of biannual resuscitation training, clinicians require continual feedback over the intervening months to ensure an adequate skill set. Whether this feedback occurs in a formal simulation environment or one on one with a manikin is probably less important than ensuring that skills are adequate and reinforced over time.
et al. Hypothermia for neuroprotection in adults after cardiopulmonary resuscitation. Cochrane Database. 2009;4:CD004128.
et al. Epidemiologic study of in-hospital cardiopulmonary resuscitation in the elderly. N Engl J Med
et al. Part 1: Executive Summary: 2010 American Heart Association Guidelines for Caridiopulmonary Resuscitation and Emergency Cardiovascular Care Science. Circulation
et al. Part 1: Executive summary: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science with Treatment Recommendations. Circulation
. 2010;122(16 Suppl 2):S250–S275.
et al. Targeted temperature management at 33°C versus 36°C after Cardiac Arrest. N Engl J Med
J. Intraosseous versus intravenous vascular access during out-of-hospital cardiac arrest: a randomized controlled trial. Ann Emerg Med
. December 2011;58:509–516.
et al. Rapid response team implementation and in-hospital mortality. Crit Care Med