Chapter 15. Cardiopulmonary Resuscitation

• Most cardiac arrests in the community setting occur as a result of coronary artery disease and cardiac ischemia.
• Given the high mortality of cardiac arrest, prevention is crucial.
• Cardiopulmonary resuscitation and rapid defibrillation are the keys to successful resuscitation from cardiac arrest.
• Advanced Cardiopulmonary Life Support (ACLS) guidelines provide treatment algorithms for the different cardiac rhythms of arrest.
• Automatic external defibrillators provide a means for rapid defibrillation by the public.

Cardiac arrest, defined as the sudden complete loss of cardiac output and therefore blood pressure, is the leading cause of death in the United States and much of the developed world, claiming at least 300,000 lives each year in the United States alone.1 In the majority of cases, myocardial ischemia in the setting of coronary artery disease represents the underlying etiology of arrest. Conversely, cardiac arrest is the initial presentation of myocardial ischemia in approximately 20% of patients.2 A wide variety of other processes can lead to cardiac arrest, including septic shock, electrolyte abnormalities, hypothermia, pulmonary embolism, and massive trauma (Table 15-1).

Survival from cardiac arrest remains dismal, even after the introduction of electrical defibrillation and cardiopulmonary resuscitation (CPR) over 50 years ago. In the best cases (witnessed ventricular fibrillation arrest with rapid defibrillation), survival to hospital discharge ranges from 30% to 46%,3,4 although overall out-of-hospital arrest survival is usually much lower, ranging from 2% to 26%.5 In large American cities, out-of-hospital arrest survival may be even worse—survival rates of 1.4% and 1.8% have been reported for New York and Chicago, respectively.6–8 Even after successful resuscitation from cardiac arrest, most patients die within 24 to 48 hours despite aggressive intensive care treatment. Reperfusion injury, a subject of much basic science investigation, is thought to be involved in this postarrest deterioration.9,10

Demographic data from multiple studies demonstrate that the mean age of patients who suffer out-of-hospital cardiac arrest is approximately 68 to 70 years, with a slightly higher incidence in men than in women.1,2,11 Over 70% of these patients experience arrest in the home or other residential location.12,13 In-hospital cardiac arrest patients exhibit similar demographics, with one survey showing a mean age of 71 years and also somewhat higher incidence in males.14 There do not appear to be significant survival differences between men and women.2

To standardize treatment during cardiac arrest, a number of treatment algorithms have been developed based on laboratory and clinical evidence. These have been compiled into the Basic Life Support (BLS) and Advanced Cardiopulmonary Life Support (ACLS) guidelines published and updated regularly by the American Heart Association's Emergency Cardiac Care Committee,15 as well as other international resuscitation organizations (International Liaison Committee on Resuscitation).15 For additional information about ACLS guidelines and their revisions, see the contact information listed in Table 15-2.

The majority of discussion in this chapter pertains to adult cardiac arrest because cardiac arrest in children, fortunately, is much less common. When it occurs, pediatric cardiac arrest more often is secondary to trauma or pulmonary derangements, such as drowning, status asthmaticus, or foreign-body obstruction, rather than due to a primary cardiac arrhythmia.16 However, ventricular fibrillation does occur in the pediatric population.17 Guidelines for pediatric resuscitation have been established and are compiled in the Pediatric Advanced Life Support (PALS) manual. For neonates, in whom cardiac arrest is yet another specialized problem, the manual Neonatal Advanced Life Support (NALS) has been developed. While many of the general principles of this chapter also apply to children, readers should refer to these additional texts for more detailed information.15

Given the poor prognosis of cardiac arrest, prevention remains the best hope to save lives. To this end, out-of-hospital and in-hospital cardiac arrests require different prevention strategies.

In the outpatient setting, careful attention to coronary artery disease risk factors such as smoking, hypertension, and hypercholesterolemia and aggressive treatment for these conditions can lower the risk of myocardial ischemia and therefore the risk of cardiac arrest. In consultation with their physicians, most patients with multiple cardiac risk factors should be treated with aspirin to lower the probability and severity of myocardial infarction. Patients otherwise at risk for sudden death, such as patients with bouts of ischemic ventricular tachycardia and/or a history of myocardial infarction with subsequently depressed ejection fraction, should be considered for implantable cardioverter defibrillator (ICD) placement (reviewed in refs. 18 and 19). The use of ICD devices remains an area of active investigation and likely will expand as smaller and less expensive devices are developed.

In the in-hospital setting, where sudden ventricular fibrillation/ventricular tachycardia (VF/VT) from coronary events is not the most common mechanism of arrest, prevention requires a different approach. Several studies have demonstrated that hospitalized patients who suffer cardiac arrest frequently exhibit signs and symptoms of destabilization up to 12 hours before they become pulseless.20,21 These symptoms include vital sign changes such as progressive hypotension, tachycardia, hypothermia, or hypoxia. They also include clinical changes such as mental status deterioration or progressive shortness of breath. Therefore, nursing staff should be appropriately vigilant in monitoring for such changes, and physicians should be duly attentive to warning signs from patients and staff. Early stabilization by such measures as intubation, initiation of vasopressor therapy, and/or transfer to an intensive care unit are far more effective than treating cardiac arrest once it has occurred. In Europe, this concept has been formalized into the ALERT course, in which hospital staff and physicians are trained to recognize and treat the early signs of destabilization that may lead to cardiac arrest.22–24

It is very important for hospitals, ICUs, and prehospital care systems to establish a clearly delineated team structure for cardiac arrest treatment. In-hospital studies have shown that a well-trained and organized arrest team is an important component in the resuscitation from sudden death.25 Team members should be ACLS trained and have a specified team leader who will lead the resuscitation efforts. Training should emphasize the need for a hierarchical structure, with the team leader making most treatment decisions, to prevent the confusion that often occurs during cardiac arrest events. It also should be emphasized that response time is critical, such that a resuscitation team should be able to arrive on the scene of a cardiac arrest within minutes to initiate treatment. Recent data demonstrate that groups with best-practice survival from cardiac arrest have mean “call to shock” times (for VF/VT arrest) of 5 minutes or less,26 and certainly the earlier the response, the more likely a better outcome will be obtained.

Resuscitation efforts during cardiac arrest require activities to be performed quickly, calmly, and in regimented fashion. Rescuer panic and disorganized efforts are counterproductive and can best be avoided by appropriate training before (and debriefing after) events take place. With the growth of medical simulation technology and sophisticated manikins for resuscitation training, it is possible for cardiac arrest teams to rehearse scenarios to supplement education and enhance preparedness.27,28

The first steps of resuscitation from cardiac arrest involve what is known as Basic Life Support (BLS). These fundamental skills are part of CPR training courses offered to the public by organizations such as the American Heart Association and the American Red Cross. Given the importance of early recognition and care for cardiac arrest, it is incumbent on all medical personnel from ward receptionists to radiology technicians to physicians to maintain BLS training. Health care workers also should encourage the public to obtain these skills, which are often summarized by the ABC's—airway, breathing, and circulation.

When a patient is first assessed, it must be determined if the patient is truly unresponsive. Stimulation by squeezing a shoulder or rubbing the sternum is reasonable, and if no movement, eye opening, or verbal response is noted, attention should be turned to the ABC's. The outdated BLS teaching that a patient should be shaken probably should be avoided given that the unresponsive state may be secondary to head and neck trauma. Care should be given to moving a patient as little as possible until this is determined. If a patient is deemed unresponsive, the initial observer immediately should call for help while assessing the ABC's.

To attempt optimal airway opening, the chin should be lifted, and the jaw should be thrust forward. A quick evaluation of the oropharynx should be performed to look for a foreign body, blood, or other occluding material. Any visualized foreign body should be removed by suction or by careful use of fingers or forceps. After this evaluation, several “rescue breaths” should be delivered via mouth-to-mouth or mask-to-mouth technique. If the chest wall does not rise with these breaths, it is possible that a complete airway obstruction exists, and abdominal thrusts should be performed to attempt airway clearance. If these fail, trained personnel may need to establish a surgical airway via cricothyrotomy.

While holding the chin and jaw in the correct position, breaths should be delivered during initial efforts until a more definitive airway can be obtained. In cardiac arrest, this is performed via endotracheal intubation, which is performed routinely by anesthesiologists, emergency physicians, respiratory therapists, and paramedics. If possible, ventilation should be performed with maximal FiO2 via bag-valve-mask until intubation is performed. Pulse oximetry can be used to monitor patient oxygen saturation during this process.

The hemodynamic status of the patient should be assessed via palpation of arterial sites. As an approximate rule of thumb, the radial, femoral, and carotid pulses are lost at systolic pressures below 80, 70, and 60 mm Hg, respectively.15 Therefore, the most sensitive site to assess is the carotid artery. If no pulse can be felt at the carotid, chest compressions should be initiated immediately. Compressions should be performed at a rate of 100 beats per minute and to a depth of at least 1 to 2 in.

Recent studies have demonstrated the importance of “good quality” chest compressions, partially defined by compressions performed at the appropriate rate and depth. This is important in light of the fact that an observational study has shown performance of chest compressions to be grossly suboptimal in practice.29 Rescuers should pay particular attention to the performance of this skill. Future generations of resuscitation devices may aid this process by monitoring the quality of chest compressions and generating alarms during suboptimal performance. Exciting recent data have suggested that chest compressions may be more important than defibrillation in the initial treatment of cardiac arrest.30,31 These observations, which might have been considered heretical just several years ago, lend support to an important new paradigm in cardiac arrest, that of the three-phase model of cardiac arrest (see discussion later in this chapter).32 It has been proposed that 90 seconds of CPR be performed before defibrillation,33 but this recommendation has not been evaluated fully for inclusion in the American Heart Association ACLS guidelines.

Monitoring the adequacy of the circulation during performance of CPR traditionally has been based on palpation of pulses. Capnography is an attractive adjunct to bedside clinical monitoring because the amount of carbon dioxide returned from peripheral tissues and then exhaled from the lung should be a measure of the adequacy of cardiac output. In one prospective, observational study,33a 150 consecutive cardiac arrests out of hospital were monitored by end-tidal carbon dioxide levels after intubation. After 20 minutes of advanced cardiac life support, end-tidal carbon dioxide levels averaged 4.4 ± 2.9 mm Hg in nonsurvivors and 32.8 ± 7.4 mm Hg in survivors (p <0.001). A 20-minute end-tidal carbon dioxide value of 10 mm Hg or less successfully discriminated between the 35 patients who survived to hospital admission and the 115 nonsurvivors. While not yet routine, capnography may be useful for both judging the adequacy of resuscitative efforts and offering prognostic information.

While not strictly part of BLS, fluid resuscitation is a crucial adjunct to circulatory support in the initial phases of resuscitation. Intravenous access should be obtained rapidly if it is not already present, and adult patients should receive a rapid infusion of at least 500 to 1000 mL of 0.9% saline or lactated Ringer's solution. In children, the crystalloid infusion should be calculated at 20 mL/kg. The ideal IV access would include a peripheral large-bore (i.e., 14 to 18 gauge) catheter and/or large-bore central catheter (i.e., not a double- or triple-lumen catheter). The optimal site for central line placement in resuscitation is the femoral vein because the chest and neck are busy sites for chest compressions and ventilatory support, respectively, making subclavian or internal jugular approaches impractical unless already present. In children, an intraosseous line may be a convenient approach to initial access.

Ventricular tachycardia (VT) may or may not generate a pulse. Therefore, it is crucial to assess the hemodynamic status before ACLS resuscitative measures are begun. If the patient has a pulse, is conscious, and has only mild complaints of palpitations, mild chest discomfort, weakness, and/or anxiety, immediate electrical cardioversion is not required. On the other hand, if the patient exhibits signs of instability, including syncope, severe chest pain, or marked hypotension, then cardioversion should proceed immediately after appropriate sedation is delivered.

The treatment of VT with a pulse includes intravenous administration of lidocaine or amiodarone and supportive care (oxygen administration and preparation for electrical cardioversion). In cases in which both VT and supraventricular tachycardia with aberrancy cannot be distinguished with certainty, the use of procainamide can be considered because it is appropriate for both conditions. Recurrent VT often requires electrophysiologic evaluation and treatment, including the placement of an implantable cardioverter defibrillator (reviewed in ref. 34).

Ventricular fibrillation and ventricular tachycardia without a pulse (VF/VT) are grouped together because both require the same treatment—immediate defibrillation. In fact, defibrillation should precede any other assessment or treatment. Studies have shown consistently that the earlier a patient is defibrillated successfully, the better are the chances for survival.35 This observation has stimulated the use of automatic external defibrillators (AEDs) in airports and other public locations (see the section on AEDs later in this chapter).

When a patient is found in VF/VT, up to three defibrillatory shocks should be delivered in rapid succession, with a brief pause between each shock to examine the cardiac monitor (see Fig. 15-1 for a VF/VT algorithm). There is no need to feel for a pulse between shocks unless the rhythm has changed. Using standard monophasic defibrillators, the first shock should be delivered at 200 J; the next two shocks can be delivered at 200 J or at 300 and 360 J, respectively, in the energy escalation proposed in the ACLS guidelines. Using biphasic defibrillators (see later in this chapter for a discussion of different defibrillator types), all shocks should be delivered at 150 J or at the energy suggested by the manufacturer.4

If a patient remains in pulseless VF/VT immediately after these three shocks, further treatments include assessment of the ABC's and pharmacologic adjuncts. Chest compressions should be initiated immediately, with care taken to ensure compression quality, as discussed previously. Patients should be intubated immediately and ventilated with 100% O2, with care taken to ensure correct endotracheal tube placement by both auscultation and end-tidal CO2 detection, if available. If not already performed, IV access should be established. Large-bore (not multilumen) central venous access by the femoral approach is most convenient and practical in this setting if skilled personnel are available. It is often useful to obtain an arterial blood gas sample at this point as well because it will take some time for the results to return to the team in any case. While these steps are taking place, the arrest team leader should rapidly obtain a very brief history from available sources, including nursing staff, family, or physicians caring for the patient. The most important details to obtain are when the patient was last seen with a pulse, what pertinent medical problems the patient has, and what has taken place in the last few hours of patient observation.

In VF/VT arrest, either epinephrine 1 mg IV or vasopressin 40 U IV should be given early, preferably within the first 5 minutes of resuscitation efforts. Vasopressin, a relatively new addition to the ACLS guidelines, has been shown to improve coronary perfusion pressure and possibly improve initial resuscitation compared with epinephrine.36–38 However, despite some optimism regarding the theoretical advantages of vasopressin over epinephrine, conclusive data showing improved survival to hospital discharge are still lacking.39 Additionally, high doses of epinephrine (e.g., 3 or 5 mg IV) previously had been hoped to have additional benefit over the standard 1 mg IV dose, but human trials have demonstrated no significant survival benefit.40–42

After the administration of epinephrine or vasopressin, another shock at 360 J should be delivered. If the patient remains in VF/VT after this, additional medications can be administered before additional defibrillation attempts. One such pharmacologic option is amiodarone, given as a 300-mg IV bolus. Two studies have shown clearly higher initial survival with amiodarone compared with lidocaine, although lidocaine and amiodarone have yet to show an effect on survival to hospital discharge.43,44

Pulseless electrical activity (PEA) is a state defined by having no detectable pulse despite the appearance of an organized electrical rhythm on cardiac monitoring. This electrical activity may appear as so-called normal sinus rhythm or similar pattern but must be differentiated from pulseless VF/VT, which requires specific treatment (see above). PEA used to be described as “electromechanical dissociation” (EMD), based on the assumption that some pathophysiologic process had separated the normal electrical conduction of the heart from its ability to cause myocardial contraction. Echocardiographic studies of myocardium in PEA, however, demonstrate some degree of visible muscle activity.45 Thus, the term EMD has fallen out of favor compared with the more appropriate PEA.

The initial approach to a patient in PEA is much the same as the approach to patients with other pulseless rhythms (see Fig. 15-2 for a PEA algorithm). CPR should be initiated immediately, a definitive airway should be established, and the patient should be ventilated with 100% O2. Large-bore central venous access should be established. An arterial blood gas sample should be obtained early, with a request to check hemoglobin and potassium, if possible. A brief history should be taken from staff by the resuscitation team leader regarding the events leading up to the pulseless state. Neck veins should be examined to consider cardiac tamponade, lungs should be auscultated to rule out tension pneumothorax, and if appropriate, the patient's temperature should be taken to evaluate for hypothermia. All patients in PEA should receive epinephrine 1 mg IV early in the resuscitation efforts.

In contrast to VF/VT, there is no specific treatment for PEA per se. There are, however, a number of appropriate therapeutic options depending on the cause of PEA. Therefore, a differential diagnosis must be considered quickly (Table 15-3). Intravascular hypovolemia, through either hemorrhage or leakage of fluid from the vascular compartment, is a common cause of cardiac arrest in the hospital setting. A high suspicion for this etiology should be maintained for patients who recently underwent surgery or for patients known to have a serious infectious process and now may be presenting in septic shock. Hypoxemia is another common cause of PEA, although ventilation with 100% O2 renders this largely a diagnosis of exclusion. A third common process underlying PEA is hypothermia, which again may be common in postoperative and/or septic patients. Aggressive treatment with warm IV fluids and warming blankets should be undertaken in these patients during resuscitation. Resuscitative efforts should not be terminated until all efforts have been made to warm the patient to normothermia.

Tension pneumothorax also can lead to PEA arrest and should be considered especially in any patient who was mechanically ventilated prior to arrest. Lung auscultation during ventilatory cycles should help to identify this problem. If lung sounds are diminished unilaterally, the endotracheal tube should be pulled back several centimeters to determine if a main stem bronchus intubation was present. If decreased breath sounds persist after this maneuver, a needle thoracostomy should be performed with a 16- to 18-gauge IV catheter in the second intercostal space on the side exhibiting diminished lung sounds.

Another major cause of PEA arrest that deserves some discussion is massive pulmonary embolism, which can lead not only to hypoxemia but also to cardiogenic shock from sudden right ventricular (RV) dysfunction. If the resources are available, rapid echocardiography can be performed during resuscitation efforts to evaluate the size and function of the right ventricle. A markedly enlarged and poorly contracting right ventricle supports the diagnosis of pulmonary embolism. This diagnosis also should be considered for patients with known deep venous thrombosis or possible thrombophilia from disease processes such as malignancy or systemic lupus and for patients who have been hospitalized and/or immobile for at least several days. If pulmonary embolism is strongly suspected, treatment with tissue plasminogen activator (tPA) can be considered. The use of tPA in cardiac arrest remains controversial, however (see discussion of tPA later in this chapter).

The patient should be evaluated for other causes of PEA. These include cardiac tamponade, for which pericardiocentesis with a long spinal needle can be lifesaving; hyperkalemia, requiring treatment with intravenous calcium, insulin, and glucose; and drug toxicity, which requires specific therapies depending on the drug in question.

In some cases, a pulseless state can occur if the heart rate slows dramatically, e.g., to less than 30 to 40 beats per minute. This can occur in cases of complete heart block, hypoxemia, hypothermia, and toxic states from certain medications, especially β blockers and calcium channel blockers. In a sense, pulseless bradycardia is a subset of PEA arrest in which additional specific treatments may be attempted beyond that for PEA itself.

After standard ACLS maneuvers including CPR, intubation, and ventilation with 100% O2 and administration of epinephrine (see preceding sections), treatment should be directed toward increasing the rate of electrical activity, which may be sufficient to generate a blood pressure (and therefore a pulse). If this increased rate does not generate a pulse, the patient truly can be considered to be in PEA arrest.

Methods to increase the heart rate include administration of atropine 1.0 mg IV, which may be repeated up to a total of 0.04 mg/kg, and a trial of electrical pacing (see Fig. 15-3 for a bradycardia algorithm). Transcutaneous electrical pacing, a standard capability of most hospital defibrillators, can be attempted as well, although no evidence-based recommendations exist regarding rate and current to be applied. We recommend starting at a rate of 80 impulses per minute and 50 mA current, ramping up the current as necessary to obtain capture, if possible at all.46

All patients in cardiac arrest eventually converge into the rhythm of asystole, which is defined as no discernible electrical activity on cardiac monitoring (see Fig. 15-4 for an asystole algorithm). An occasional wide complex can be seen in asystole, which is known as an agonal rhythm—this carries the same grave prognosis as asystole itself. Unwitnessed cardiac arrest with the presenting rhythm of asystole has a dismal rate of survival, usually considered to be less than 1%.47 There are very few treatment options for rescuers confronted with asystolic patients, and therefore, a rapid search for reversible causes combined with standard ACLS measures in most cases should not lead to lengthy resuscitation efforts.

Besides standard resuscitation techniques (described earlier), including chest compressions, intubation, ventilation with 100% O2, and administration of epinephrine, atropine infusion and a trial of electrical pacing may be used. Atropine should be dosed intravenously in 1-mg increments every 3 to 5 minutes during arrest, up to 0.04 mg/kg. Transcutaneous electrical pacing may be attempted as well, following the same recommendations as those for bradycardia (see discussion on bradycardia above).

After three rounds of epinephrine and atropine administration, 10 to 15 minutes of chest compressions, and possibly transcutaneous pacing, the clinical situation should be reassessed. If the rhythm remains in asystole, efforts should be terminated, because survival after 10 to 15 minutes of confirmed asystole despite pharmacologic intervention and intubation is essentially zero.48,49 An important caveat in the assessment of asystole is that at least two cardiac monitoring leads should be examined for a rhythm—often what appears to be asystole in one lead actually represents a loose electrical connection, and one might find a treatable rhythm in another lead.

The subject of resuscitation team function and performance remains poorly studied, and usually ACLS training gives short shrift to team cooperation and leadership skills. The decision to terminate efforts represents a difficult moment for the resuscitation team and the team leader.50 One simple recommendation to ease the tension of this moment and ensure that all reasonable effort has been given to save the life of the patient is to involve the entire team in the termination process. We recommend that the team leader, sensing that effort has become futile, should verbally summarize to the team all the treatment rendered so far, e.g., how long CPR has been performed, what drugs and shocks were given, and what underlying arrest etiologies were considered. The team leader then can ask if any team member has final recommendations or suggestions before efforts are halted. In this fashion, the decision to stop resuscitation procedures is made by the group, and staff will feel satisfied that resuscitation was not terminated prematurely. After termination, it is often useful to conduct a debriefing session among key team personnel before disbanding, especially to troubleshoot any technical or team function problems. Hospitals should establish CPR review committees to monitor the quality of resuscitations on a periodic basis and implement system changes as necessary to improve outcomes.

Modern electrical defibrillation, or the use of electric current applied directly to a patient's chest to restore a viable heart rhythm, grew out of research into electrocution deaths among maintenance workers at Consolidated Edison of New York. The first human defibrillation was performed intraoperatively by Claude Beck in 1947; the first external defibrillation was undertaken by Paul Zoll in 1955.51 Since that time, defibrillation has become a cornerstone of cardiac resuscitation and has been used successfully by physicians, nurses, paramedics, police, and even the public at large.

The exact mechanism of defibrillation remains uncertain. Whether a critical number of myocardial cells require membrane depolarization to overcome ventricular fibrillation or whether certain regions of the heart must achieve a critical current density remains a subject of active study. Several mechanistic aspects are clear, however. The energy discharged (measured in joules, or watt-seconds) appears to have both dose-response and therapeutic window characteristics. That is, the chance of successful defibrillation rises with increasing energies delivered; however, as energy is increased further, functional myocardial injury predominates over useful resuscitative properties. With standard defibrillators, 200 J is the recommended energy for initial shock; 300 and 360 J are accepted levels for subsequent attempts to defibrillate. For children, the recommended initial dose is 2 J/kg, with a second shock at 4 J/kg if the first shock fails.

Technique of defibrillation is also important. Firm pressure must be applied with defibrillation paddles to ensure proper delivery of energy without electrical arc or skin burn. Similarly, defibrillation pads must be well applied to the chest. Positioning of paddles or pads must ensure that the imaginary line connecting the two electrodes runs through the heart. That is, in one standard approach, an electrode should be placed at the right upper sternal border and the other at the left midaxillary line near the apex of the heart.

Perhaps the most important observation regarding electrical defibrillation is that the longer the delay before a shock is delivered, the less chance there is for a successful resuscitation. Ventricular fibrillation or tachycardia should be defibrillated immediately; this is the fundamental principle underlying implantable cardioverter defibrillators (ICDs), a commonly placed device for patients with recurrent ventricular tachycardia or history of cardiac arrest.18 If VF/VT persists for even 5 minutes without CPR or defibrillation, the chance for a successful outcome falls dramatically.

Given the need for early defibrillation, automatic external defibrillators (AEDs) have become an important tool for paramedics and the lay public. These devices, commonly found in airports and other heavily trafficked public locales, perform rhythm analysis and provide defibrillatory shocks if needed. In theory, no prior experience should be required to operate such a device. AEDs are discussed in more detail below.

Most defibrillators in use today generate monophasic waveform shocks. Over the past few years, biphasic defibrillators have become available commercially and likely will become the standard of care over the next decade. These defibrillators have been shown to be equally effective as monophasic devices at lower energies, which may optimize the benefits of the shock while minimizing myocardial injury.4

There is hope that in the coming years resuscitation science will offer substantially improved survival for victims of cardiac arrest. With the success of early defibrillation programs in airports, casinos, and other public places, survival rates in these special locales have soared to greater than 50%.52 However, a new paradigm has emerged in our understanding of sudden death. The three-phase time-sensitive model of cardiac arrest, based on data from the past several years, offers the hope of better survival with therapy tailored to the time after initial arrest.32

This model proposes that time in cardiac arrest can be divided into different phases: the electrical phase (the first 4 minutes after arrest), the circulatory phase (minutes 4 through 10), and the metabolic phase (after 10 minutes), with each requiring different therapeutic approaches (Fig. 15-5). The electrical phase calls for defibrillation as the first therapy for VF/VT and is currently our standard of care regardless of time spent in cardiac arrest. This fits well with national efforts to get more rapid defibrillation with AEDs—because the evidence suggests that defibrillation within the first few minutes is associated with a better than 50% chance of initial survival. However, a challenge during this electrical phase is the need to get defibrillators rapidly to victims at home, where over 70% of cardiac arrests take place. The circulatory phase appears to be best treated initially with chest compressions and ventilations and then followed by defibrillation after several minutes of CPR. Note that this is not the current standard of care according to the American Heart Association at the time of this writing. Using this “CPR first” algorithm, paramedic services in Norway have improved survival rates from 4% to 20% over standard advanced cardiac life support during this circulatory phase.30 However, another challenge becomes apparent during this phase: our current quality of CPR remains unmeasured and poorly controlled. Recent studies would suggest that CPR quality in real resuscitation falls far short of the high quality required for survival—and new technology offers us the ability to markedly improve on this in the next few years.29 This circulatory phase may be difficult to identify because we usually do not have accurate information on time of collapse and thus may not know in which phase a patient resides. The circulatory phase depends on very good CPR, so prioritizing good compression rate, compression depth, minimal pauses in compression, and proper ventilatory management all become critical priorities.

The third so-called metabolic phase is the most lethal and challenges our basic scientific understanding of ischemia and reperfusion injury. Novel therapies, such as advanced cardioprotective pharmacologic agents, cardiopulmonary bypass, induced hypothermia (see Chap. 16), preconditioning pathways, inflammatory mediators, apoptosis signaling, and hibernation may offer promise in the understanding and treatment for this phase—but the need for new translational research in this area is vital. The tools of molecular biology, proteomics, and cellular physiology are likely to provide important insights and to create new biosensors that can guide clinical therapies. It is not unrealistic to believe that major improvements in survival rates will result as we change our current practices in the near future.53

Given the assumption that early defibrillation remains the best treatment for VF/VT cardiac arrest, a number of devices have been developed to allow inexperienced users to defibrillate victims before the arrival of medical personnel (reviewed in ref. 54). These devices, known as automatic external defibrillators (AEDs), have become ubiquitous in airports and other public locations. These simple-to-use defibrillators contain waveform analysis software that determines whether a shock is warranted when a layperson attaches sensing pads to the chest of a comatose individual. Appropriate shocks are then delivered. Audio prompts guide the user through the process.

The placement of AEDs in public places has been shown to affect survival from cardiac arrest, supporting the concept that earlier defibrillation correlates with improved outcomes.52 However, the majority of cardiac arrests occur in the home, not in public. Data from Seattle suggest that as many as 70% of out-of-hospital cardiac arrests take place in residences, and only 21% occur in public locales.1 Whether AEDs should be available for home installation, much like fire extinguishers, remains an active question. As AEDs become smaller, smarter, and cheaper, this debate may tip toward home availability.55

Whether AEDs should be placed in hospital wards remains another topic under current discussion.56 Although hospital resuscitation teams include ACLS-trained personnel, most “first responders” in the hospital setting are nurses or other health care staff who may not be ACLS proficient and therefore unlikely to perform defibrillation. It has been argued that the availability of AEDs in the hospital would allow for rapid defibrillation attempts before the arrival of resuscitation teams. However, the presence of AEDs would not be sufficient—nurses and other health care workers would have to accept defibrillation as a possible primary responsibility. There are some data to suggest nurses would support such a role.57

In the search for novel cardiac arrest therapies, induced hypothermia has generated a great deal of recent interest, spurred by two well-conducted studies showing improved survival when patients were cooled to 32 to 34°C after resuscitation from cardiac arrest.58,59 An international recommendation has been issued based on this evidence that patients should be cooled after out-of-hospital cardiac arrest; data on in-hospital cardiac arrest are still under discussion.60 Much work remains to further define this treatment, regarding both depth and duration of hypothermia. Novel techniques for cooling patients are under development as well, including multiphase coolant fluids and cooling catheters. Chapter 16 of this book is devoted to this exciting field of induced hypothermia.

CPR and electrical defibrillation are the central treatment modalities for cardiac arrest in current practice. While medications such as epinephrine, atropine, lidocaine, and amiodarone have been incorporated into treatment algorithms for cardiac arrest, to this day they do not have any proven survival benefit.61 A surge of interest in “high dose” epinephrine in recent years was quelled when a number of studies demonstrated no benefit from this approach.42,62 Current interest in amiodarone as a treatment for VF/VT is based largely on one study that showed an improvement in initial resuscitation but did not demonstrate an improved survival to hospital discharge.43 There are no definitive data to demonstrate a survival benefit from atropine or lidocaine. Similarly, bicarbonate, while widely administered during cardiac arrest, has not been proven to aid resuscitation. In fact, ACLS guidelines only recommend bicarbonate infusion in a small subset of cardiac arrest patients, namely, those known to be hyperkalemic.15 Doses of standard ACLS medications are given in Table 15-4.

Thrombolytic therapy in cardiac arrest has received recent interest because a number of uncontrolled studies and cohort series have suggested a benefit from the use of urokinase or tPA.63–65 A small but well-executed controlled study recently demonstrated no improvement in return of spontaneous circulation or survival with tPA in the treatment of PEA arrest.66 A larger European study is currently ongoing and may help to resolve this controversy, and certain subsets of cardiac arrest patients may be found to benefit from this treatment modality. At this point, it is fair to say that thrombolytic therapy may be attempted if there is strong evidence to suspect pulmonary embolism as the cause of arrest.61,67

The idea of a chemical code, i.e., performing resuscitation with pharmacologic agents only and not with chest compressions or defibrillation, is not controversial insofar as there is no disagreement among expert providers. Studies have demonstrated clearly that the concept lies much more in the realm of mythology or wishful thinking than in science. The only controversy is that the concept has persisted in hospitals and among health care workers across the world to this day.68 It is important to stress the following point because the chemical code is often presented to family members as an option for care of their loved one: cardiac arrest is not a medical problem treatable by medications only.

In a similar vein, the slow code, in which efforts to resuscitate are intentionally delayed or limited by rescuers, is ethically unacceptable.69 Patients who are full code should have every appropriate effort made to resuscitate them; decisions regarding appropriateness of resuscitation efforts should be made by patients and their primary physicians, not by a resuscitation team at the time of arrest.

The ethical dimensions of cardiac arrest treatment are complex and important for physicians to consider.70,71 Decisions regarding termination of efforts and even the decision not to initiate efforts in the first place should be calibrated carefully depending on the individual case in question. The growing establishment of do not resuscitate (DNR) protocols has allowed patients and their families to avoid the traumatic and often futile efforts of resuscitation.

It cannot be stressed enough that physicians should initiate frank and truthful end-of-life discussions with patients early in their care, before hospitalization or cardiac arrest appear on the horizon. In this fashion, the use of cardiac arrest treatment can be judiciously tailored to the appropriate patients.72 Physicians must emphasize the distinction to patients between DNR and comfort care. That is, a DNR order means that all curative measures could be employed except chest compressions and defibrillation. This distinction is also important for hospital personnel and physicians to understand, lest a DNR order influence other care decisions in the critically ill. In short, do not resuscitate should never mean do not treat.73