Chapter 82. Cardiopulmonary Resuscitation

1. During many in- and out-of-hospital resuscitation attempts less than half the time is devoted to chest compressions. Interrupting compressions reduces myocardial perfusion and is detrimental to the ultimate success of resuscitation. Maintaining uninterrupted chest compressions must be the first priority during cardiopulmonary resuscitation.

2. The relative importance of chest compressions, ventilation, and defibrillation during resuscitation must be adjusted for the context of the rescue situation and the personnel available.

3. When an arrest is witnessed, the arrest is likely to be of cardiac (rather than respiratory) cause, and advanced care will be available within a short time, closed chest compressions alone may be as efficacious as compressions and mouth-to-mouth ventilation.

4. Fluctuations in intrathoracic pressure play a significant role in blood flow during most resuscitations, and the cardiac pump mechanism contributes under some circumstances.

5. The critical myocardial blood flow is associated with aortic "diastolic" pressure exceeding 40 mm Hg.

6. Cardiac output is severely depressed during cardiopulmonary resuscitation (CPR), ranging from 10% to 33% of prearrest values in experimental animals. Nearly all of the cardiac output is directed to organs above the diaphragm.

7. During CPR, measurement of blood gases reveals an arterial respiratory alkalosis and a venous respiratory acidosis because the arterial PCO2 is reduced and the venous PCO2 is elevated.

8. Exhaled end-tidal CO2 is an excellent noninvasive guide to the effectiveness of standard CPR.

9. Immediate defibrillation is most effective if applied within 4 to 5 minutes of collapse. Otherwise, a brief period of 2 to 3 minutes of chest compressions before defibrillation may improve survival.

10. Amiodarone and lidocaine are used during cardiac arrest to aid defibrillation when ventricular fibrillation is refractory to electrical countershock therapy or when fibrillation recurs following successful conversion.

11. Outcome studies prospectively comparing standard and high-dose epinephrine have not demonstrated conclusively that higher doses will improve survival.

12. Evidence currently suggests that, like other potent vasopressors, vasopressin is equivalent to but not better than epinephrine for use during CPR.

13. In contrast to pharmacologic therapy, 2 studies demonstrate improved neurologic outcomes when mild therapeutic hypothermia (89.6°F-93.2°F [32°C-34°C]) is induced for 12 to 24 hours in cardiac arrest survivors who remained comatose after hospital admission.

In the hospital, resuscitation from cardiac arrest depends on the rapid response of a well-trained cardiopulmonary resuscitation (CPR) team that usually includes physicians, nurses, respiratory care providers, and pharmacists. The anesthesiologist is usually a team member and frequently is expected to be the team leader. The team must be coordinated and communicate readily. The goals of published CPR guidelines and widely taught courses in basic life support (BLS), advanced cardiac life support (ACLS), and pediatric advanced life support (PALS) are to provide the knowledge base and framework for effective teamwork.1 To function within the team, the anesthesiologist must be thoroughly familiar with ACLS protocols. Team leadership requires in-depth, current knowledge of physiology, pharmacology, and alternative CPR techniques. This chapter provides the scientific background on which current CPR practice is based. It focuses exclusively on cardiac arrest. Other circumstances requiring cardiovascular support, such as shock and dysrhythmias, are discussed in other chapters.

Approximately 50% of patients suffering in-hospital cardiac arrest are resuscitated, and 18% of adults and 27% of children survive to discharge.2 The operating room is the location where CPR has the highest rate of success. Cardiac arrest occurs approximately 7 times for every 10000 anesthesia events.3 The cause for the arrest is anesthesia related approximately 4.5 times out of every 10000 anesthesia events, but mortality from these arrests is only 0.4 per 10000 anesthesia events. Resuscitation is successful approximately 90% of the time in anesthesia-related cardiac arrests. Excluding the operating room, the best initial resuscitation rates are in reports from the intensive care unit (ICU), and the best survival rates are shown in patients arresting in the emergency department. The in-hospital success is similar to resuscitation and survival rates from out-of-hospital arrest in cities with rapid response emergency medical systems.4 In out-of-hospital arrests, poor outcomes are associated with (a) long arrest times before CPR is begun, (b) prolonged ventricular fibrillation (VF) without definitive therapy, and (c) inadequate coronary and cerebral perfusion during CPR.5 Therefore, optimum survival is obtained only if uninterrupted chest compressions are started immediately and the emergency system is activated so that defibrillation can be applied within 8 minutes.4 Rapid response times and expert personnel factor into better outcomes for in-hospital arrests. Intercurrent illnesses in hospitalized patients reduce the likelihood of survival and the arrest victim is more likely to be elderly, a factor that may reduce survival. The cause of arrest associated with the best outcome is ventricular fibrillation secondary to myocardial ischemia. This initiating event is less common in hospitalized patients. When applying CPR, attention to the details of effective resuscitation is important. However, CPR is only symptomatic therapy, thus attention should not only be paid to the mechanics of CPR but to a search for a treatable cause of the arrest.

Discoveries contributing to modern CPR have a long history recorded in many famous works.6,7 These discoveries begin with the Bible story of Elisha breathing life back into the son of a Shunammite woman (II Kings 4:34) and continue with Andreas Vesalius's description of tracheotomy and artificial ventilation, William Harvey's manual manipulation of the heart, and the teachings of the Society for the Recovery of Persons Apparently Drowned, founded in London in 1774. Nevertheless, the combined techniques of modern CPR are just 50 years old. Although there were significant contributions from other centers in the United States and Europe, modern CPR developed primarily from the fortuitous assemblage of innovative clinicians and researchers in Baltimore in the 1950s and early 1960s. In the late 1950s Elam, Safar, and Gordon established mouth-to-mouth ventilation as the only effective means of artificial ventilation.8-11 The internal defibrillator was developed in 1933 but was not applied successfully until 1947.12,13 It was another decade before its general use was made possible by the development of external transthoracic defibrillation.14,15 Despite these advances, widespread resuscitation from cardiac arrest was not possible until Kouwenhoven, Jude, and Knickerbocker described success with closed chest cardiac massage in a series of patients.16 The final major component of modern CPR was added in 1963 when Redding and Pearson described the improved success obtained by administering epinephrine or other vasopressor drugs.17

The process of cardiopulmonary resuscitation consists of multiple elements including recognition of arrest; activation of an emergency response system; attention to airway, breathing, circulation, drugs, electrical therapy (ABCDEs); and postresuscitation care. Traditionally, basic life support (BLS) included the ABCs and was considered appropriate for the lay rescuer as well as the health care provider, without the use of additional equipment. Advanced life support (ALS) included BLS plus all other available modalities. The distinction between BLS and ALS has become blurred with the addition of public access defibrillators, more individuals trained in CPR, and the recognition that optimum performance of BLS provides better outcomes than the addition of any advanced therapies.

Physiologically, the objective of performing CPR is to provide oxygen to the vital organs—the brain and especially the heart—until normal circulation is restored. Cardiac arrest caused by ventricular fibrillation can be characterized in 3 phases (Table 82-1).18 The electrical phase occurs during the first 4 to 5 minutes of the arrest, and early defibrillation is critical for success during this time. The hemodynamic phase follows for the next 10 to 15 minutes when perfusing the myocardium with oxygenated blood is critical. This is followed by what has been called the metabolic phase, when the ischemic injury to the heart is so great that it is not clear what interventions will be successful. This mechanism of cardiac arrest (that caused by VF) is most common in adults and obviously requires primary attention to defibrillation and perfusing the myocardium. Because arrest occurs with normal oxygenation of the blood and a reservoir of oxygen in the lung, ventilation is relatively less important. Cardiac arrest in children and certain adult situations, such as drowning, is more likely to be caused by hypoxia. In this circumstance, although restoring circulation with chest compressions remains important, providing adequate ventilation to reoxygenate the blood is also necessary. Consequently, rescuers are encouraged to take the context of resuscitation into account when applying the elements of CPR, adjusting the priorities to the situation at hand. These 2 different circumstances account for the slightly different approach to the airway-breathing-compression sequence in adults and children (Boxes 82-1 and 82-2).

Traditionally, CPR has been presented as a sequence of distinct steps (algorithm) to help a single rescuer prioritize actions or decisions. This approach has been continued in Boxes 82-1 and 82-2. However, it is common for multiple providers to be present at resuscitation and each should take responsibility for simultaneously instituting a critical action (eg, one to activate the emergency response, one to begin chest compressions, and one to retrieve a defibrillator). The actual priority of actions may vary depending on the context of the cardiac arrest.

It has been repeatedly demonstrated that early intervention is necessary for resuscitation attempts to be successful. CPR has had the most dramatic effect when prompt defibrillation is applied during the electrical phase of ventricular fibrillation, and the circumstance in which public-access automated external defibrillation has proven beneficial. The high oxygen consumption of the fibrillating heart causes the rapid depletion of myocardial high-energy phosphates. Myocardial adenosine-5′-triphosphate (ATP) levels during VF correlate with the success of defibrillation and, by about 4 minutes, the ATP levels in the heart have fallen to levels that make restoration of normal contractile function problematic. Effective chest compressions help replete or delay reductions in ATP by restoring myocardial blood flow. Therefore, the most important intervention during the hemodynamic phase of cardiac arrest is producing coronary perfusion by means of chest compressions. In the absence of prompt defibrillation, the most important intervention for neurologically normal survival from cardiac arrest is restoration and maintenance of cerebral and myocardial blood flow. Because perfusion pressures generated by chest compressions are quite low compared to the intact circulation, any interruption of chest compressions markedly reduces the chances for neurologically normal survival. Therefore, any intervention that interrupts chest compressions is strongly discouraged.

Early intervention in cardiac arrest requires not only rapid response emergency medical systems for out-of-hospital arrests and appropriate code teams for in-hospital arrests, but also immediate intervention by bystanders when cardiac arrest victims collapse. Unfortunately, the incidence of bystander CPR has been falling for 3 decades. The reasons for this reluctance to intervene are multiple but primarily seem to be (a) a lack of training, (b) the complexity of the task, and (c) a fear of doing harm. Many of these concerns focus on the mouth-to-mouth ventilation part of the CPR intervention and many experts are now stressing the importance of continuous chest compressions, even at the expense of ventilation, by bystanders and early rescuers. There is no doubt that chest-compression-only CPR given by bystanders is more acceptable for most, easier to learn and retain, and provides better hemodynamic support.

Early studies in anesthetized humans suggested that the airway would not remain open in the unconscious, leading to the teaching that airway control and artificial ventilation must accompany chest compressions.8-11 However, there are now considerable data to suggest that eliminating mouth-to-mouth ventilation early in the resuscitation of witnessed fibrillatory cardiac arrest is not detrimental to outcome and may improve survival. Data from the Belgian Cardiopulmonary Cerebral Resuscitation (CPCR) Registry demonstrates that 14-day survival and neurologic outcome are the same regardless of whether bystanders initiate full basic life support or only do chest compressions. Both are significantly better than if the bystanders only do mouth-to-mouth ventilation or attempt no CPR.19,20

The necessity for ventilation during basic life support has been studied in animal models. Multiple studies have reported that in prolonged fibrillatory cardiac arrest, neurologically intact survival is the same with chest-compression-only resuscitation as with idealized standard CPR (as recommended by the 2000 American Heart Association [AHA] Guidelines), with a 15:2 compression-to-ventilation ratio and when compressions are only interrupted for 4 seconds to provide ventilation.21-23 However, it has been demonstrated that a single lay rescuer interrupts chest compressions for an average of 16 seconds to deliver the 2 recommended mouth-to-mouth ventilations.24 When the 16-second pause for 2 ventilations is incorporated, the 24-hour neurologically intact survival rate in the animal model is reduced. Survival is 13% with a 15:2 compression-to-ventilation ratio and 42% with a 30:2 ratio compared to approximately 70% survival with continuous chest compressions.25,26 The animal studies are supported by a study of CPR technique instructions given by telephone dispatchers to inexperienced bystanders. Instruction in chest compressions alone resulted in the same survival as that of instruction in both compressions and ventilation.27

These observations suggest that when arrest is witnessed and likely to be of cardiac (rather than respiratory) cause, and when intubation will be available within a short time, closed chest compressions alone may be as efficacious as compressions and mouth-to-mouth ventilation. In many venues, basic life support teaching for the lay rescuer is being simplified to eliminate mouth-to-mouth ventilation. Yet the recognition that interrupting chest compressions is harmful and occurs frequently may be more important to the efficacy of intervention than dispensing with mouth-to-mouth. Compressions are interrupted during repeated patient assessment, ventilations, intubation, central line placement, change of rescuers, and defibrillation (especially with automatic external defibrillators [AEDs]). In many resuscitation efforts both in and out of hospital, chest compressions are performed during less than 50% of the attempt. Interrupting compressions is significantly detrimental to maintaining myocardial perfusion, and therefore to the ultimate success of the resuscitation attempt.

The goal of airway management during cardiorespiratory arrest is the same as that during general anesthesia: to provide a clear path for respiratory gas exchange while minimizing gastric insufflation and the risk of pulmonary aspiration. Airway maintenance in an unconscious patient is a basic part of anesthesia practice, and all the techniques learned for its use during anesthesia are applicable for use in the cardiac arrest victim. Just as in the operating room, the most commonly used technique for opening the airway is the "head tilt–chin lift" method. If this is ineffective, the "jaw thrust" maneuver is frequently helpful. Oropharyngeal and nasopharyngeal airways are useful for helping maintain an open airway in patients who are not intubated. Care must be taken to ensure that airways are inserted correctly and do not worsen airway obstruction. Insertion in the semiconscious patient can induce vomiting or laryngospasm.

Effective airway management during CPR is a major problem, even for medical professionals. Many individuals cannot effectively manage a self-inflating resuscitation bag and mask. Larger tidal volumes at lower pressures are delivered by mouth-to-mouth or mouth-to-mask ventilation.28 The bag and mask apparatus is more effective if 2 individuals manage the airway: 1 to hold the mask and maintain the airway and 1 to squeeze the bag.29 A number of advanced airway devices designed for blind placement have been described for use by individuals who are not skilled laryngoscopists. The combination esophageal-tracheal tube (Combitube) and the laryngeal mask airway (LMA) have been studied for use during CPR. Neither of these methods allows the degree of airway control that is obtained with an endotracheal tube. However, intubation has not proven to be superior to the others in promoting successful resuscitation.30 Intubation may be performed in prolonged resuscitations if a skilled laryngoscopist is available. It should not be performed, however, until adequate ventilation by other means (preferably with supplemental oxygen) and circulation by chest compressions have been established and defibrillation performed, if appropriate. When other methods of establishing an airway are unsuccessful, translaryngeal ventilation or tracheotomy by cricothyroid puncture may be lifesaving (Fig. 82-1).

Mouth-to-mouth or mouth-to-nose ventilation is the most expeditious and effective method of immediately available rescue ventilation. Although inspired gas with this method contains approximately 4% carbon dioxide and only approximately 17% oxygen (composition of exhaled air), it is sufficient to maintain viability. Mouth-to-mask ventilation using a simple anesthesia face mask is also very effective. If the mask is fitted with a nipple adapter, a flow of oxygen can also supplement inspired oxygen concentration. The self-inflating resuscitation bag with mask can also be used with or without supplemental oxygen. Recommendations for ventilation are 30 chest compressions followed by 2 breaths during basic life support, as summarized in Box 82-3. Once an advanced airway (endotracheal tube, LMA, Combitube) is in place, ventilation should proceed at 8 to 10 breaths/min without pauses in chest compressions. Do not hyperventilate. Blood flow during CPR slows rapidly when chest compressions are stopped and recovers slowly when they are restarted. Consequently, unless the cause of arrest is hypoxia, emphasis should be placed on maintaining chest compressions with as little interruption as possible.

Physiology of Ventilation during CPR

In the absence of an endotracheal tube, the distribution of gas between the lungs and stomach during mouth-to-mouth or mask ventilation will be determined by the relative impedance to flow into each: to the stomach, by the opening pressure of the esophagus; to the lungs, by lung-thorax compliance. It is likely that during cardiac arrest, esophageal opening pressure is no greater than that found in anesthetized individuals (∼20 cm H2O) and lung-thorax compliance is likely reduced.

Insufflation of air into the stomach during resuscitation leads to gastric distension, impeding ventilation and increasing the danger of regurgitation and gastric rupture. If gastric insufflation is to be avoided, inspiratory airway pressures must be kept low. Partial airway obstruction by the tongue and pharyngeal tissues often leads to increased airway pressures and gastric insufflation. During rescue breathing meticulous attention is necessary to maintain an open airway. To cause a rise in the chest of most adults, a tidal volume of 0.5 to 0.6 L will be needed. Rescue breaths should be given as quickly as possible without causing gastric insufflation.

Physiology of Circulation during Closed Chest Compressions

Two mechanisms of blood flow during closed chest compression have been described.15,31 In the cardiac pump mechanism, the heart is compressed between the sternum and spine, resulting in ejection of blood from the heart into the aorta with the atrioventricular valves preventing backward blood flow. In the thoracic pump mechanism, chest compression raises intrathoracic pressure, forcing blood out of the chest; the venous valves and dynamic venous compression prevent backward flow and the heart acts as a passive conduit. The 2 mechanisms are not mutually exclusive (Fig. 82-2). Fluctuations in intrathoracic pressure play a significant role in blood flow during most resuscitations and the cardiac pump mechanism contributes to blood flow under some circumstances. The predominant mechanism probably varies from victim to victim and even during the course of resuscitating a single victim.

Successful resuscitation in experimental models is associated with myocardial blood flows of 15 to 20 mL/min/100 g (Table 82-2).32 Obtaining such flows requires that closed chest compressions generate adequate cardiac output and coronary perfusion pressure. During CPR, coronary perfusion occurs primarily during the relaxation phase (diastole) of chest compression. In animal models the critical myocardial blood flow is associated with aortic "diastolic" pressure exceeding 40 mm Hg and coronary perfusion pressure (aortic diastolic minus right atrial diastolic pressure) exceeding 25 mm Hg.32-42 One report has confirmed similar findings in humans, noting that all patients with successful return of spontaneous circulation had coronary perfusion pressures higher than 15 mm Hg.42 Critical coronary perfusion pressure develops slowly when chest compressions are begun and is lost quickly when compressions are paused (Fig. 82-3). Again, for successful resuscitation, interruptions of chest compressions must be avoided whenever possible.

During CPR, cardiac output is severely depressed, ranging from 10% to 33% of prearrest values in experimental animals. Nearly all of the cardiac output is directed to organs above the diaphragm. Brain blood flow is 50% to 90% of normal and myocardial blood flow 20% to 50% of normal, and lower extremity and abdominal visceral flow is reduced to less than 5% of normal. Total blood flow tends to decrease with time, but the relative distribution of flow does not change. Flow to the brain and heart are improved by the administration of vasopressors, and flow to organs below the diaphragm is unchanged or further reduced.

Physiology of Gas Transport during CPR

During CPR, measurement of blood gases reveals an arterial respiratory alkalosis and a venous respiratory acidosis because the arterial PCO2 is reduced and the venous PCO2 is elevated. The cause is not respiratory in origin. Rather, these changes result from reduced cardiac output. During the low-flow condition of CPR, excretion of CO2 (milliliters of CO2 per minute in exhaled gas) is decreased to approximately the same extent as is cardiac output. This reduced CO2 excretion is primarily a result of shunting of blood flow away from the lower half of the body. The exhaled CO2 reflects only the metabolism of the part of the body that is being perfused. In the nonperfused areas, CO2 accumulates during CPR. When normal circulation is restored, the accumulated CO2 is washed out and a temporary increase in CO2 excretion is seen.

Although CO2 excretion is reduced during CPR, the mixed venous partial pressure of CO2 (PvCO2) is usually increased.43 Two factors account for this elevation. Buffering acid causes a reduction in serum bicarbonate so that the same blood CO2 content results in a higher PvCO2. In addition, the mixed venous CO2 content is elevated. When flow to a tissue is reduced, not all the produced CO2 is removed, and CO2 accumulates, increasing the tissue partial pressure of CO2. This allows more CO2 to be carried in each aliquot of blood, and mixed venous CO2 content increases. If flow remains constant, a new equilibrium is established where all CO2 produced in the tissue is removed but at a higher venous CO2 content and partial pressure. In contrast to the venous blood, arterial CO2 content and partial pressure (PaCO2) are decreased during CPR. This accounts for most of the observed increase in the arterial–venous CO2 content difference. Although venous blood may have an increased CO2, the marked reduction in cardiac output with maintained ventilation results in very efficient CO2 removal.

Decreased pulmonary blood flow during CPR causes lack of perfusion to many nondependent alveoli. The alveolar gas of these lung units has no CO2. Consequently, the level of mixed alveolar CO2 (ie, end-tidal CO2) will be very low and will correlate poorly with arterial CO2. However, end-tidal CO2 does correlate well with cardiac output during CPR. As flow increases, more alveoli become perfused, there is less alveolar dead space, and end-tidal CO2 measurements rise.

Technique of Closed Chest Compression

Cardiac arrest should be presumed in a patient unresponsive to vigorous stimulation with absent or gasping (agonal) respirations. Even experienced health care providers take too long and have difficulty detecting the presence or absence of a pulse in arrest victims. Therefore, if a pulse check is done before or during rescue efforts, it should not take more than 10 seconds and should not be relied upon to determine successful resuscitation. After that time, if the patient remains unresponsive and apneic, the emergency response system should be activated and CPR begun (Boxes 82-1 and 82-2).

Standard chest compression technique consists of the rhythmic application of pressure over the lower half of the sternum. For compressions to be effective in providing blood flow to the brain and heart, the patient must be on a firm surface with the head level with the heart. The rescuer should stand or kneel at the side of the patient so that the hips are level with the victim's chest. Using the weight of the entire upper body, the compression is delivered straight down with enough force to depress the sternum at least 2 in (5.0 cm) in adults and teens, and to one-third to one-half the depth of the chest in children and infants. Figure 82-4 illustrates the technique of chest compressions in infants. Following maximal compression, pressure is released completely from the chest. Chest compressions should be performed at a rate of 100 per minute. They are most effective if the compression and relaxation phases of the cycle are equal in length. This 50% compression time (compressions to relaxation) is easier to achieve at faster compression rates.

Alternative Techniques of Circulatory Support

In recent years, better understanding of circulatory physiology during CPR, especially involving the thoracic pump mechanism, has resulted in several proposals for alternative techniques or adjunct devices. Most are intended to provide better hemodynamics and extend the duration during which CPR can support viability. Unfortunately, none has proven reliably superior to the standard technique, and no improvement in survival from cardiac arrest has been demonstrated consistently. Initial studies suggested hemodynamic advantages for techniques that raised intrathoracic pressure (such as simultaneous ventilation/compression, abdominal binding with compression, pneumatic antishock garment). Further investigations found that these techniques raised right atrial pressure and intracranial pressure, often more than aortic pressure. Consequently, there was no improvement in cerebral or myocardial blood flow. Outcome studies found no improvement in resuscitation success compared to standard CPR. These techniques are currently not recommended for support of the cardiac arrest victim.

Interposed Abdominal Compression CPR

Interposed abdominal counterpulsation (IAC)-CPR consists of a dedicated rescuer providing manual abdominal compression between the xiphoid and umbilicus during the relaxation phase of conventional CPR.44 IAC-CPR increases venous return and compresses the abdominal aorta to produce retrograde aortic flow, closing the aortic valve and augmenting diastolic pressure. Although initial hemodynamic studies were encouraging, a large randomized study of out-of-hospital arrest found no improvement in survival compared to standard CPR.45 Two subsequent in-hospital studies using trained providers found improved return of spontaneous circulation (ROSC) and short-term survival with IAC-CPR compared to the standard technique.46,47 In general, there has not been improvement in long-term survival with the use of this technique. Reports of human studies demonstrate that the frequency of complications, including laceration of abdominal viscera or esophageal regurgitation, is not increased with IAC-CPR. During in-hospital resuscitation, IAC-CPR may be considered when there are sufficient personnel trained in its use. Its use for out-of-hospital arrest remains experimental.

Active Compression–Decompression CPR

Active compression–decompression (ACD)-CPR developed from the anecdotal report of CPR performed with a plumber's helper applied to the anterior chest wall.48 This suggested that active decompression of the chest wall might reduce intrathoracic pressure during the relaxation phase of chest compressions, leading to improved venous return, increased stroke volume with compression, and better blood flow. Devices that can be applied to the chest wall in order to enable active compression and decompression have been developed, but none is currently approved by the Food and Drug Administration for sale in the United States. Hemodynamic studies in animals and humans show that coronary and cerebral perfusion may be somewhat improved with this method compared to standard CPR, although when epinephrine is used, there is no difference between techniques.49,50 Clinical trials have found mixed results, with some showing short-term benefit but none showing long-term improvement in neurologically intact survival. A meta-analysis of 4162 patients in 10 out-of-hospital trials and 826 patients in 2 in-hospital trials found no difference in early or late survival with ACD-CPR compared to standard CPR.51

The Lund University Cardiac Arrest System (LUCAS) is a gas (oxygen or air) or electric powered piston-type device that produces a consistent chest compression rate of 100/min and a maximum compression depth of 5 cm, and incorporates a suction cup attached to the sternum to return the sternum to the starting position or to cause active decompression.52 Case series have reported variable success with the device.

Impedance Threshold Device

The impedance threshold device (ITD) is a valve that limits the flow of air into the lungs during the relaxation phase of chest compressions, reducing intrathoracic pressure during chest recoil and enhancing venous return.53 Originally designed for use with an endotracheal tube, it has been adapted for use with a face mask, assuming the maintenance of a tight seal with the face. It has frequently been studied in association with ACD-CPR in the belief that the 2 techniques would act synergistically to improve venous return. One meta-analysis using data from randomized trials of both standard CPR and ACD-CPR found an improved short-term outcome with the use of an ITD but no significant improvement in survival to discharge.54 Recently, a randomized trial (ROC PRIMED) involving 11500 patients was stopped for futility because "researchers found that ITD use did not significantly improve or worsen survival rates for cardiac arrest patients" (https://roc.uwctc.org/tiki/tiki-index.php; statement released Nov. 6, 2009).

Pneumatic Vest CPR and Load-Distributing Band CPR

Pneumatic vest CPR is a method of increasing intrathoracic pressure by phasically inflating a bladder around the chest (with or without simultaneous ventilation) yet without significantly changing the dimensions of the chest.55,56 Experimental animal studies show excellent hemodynamics and the ability to maintain viability for prolonged periods. The technique continues to be investigated with a number of modifications from the original method. One modification of this technique is the load-distributing band (LDB) device consisting of a pneumatically or electrically actuated constricting band and backboard.57 A multicenter randomized controlled trial comparing this device against manual CPR demonstrated no improvement in 4-hour survival.58

Invasive Techniques

In contrast to the closed chest techniques, invasive methods have been able to maintain cardiac and cerebral viability during long periods of cardiac arrest. In animal models, open-chest cardiac massage can provide better hemodynamics and myocardial and cerebral perfusion than closed chest techniques.59,60 When performed after 15 minutes of closed-chest CPR, open-chest CPR significantly improves coronary perfusion pressure and the rate of successful resuscitation.61 Furthermore, when initiated early (probably within 20-30 min of arrest) following failure of closed-chest CPR, open-chest CPR may improve resuscitation.62-64 However, if open-chest massage is begun after 30 minutes of ineffective closed-chest compressions, there is no better survival even though hemodynamics are improved.65 This may be a useful technique when the chest or abdomen is already open in the operating room or early in the postoperative period after cardiothoracic surgery.

Preliminary trials of percutaneous cardiopulmonary bypass (through the femoral artery and vein using a membrane oxygenator) for refractory human cardiac arrest have been reported66 and this is a technique available in some institutions. In canines, prompt restoration of blood flow and perfusion pressure with cardiopulmonary bypass can provide resuscitation with minimal neurologic deficit after 20 minutes of fibrillatory cardiac arrest.67

Monitoring during CPR

Assessment of the patient during CPR is similar to that in other clinical situations; Box 82-4 lists the major points. A basic clinical examination and adherence to basic principles, including inspection, palpation, and auscultation of the patient, are performed. The chest is carefully observed for adequacy of expansion with artificial ventilation and for equal and normal breath sounds. In addition, the depth of compression and the position of the rescuer's hands in performing chest compressions should be reevaluated constantly.

Physiologic variables associated with successful resuscitation are listed in Table 82-2. When they are available, these variables should be used to guide resuscitation efforts. The use of an indwelling arterial catheter is an invaluable monitor in assessing the arterial blood pressure and estimating these critical perfusion pressures. In addition, an arterial line allows for the determination of arterial blood gases. If pressures are below the critical levels, adjustments should be made to improve chest compressions and/or additional epinephrine should be administered. Greater pressures do not ensure success. Damage to the myocardium from underlying disease may preclude survival no matter how effective the CPR efforts. However, inadequate vascular pressures consistently result in poor outcomes.

Although invasive pressure monitoring may be the ideal, exhaled end-tidal CO2 is an excellent noninvasive guide to the effectiveness of standard CPR.68 Carbon dioxide excretion during CPR with an endotracheal tube in place is dependent primarily on flow rather than ventilation. Because alveolar dead space is large during low-flow conditions, end-tidal CO2 is very low (frequently <10 mm Hg). If cardiac output increases, more alveoli are perfused and end-tidal CO2 rises (usually to >20 mm Hg during successful CPR). When spontaneous circulation resumes, the earliest sign is a sudden increase in end-tidal CO2 to greater than 40 mm Hg. Within a wide range of cardiac outputs, end-tidal CO2 during CPR correlates with coronary perfusion pressure,69 cardiac output,70 initial resuscitation,71,72 and survival.73 End-tidal CO2 measured during human CPR has been used to predict outcome. No patient with an end-tidal CO2 of less than 10 mm Hg could be successfully resuscitated.73 In the absence of invasive pressure monitoring, end-tidal CO2 monitoring can be used to judge the effectiveness of chest compressions, and quantitative waveform capnography is encouraged for use during all resuscitations. Attempts should be made to maximize the value by alterations in technique or drug therapy. Sodium bicarbonate administration results in the liberation of CO2 in the venous blood and a temporary rise in end-tidal CO2. Therefore, end-tidal CO2 monitoring will not be useful for judging the effectiveness of chest compressions for 3 to 5 minutes following bicarbonate administration.

Duration and Electrical Pattern of Fibrillation

Ventricular fibrillation is a common rhythm disturbance in cardiac arrest. The only consistently effective treatment is electrical defibrillation. The fibrillating heart consumes considerable oxygen, which increases myocardial ischemia and decreases the time to irreversible cell damage. The longer that fibrillation continues, the more difficult it is to defibrillate and the less likely is successful resuscitation.5,62,74-77 During the electrical phase of arrest (within 5 minutes of collapse), defibrillation should be the first priority of resuscitation (Box 82-2). The earlier that defibrillation is accomplished, the better the success of initial resuscitation following out-of-hospital fibrillation and the better the survival to hospital discharge.5,76

The application of defibrillation during the electrical phase has been facilitated by the development of AEDs that can recognize ventricular fibrillation, charge automatically, and give a defibrillatory shock. Minimally trained individuals can incorporate defibrillation into BLS skills using an AED, thus improving survival in out-of-hospital arrest by reducing the time to delivery of the first shock.5,77-80 The value of public access defibrillation was dramatically demonstrated by early data after AEDs were installed in Las Vegas casinos and in Chicago airports.78,79 Results included a 53% survival to hospital discharge78 and a 55% 1-year neurologically intact survival rate over the first 2 years,79 respectively. Early defibrillation is the paramount intervention during the electrical phase of cardiac arrest. However, in the usual out-of-hospital rescue, in which emergency medical technicians (EMTs) or paramedics perform the defibrillation, and in many in-hospital arrests, the first shock frequently is delayed until 6 to 10 minutes following collapse, well into the hemodynamic phase of arrest. A retrospective analysis in Seattle found that when response time was greater than 4 minutes, survival was improved if CPR was provided before defibrillation.81 In a randomized trial of 200 out-of-hospital cardiac arrests in Oslo, a highly significant improvement in outcome resulted if CPR was provided before defibrillation when the response time was greater than 5 minutes.82 Consequently, it now appears that immediate defibrillation is most effective if applied within 4 to 5 minutes of collapse. Otherwise, a brief period of 2 to 3 minutes of chest compressions before defibrillation may improve survival.

The coarseness of the fibrillatory waves on the electrocardiogram (ECG) may reflect the severity and duration of the myocardial insult, and thus have prognostic significance.83 Increasing myocardial ischemia results in less-vigorous fibrillation, reduced-amplitude electrical activity, and more difficult defibrillation. Low-amplitude and low-frequency fibrillatory waveforms are associated with poor outcome.83 Retrospective analysis of VF waveforms suggests that the probability of success of defibrillation can be predicted from the fibrillation waveform and that the shorter the time interval between the last chest compression and shock delivery, the more likely the shock will be successful.84 A reduction of even a few seconds in the interval from pausing compressions to shock delivery can increase the probability of shock success. Catecholamines with β-adrenergic activity, such as epinephrine, increase the amplitude of the electrical activity but have no influence on the ability to defibrillate.75,85 Consequently, defibrillation should not be delayed for drug therapy.

Defibrillators

The modern defibrillator is a variable transformer that stores a direct current in a capacitor until discharged through the electrodes. Until recently, the current waveform of most defibrillators used for transthoracic defibrillation was a monophasic (single direction of current flow between paddle electrodes), damped sinusoid, although some delivered truncated exponential waves. Nearly all AEDs and defibrillators sold today deliver a biphasic current (direction of flow reverses during the shock) in either a truncated, exponential, or rectilinear waveform.86 The output of most defibrillators is indicated in energy units (joules or watt-seconds). At a constant stored energy, the energy delivered to the patient will be inversely related to the impedance (resistance) between the paddle electrodes. For consistency, the energy level indicated on most commercially available defibrillators is the output when discharged into a 50-ohm load.

Transthoracic Impedance

Defibrillation is accomplished by the current passing through a critical mass of myocardium causing simultaneous depolarization of the myofibrils. Even at a constant delivered energy, delivered current will be reduced as impedance increases. At high impedance and relatively low energy levels, current could be too low for defibrillation.

Transthoracic impedance has been measured at 15 to 143 ohms in human defibrillation.87 The major determinants of transthoracic impedance are known, and many are under the control of rescuers (Box 82-5). Impedance decreases with increasing electrode size, but too large a paddle size may diffuse the current over too great an area for effective defibrillation. Handheld paddles and self-adhesive pad electrodes of 8- to 12-cm diameter work well for adults and children and a 4.5-cm diameter is best for infants. The high impedance between metal electrode and skin requires the use of carefully applied self-adhesive pads or electrode paste/gel specifically designed to conduct electricity in the defibrillation setting. Transthoracic impedance decreases with successive shocks, a partial explanation for why an additional shock of the same energy can cause defibrillation when previous shocks have failed. Transthoracic impedance is slightly but significantly higher during inspiration than during exhalation. Air is a poor electrical conductor. Firm paddle pressure of at least 11 kg reduces resistance by improving paddle–skin contact and by expelling air from the lungs.87

Optimum success of defibrillation is obtained by keeping impedance as low as possible. The average transthoracic impedance in human defibrillation is 70 to 80 ohms. Impedance is probably of little clinical significance when reasonably proper technique and high-energy shocks are used. For lower-energy shocks, great care should be taken to minimize resistance. Some defibrillators measure transthoracic impedance during the charge cycle, allowing the use of low-energy shocks in appropriate patients and identification of victims needing higher energy.88 Although not clinically available, current-based defibrillation may be a better way to describe defibrillation dose and may be superior to energy-based defibrillation.89,90

Energy Requirements and Adverse Effects

The incidence and severity of myocardial damage from defibrillation in humans is not clear. In animal studies using the traditional monophasic defibrillator, repeated high-level shocks result in dysrhythmias, ECG changes, and myocardial necrosis.91,92 In humans, dysrhythmia frequency and the degree of ST-segment depression is greater with increased energy doses.93 Although there is no definitive evidence demonstrating myocardial damage within the range of clinically used doses, it would seem prudent to keep energy levels as low as possible during defibrillation attempts. However, if energy is too low, the delivered current may be insufficient for defibrillation, especially when transthoracic impedance is high.

With monophasic defibrillation, there is a general relationship between body size and energy requirements for defibrillation.94 Children need lower energies than adults, perhaps as low as 0.5 J/kg.95 However, over the size range of adults, body size does not seem to be a clinically important variable.96 Multiple studies have demonstrated high rates of successful defibrillation using relatively low levels (160-200 J) of delivered energy. Studies of out-of-hospital and in-hospital arrests have demonstrated equal success when using 200 J or less initial energy compared to administering monophasic shocks at energies of 300 J or greater.93,97

Biphasic defibrillators with different waveforms have been demonstrated to be effective over specific dose ranges. AEDs have the shock energy preselected within the defibrillator algorithm. For manual biphasic defibrillation, most manufacturers display the effective dose range on the defibrillator. For biphasic truncated exponential waveforms, initial energies of 150 to 200 J have been shown to be successful more than 90% of the time.1 For rectilinear biphasic waveforms, initial energy of 120 J has had similar success. Escalating doses of biphasic shocks have not clearly been shown to improve the success of defibrillation. Second and subsequent energy levels should be at least equivalent, and higher energy levels may be considered if available.

The current recommendation in adult patients using a monophasic defibrillator is to give a single shock of 360 J followed by immediate resumption of CPR for 2 minutes before rechecking rhythm and pulse (Box 82-6).1 Because most defibrillations result in asystole or pulseless electrical activity (PEA), a period of CPR will be necessary; beginning compressions immediately following the shock minimizes interruption of critical blood flow. With an AED or biphasic defibrillator, use the preset or manufacturer's recommended energy. If that is unknown or unclear, a dose of 200 J may be selected. In children, with a manual defibrillator (monophasic or biphasic), an initial dose of 2 J/kg is used. As with adults, CPR is immediately restarted after a shock for 2 minutes before repeat rhythm and pulse check. If the first shock is unsuccessful, additional shocks of 4 J/kg are given separated by at least 2 minutes of CPR and appropriate drug and other therapy. If defibrillation attempts are unsuccessful, attention should be directed to the oxygenation, acid-base status, and administration of CPR and drugs such as epinephrine. In addition, factors that may increase transthoracic impedance should be evaluated, including pneumothorax, inadequate paddle–chest wall interface, improper paddle position, excessive distance between paddles, and inadequate paddle pressure on the chest.

Open-chest defibrillation can be used in the operating room when the thorax is already opened during surgery. Appropriate internal paddle size includes a diameter of 6 cm for adults, 4 cm for children, and 2 cm for infants. The paddles are applied with saline-soaked pads, one placed behind the left ventricle and the other over the right ventricle. When defibrillation is performed in this manner, 5 J should be used on the first attempt. If unsuccessful, repeated doses can be used with energy levels of up to 50 J.

Supplemental Therapy

Amiodarone and lidocaine are used during cardiac arrest to aid defibrillation when ventricular fibrillation is refractory to electrical countershock therapy or when fibrillation recurs following successful conversion. However, no antiarrhythmic agent has been shown to be superior to electrical defibrillation or more effective than placebo in the treatment of ventricular fibrillation. Consequently, defibrillation should not be withheld or delayed for drug therapy, but should be applied at the earliest possible time when treating ventricular fibrillation (Box 82-2).

Amiodarone and Lidocaine

Amiodarone is a pharmacologically complex drug with sodium, potassium, calcium, and α- and β-adrenergic blocking properties that is useful for treatment of atrial and ventricular arrhythmias. Amiodarone can cause hypotension and bradycardia when infused too rapidly in patients with an intact circulation. This can usually be prevented by slowing the rate of drug infusion, or treating with fluids, vasopressors, chronotropic agents, or temporary pacing. Lidocaine is primarily an antiectopic agent with few hemodynamic effects. It depresses automaticity by reducing the slope of phase 4 depolarization and the heterogeneity of ventricular refractoriness. Lidocaine tends to restore the ventricular fibrillation threshold reduced by ischemia or infarction. It has no effect on conduction times through the atrioventricular node or on intraventricular conduction time.

Amiodarone is effective in treating multiple types of supraventricular and ventricular dysrhythmias. In contrast, lidocaine is effective in terminating ventricular premature beats (VPBs) and ventricular tachycardia associated with cardiac surgery, acute myocardial infarction, and digitalis intoxication. It is also effective in preventing and treating ventricular dysrhythmias during cardiac catheterization. Lidocaine is not effective in the treatment of atrial or atrioventricular junctional dysrhythmias (Box 82-7).

In cardiac arrest, antiarrhythmic drugs are administered when ventricular tachycardia or ventricular fibrillation have not responded to or have recurred following epinephrine and defibrillation. Two randomized, blinded clinical trials of shock-resistant cardiac arrest victims demonstrated improved admission alive to hospital with amiodarone treatment compared to placebo or lidocaine, although there was no difference in survival to discharge.98,99 Amiodarone is initially administered as a 300-mg rapid infusion. Supplemental infusions of 150 mg can be repeated as necessary for recurrent or resistant arrhythmias to a maximum total daily dose of 2 g.

Lidocaine is an alternative therapy in refractory fibrillation when amiodarone is unavailable, but it has no proven short- or long-term benefits when used during cardiac arrest. To rapidly achieve and maintain therapeutic blood levels during CPR, relatively large doses are necessary. An initial bolus of 1 to 1.5 mg/kg should be given and additional boluses of 0.5 to 0.75 mg/kg can be given every 5 to 10 minutes during CPR up to a total dose of 3 mg/kg. Only bolus dosing should be used during CPR, but an infusion of 2 to 4 mg/min can be started after successful resuscitation.

Temporary Pacemaker Therapy

When emergency pacing is indicated, intravenous or transthoracic pacing can be initiated. Temporary pacing can usually be quickly instituted using external, noninvasive, temporary pacemakers that use large, high-impedance electrodes, allowing relatively nonpainful stimuli.100

Temporary pacing is indicated for patients with preserved myocardial function whose primary problem is impulse formation or conduction, such as patients with severe bradycardia or high-grade heart block who have a palpable pulse. Studies of temporary pacing during cardiac arrest (out-of-hospital or in the emergency department) have found no improvement in admission to hospital or survival. Consequently, withholding chest compressions in patients with asystole to attempt pacing is not recommended.1

Pulseless electrical activity (PEA) and asystole are increasingly seen as the presenting rhythm in cardiac arrest and commonly follow defibrillation. They are seen more often in children and are associated with a poor outcome. PEA is the circumstance where no palpable pulse is present but an ECG rhythm is detectable. This includes a number of different rhythms, most commonly electromechanical dissociation (EMD; Box 82-8). Asystole is the absence of electrical activity on the ECG. Fine ventricular fibrillation can masquerade as asystole if the ECG lead being monitored is located at right angles to the fibrillatory wave.101 For this reason, the trace from a second lead or from a different position of paddle electrodes should always be inspected before a decision is made not to defibrillate.

The predominant cause of primary PEA and asystole is severe myocardial ischemia. These rhythms may represent a failure of myocardial contractility caused by a depletion of intracellular energy stores and they carry a very poor prognosis. However, these ECG patterns also may be secondary, caused by a variety of noncardiac mechanisms that may be amenable to treatment. Therapy directed at the underlying disorder may result in a successful outcome. Consequently, consideration of a noncardiac cause for arrest (frequently referred to as the 5 H's and T's) should always be part of the resuscitation sequence when these rhythms are detected (Box 82-9).

During cardiac arrest, drug therapy is secondary to more fundamental interventions (Box 82-2). Chest compressions and defibrillation (if appropriate) should take precedence over medications. Establishing intravenous (IV) access and administering drugs, although important, should not interrupt sustained chest compressions and ventilation.

Oxygen

Patients in cardiac arrest or low cardiac output states should receive 100% oxygen as soon as possible. Oxygen will increase arterial oxygen tension and hemoglobin saturation if ventilation is supported and will improve tissue oxygenation when circulation is supported. Although exposure to 100% inspired oxygen (FIO2 = 1.0) during CPR and after resuscitation has potential toxicity, there is insufficient evidence to indicate that this occurs during adult CPR.102-104

Vasopressor Agents

Mechanism of Action

Epinephrine has been used in resuscitation since the 1890s and has been the vasopressor of choice in modern CPR since the studies of Redding and Pearson in the 1960s.17,105 Its efficacy lies entirely in its α-adrenergic properties.39 When epinephrine is administered during CPR, peripheral vasoconstriction results in higher aortic pressure, which causes an increase in coronary and cerebral perfusion pressures and myocardial and brain blood flows.38,106,107 Flow to other organs either does not improve or diminishes further when epinephrine is given, despite the increase in aortic pressure. Animal studies demonstrate that all strong α-adrenergic drugs (epinephrine, phenylephrine, methoxamine, dopamine, norepinephrine) and nonadrenergic vasopressors (vasopressin) are equally successful in aiding resuscitation regardless of the β-adrenergic potency. β-Adrenergic agonists without α activity (isoproterenol, dobutamine) are no better than placebo.17,34,105,108,109 α-Adrenergic blockade precludes resuscitation, whereas β-adrenergic blockade has no effect on the ability to restore spontaneous circulation.36,37 Although it has been suggested that the ability of epinephrine to increase the amplitude of ventricular fibrillation (a β-adrenergic effect) makes defibrillation easier, animal studies show that epinephrine does not improve the success of, nor decrease the energy necessary for, defibrillation.75,85 Retrospective clinical studies show no effect of epinephrine on defibrillation success.83

β-Adrenergic stimulation during cardiac arrest is potentially deleterious. In the fibrillating heart, epinephrine increases oxygen consumption. Large doses of epinephrine increase deaths in swine early after resuscitation as a consequence of tachyarrhythmias and hypertension, an effect partially offset by metoprolol treatment.110 Despite these theoretical considerations, survival and neurologic outcome studies show no difference when epinephrine is compared to a pure α-agonist (methoxamine or phenylephrine) or to a nonadrenergic vasopressor (vasopressin) during CPR in animals or humans.111-114 Recently, it has been questioned whether any drug therapy improves outcome from cardiac arrest. One study, comparing various outcomes after out-of-hospital cardiac arrest, was not able to demonstrate any improvements after introduction of advanced life support (epinephrine, atropine, lidocaine).115 Another study demonstrated improvement in short-term outcomes (ROSC and survival to hospital admission) but no difference in survival to discharge, comparing placebo with the use of all drugs (epinephrine, amiodarone, atropine, vasopressin).116 However, this study was not powered to detect clinically meaningful differences in long-term outcome. Neither of the studies was able to isolate outcomes specifically related to individual drug administration. If any effect was found, most drug trials demonstrated only short-term outcome advantage.

Epinephrine

When added to chest compressions, epinephrine helps to develop the critical coronary perfusion pressure necessary to provide enough myocardial blood flow for restoration of spontaneous circulation. If invasive monitoring is present during CPR, an arterial diastolic pressure of 40 mm Hg or coronary perfusion pressure of 20 mm Hg must be obtained with good chest compression technique and/or epinephrine therapy (Table 82-2). In the absence of such monitoring, the dose of epinephrine must be chosen empirically. The standard dose used in animals and humans for many years has been 0.5 to 1.0 mg intravenously. On a weight basis, this dose is approximately 0.1 mg/kg in animals but only 0.015 mg/kg in humans. Animal studies in the 1980s suggested that higher doses of epinephrine in human CPR might improve myocardial and cerebral perfusion and improve success of resuscitation.

Outcome studies prospectively comparing standard and high-dose epinephrine do not demonstrate conclusively that higher doses improve survival. Two randomized, blinded animal studies (1 with fibrillatory arrest and 1 with asphyxial arrest) comparing standard and high-dose epinephrine found no difference in 24-hour survival or neurologic outcome, but more of the high-dose epinephrine animals died in the early postresuscitation period as a result of a hyperdynamic state.110,117 Eight adult prospective randomized clinical trials involving more than 9000 cardiac arrest patients found no improvement in survival to hospital discharge or neurologic outcome, even in subgroups, when initial high-dose epinephrine (5-18 mg) was compared to standard doses (1-2 mg). Some of the studies and a meta-analysis suggest that there may be an improvement in immediate resuscitation with high-dose epinephrine.118

Because of the long experience with epinephrine, it remains the vasopressor of choice in CPR. It should be administered whenever resuscitation has not occurred after adequate chest compressions and ventilation have been started and defibrillation (if appropriate) attempted (Boxes 82-1 and 82-2). High doses of epinephrine apparently are not needed as initial therapy for most cardiac arrests and potentially could be deleterious under some circumstances. Current recommendations are to give intravenous epinephrine, 1 mg in the adult or 0.01 mg/kg in children, every 3 to 5 minutes. Higher doses, given as "rescue" therapy, may be indicated in specific circumstances, or if treatment has been delayed and the standard dose seems ineffective.

Vasopressin

The newest addition to the pharmacologic armamentarium in CPR is arginine vasopressin. It is currently recommended as an alternative to either the first or second dose of epinephrine in a dose of 40 units IV. If additional vasopressor doses are needed, epinephrine should be used. Vasopressin is a naturally occurring (antidiuretic) hormone that, when administered in high doses, is a potent nonadrenergic vasoconstrictor, acting by stimulation of smooth muscle V1 receptors. The half-life in the intact circulation is 10 to 20 minutes and longer than epinephrine during CPR. Animal studies demonstrate that vasopressin is as effective as, or more effective than, epinephrine in maintaining vital organ blood flow during CPR.119-121 Postresuscitation myocardial depression and splanchnic blood flow reduction are more marked with vasopressin than epinephrine, but they are transient and can be treated with low doses of dopamine.122 Three randomized controlled trials and a meta-analysis have found no difference in short- or long-term outcomes with vasopressin compared to epinephrine as a first-line vasopressor in cardiac arrest.114,123-125 Overall, evidence currently suggests that, like other potent vasopressors, vasopressin is equivalent to, but not better than, epinephrine for use during CPR.

Atropine

Atropine sulfate enhances sinus node automaticity and atrioventricular conduction by its vagolytic effects. Atropine is indicated when bradycardia coexists with hypotension, ventricular ectopy, or symptoms associated with myocardial ischemia. The drug can also be used to treat second- and third-degree heart block, and slow idioventricular rates. Although atropine is frequently given during cardiac arrest associated with an ECG pattern of asystole or slow PEA, no studies provide evidence that it actually improves outcome from asystolic or bradysystolic arrest.126,127 The predominant cause of asystole and EMD is severe myocardial ischemia. Excessive parasympathetic tone probably contributes little to these rhythms during cardiac arrest in adults. Even in children, the significance of autonomic tone during arrest is doubtful. The most effective treatment for asystole or PEA is improvement in coronary perfusion and myocardial oxygenation with chest compressions, ventilation, and epinephrine. However, cardiac arrest with these rhythms has a very poor prognosis. Because atropine has few adverse effects, it can be tried in arrest refractory to epinephrine and oxygenation, but its routine use is not recommended. The recommended dose for bradycardia in adults is 0.5 mg IV every 3 to 5 minutes to a total dose of 3.0 mg. The pediatric dose for treating bradycardia is 0.02 mg/kg with a minimum dose of 0.1 mg and a maximum total dose of 1.0 mg in a child and 3.0 mg in an adolescent. The dose may be repeated every 3 to 5 minutes. When treating pulseless arrest in the adult, the dose is 1.0 mg IV every 3 to 5 minutes to a total dose of 3.0 mg. Occasionally, the use of atropine during CPR may cause a sinus tachycardia postresuscitation.

Calcium Salts

With normal cardiovascular physiology, calcium increases myocardial contractility and enhances ventricular automaticity. For many years, consequently, calcium salts have been administered during attempted resuscitation of asystole and EMD. However, multiple clinical studies have found that calcium is no better than placebo in promoting resuscitation and survival from asystole or EMD.128,129 Calcium is not indicated for use during cardiac arrest in adults or children. It may be useful for treatment of hyperkalemia, ionized hypocalcemia, hypermagnesemia, or calcium channel blocker toxicity. If calcium is administered, the chloride salt (2-4 mg/kg) is recommended because it produces higher and more consistent levels of ionized calcium than other salts.

Sodium Bicarbonate

Although in the past sodium bicarbonate was used commonly during CPR, there is little evidence to support its efficacy. Its use during resuscitation was predicated on the adverse cardiovascular consequences of acidosis, including impaired myocardial function, decreased catecholamine responsiveness, and peripheral vasodilatation. However, most studies have been unable to demonstrate improvement in the success of defibrillation or resuscitation with the use of bicarbonate.130,131 The observation that metabolic acidosis develops very slowly during CPR may explain the absence of effect of buffer therapy. Acidosis does not become severe until 15 to 20 minutes of cardiac arrest have passed.43,132,133

Sodium bicarbonate use during CPR should be restricted, not only because of its lack of efficacy, but because of the documented complications from excessive use, including hyperosmolality, hypernatremia, metabolic alkalosis, and hypercapnia from CO2 liberation (Box 82-10).132,134 These abnormalities are associated with low resuscitation rates and poor survival.

Theoretically, sodium bicarbonate could cause a paradoxical worsening of intracellular and intracerebral acidosis as the CO2 liberated during the reaction with acid readily diffuses across cell membranes and the blood-brain barrier, whereas bicarbonate diffuses much more slowly. Direct evidence for this effect has not been found. Animal studies have found no change in spinal fluid acid-base status with clinically relevant doses of sodium bicarbonate135 and no worsening of myocardial intracellular acidosis during bicarbonate administration.136 Consequently, paradoxical acidosis from sodium bicarbonate remains a concern primarily on theoretical grounds.

Current practice suggests that sodium bicarbonate should be considered for use during CPR only in arrests associated with hyperkalemia, severe preexisting metabolic acidosis, and tricyclic or phenobarbital overdose. It may be considered for use in protracted resuscitation attempts after other modalities have been instituted. These recommendations stem from its unproven efficacy in increasing patient survival and its known side effects. When bicarbonate is used, 1 mEq/kg should be given as the initial dose and no more than half this dose given every 10 minutes thereafter. However, dosing of sodium bicarbonate should be guided by blood gas determination, whenever possible.

Routes of Administration and Vascular Access

The preferred route of administration of all drugs during CPR is intravenous or intraosseous. If one of these routes cannot be established rapidly because of technical difficulties, one of the other alternatives—endotracheal or intracardiac—can be used.

Intravenous Access

The most rapidly reached and highest drug levels occur with administration into a central vein. Therefore, when a central venous catheter is available during CPR, as is often the case in the operating room, it should be used for drug therapy. However, peripheral intravenous administration is also effective. The antecubital or external jugular vein should be the site of first choice for starting an infusion during resuscitation because starting a central line usually necessitates stopping CPR. Sites in the upper extremity and neck are preferred because of the paucity of blood flow below the diaphragm during CPR. Drugs administered in the lower extremity may be extremely delayed or not reach the sites of action. Even in the upper extremity, drugs may require 1 to 2 minutes to reach the central circulation.137 Onset of action may be speeded if a peripheral drug bolus is followed by a 20 to 30-mL bolus of intravenous fluid.

Intraosseous Administration

The intraosseous route of fluid and medication administration, originally described in 1934,138 has regained popularity recently for emergency vascular access. During CPR, all medications and fluids used, including whole blood, have been given by the intraosseous route. This technique should be considered a temporary measure during emergencies when other vascular sites are not available.

The technique of placing an intraosseous line is straightforward. A standard 16- or 18-gauge needle, spinal needle with stylet, or bone marrow needle is inserted into the anterior surface of the tibia, 1 to 3 cm below the tibial tuberosity. Commercially available kits can facilitate access in adults. The needle is directed to a 90° angle to the medial surface of the bone or slightly inferior to avoid the epiphyseal plate. There is a loss of resistance after the needle passes through the bony cortex of the tibia. Placement is successful if the needle is in the marrow cavity, as evidenced by its standing upright without support, and bone marrow can be aspirated into a syringe connected to the needle. The needle will lose the upright position if it has slipped into the subcutaneous tissue. Free flow of the drug or fluid infusion without significant subcutaneous infiltration also should be demonstrated (Fig. 82-5).139 There are a number of reports of successful intraosseous infusions in children, with minimal complications. Animal models have demonstrated successful reversal of hemorrhagic shock,140 effective buffering with sodium bicarbonate,141 and equal hemodynamic response to epinephrine with intraosseous and central intravenous administration.142

Endotracheal Administration

Drugs that can be absorbed from tracheal mucosa include lidocaine, atropine, naloxone, epinephrine, and vasopressin (Box 82-11). During CPR the rapid establishment of vascular access can be difficult, especially in the obese or in infants and small children. The endotracheal route can be used as an alternative following intubation in these settings. However, intravenous or intraosseous drug administration is the preferred route because the time to effect and the drug levels achieved are inconsistent using the endotracheal route.143-145 Better results may be obtained by administering 5- to 10-mL volumes. It is unclear whether deep injection is better than simple instillation into the endotracheal tube.32 Doses 2 to 2.5 times higher than the recommended IV dose should be administered when this route is used.

In the past, the lack of optimal postresuscitation care has contributed significantly to the poor outcome from cardiac arrest. There is growing awareness that aggressive postresuscitation care can significantly improve survival. Out-of-hospital cardiac arrest victims should be transported to a facility with a comprehensive postresuscitation system of care that includes acute coronary interventions, neurologic care, goal-directed critical care, and the institution of therapeutic hypothermia. In-hospital victims should be moved to a critical care unit that provides comprehensive postresuscitation care. The precipitating cause for the arrest should be sought (review the 5 H's and T's; Box 82-9). Because cardiovascular disease and coronary ischemia are the most common causes of cardiac arrest, the search for acute coronary syndromes with possible coronary reperfusion should be a high priority. Coronary interventions can be performed while efforts to stabilize hemodynamics and institute therapeutic hypothermia are ongoing.

The major factors contributing to mortality following successful resuscitation are progression of the primary disease and cerebral damage suffered as a result of the arrest. Any cardiac arrest, even of brief duration, causes a generalized decrease in myocardial function similar to the regional hypokinesis seen following periods of regional ischemia. This is usually referred to as global myocardial stunning and can be mitigated with inotropic agents, if necessary. Active management following resuscitation appears to mitigate postischemic brain damage and improve neurologic outcome without increasing the number of patients surviving in a vegetative state.146 When flow is restored following a period of global brain ischemia, 3 stages of cerebral reperfusion are seen in the ensuing 12 hours. Immediately following resuscitation there are multifocal areas of the brain with no-reflow. Within an hour there is global hyperemia followed quickly by prolonged global hypoperfusion.

Postresuscitation support is focused on providing stable oxygenation and hemodynamics so as to minimize any further cerebral insult and to detect and treat any other organ failure. A comatose patient should have hypothermia instituted and be maintained on mechanical ventilation for 24 hours to ensure adequate oxygenation and ventilation. Restlessness, coughing, or seizure activity should be aggressively treated with appropriate medications including neuromuscular blockers, if necessary. Oxygen free radicals contribute significantly to brain reperfusion injury, which may be worsened by hyperoxia.102-104 Although 100% oxygen is appropriate for use during initial resuscitation, postresuscitation oxygen therapy should be titrated to the lowest level to maintain arterial oxygen saturation above 94%. Hyperventilation should be avoided (PaCO2 35-40 mm Hg). Blood volume should be maintained at normal, and moderate hemodilution to a hematocrit of 30% to 35% may be helpful. A brief 5-minute period of hypertension to mean arterial pressure of 120 to 140 mm Hg may help overcome the initial cerebral no-reflow. This frequently occurs secondary to the effects of epinephrine given during CPR. However, both prolonged hypertension (>110 mm Hg) and hypotension are associated with a worsened outcome. Hyperglycemia during cerebral ischemia is known to result in increased neurologic damage. Thus it seems prudent to control glucose in the 100 to 180 mg/dL range. Specific pharmacologic therapy directed at brain preservation has not been shown to have further benefit. Although animal trials of barbiturates and calcium channel blockers were promising, large, multicenter trials found no improvement in neurologic status when drugs were given following cardiac arrest.146,147

In contrast to pharmacologic therapy, 2 studies have demonstrated improved neurologic outcome when mild therapeutic hypothermia (89.6°F-93.2°F [32°C-34°C]) was induced for 12 to 24 hours in cardiac arrest survivors who remained comatose after admission to the hospital.148,149 Both investigations studied only patients whose initial rhythm was ventricular fibrillation, and the larger of the trials included only witnessed arrests. Nevertheless, these are the first studies to document improved neurologic outcome with a specific postarrest intervention. The International Liaison Committee on Resuscitation (ILCOR) now recommends that "unconscious adult patients with return of spontaneous circulation after out-of-hospital cardiac arrest should be cooled to 32°C to 34°C for 12 to 24 hours when the initial rhythm was ventricular fibrillation. Such cooling may also be beneficial for other rhythms or in-hospital cardiac arrest."

For the comatose survivor of CPR, the question of ultimate prognosis is important. Neurologic prognosis may be difficult to determine during the first 72 hours in patients not undergoing hypothermia, and this time frame is likely extended in patients being cooled.150 Many initially comatose survivors of cardiac arrest have the potential for full recovery. A neurologic exam for prognosis must not be undertaken if confounding factors (hypothermia, sedatives, neuromuscular blockers, seizures) exist. Even in patients not treated with hypothermia, no clinical signs reliably predict poor outcome less than 24 hours after cardiac arrest. Most patients who completely recover show rapid improvement in the first 48 hours. One older study found that if a patient is not awake by 4 days following arrest, the chance of ever awakening is less than 20%, and all those patients awakening had marked neurologic deficits.151 Summarized in Box 82-12 are the clinical signs that strongly predict death or poor neurologic outcome in comatose patients not treated with hypothermia.152-154 The time frame for similar assessment in patients treated with hypothermia is unknown.

• Emergency Cardiac Care Committee, Subcommittees and Task Forces, American Heart Association: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2010;122:S640-S946.
• Weisfeldt ML, Becker LB. Resuscitation after cardiac arrest: a 3-phase time-sensitive model. JAMA. 2002;288:3035.
• Berg RA, Kern KB, Sanders AB, et al. Bystander cardiopulmonary resuscitation: is ventilation necessary? Circulation. 1993;88:1907.
• Ewy GA, Zuercher M, Hilwig RW, et al. Improved neurological outcome with continuous chest compressions compared with 30:2 compressions-to-ventilations cardiopulmonary resuscitation in a realistic swine model of out-of-hospital cardiac arrest. Circulation. 2007;116:2525.
• Ralston SH, Voorhees WD, Babbs CF. Intrapulmonary epinephrine during prolonged CPR. Improved regional blood flow and resuscitation in dogs. Ann Emerg Med. 1984;13:79.
• Otto CW, Yakaitis RW, Blitt CD. Mechanism of action of epinephrine in resuscitation from asphyxial arrest. Crit Care Med. 1981;9:321.
• Sanders AB, Kern KB, Otto CW, et al. End-tidal carbon dioxide monitoring during cardiopulmonary resuscitation: a prognostic indicator for survival. JAMA. 1989;262:1347.
• Valenzulea TD, Roe DJ, Nichol G, et al. Outcomes of rapid defibrillation by security officers after cardiac arrest in casinos. N Engl J Med. 2000;343:1206.
• Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic cardiopulmonary resuscitation to patients with out-of-hospital ventricular fibrillation. JAMA. 2003;289:1389.
• Kudenchuk PJ, Cobb LA, Copass MK, et al. Amiodarone for resuscitation after out of hospital cardiac arrest due to ventricular fibrillation. N Engl J Med. 1999;341:871.
• Wenzel V, Krismer AC, Arntz HR, et al. A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation. N Engl J Med. 2004;350:105.
• Olasveengen TM, Sunde K, Brunborg C, et al. Intravenous drug administration during out-of-hospital cardiac arrest: a randomized trial. JAMA. 2009;302:2222.
• Holzer M, The Hypothermia After Cardiac Arrest Study Group. Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346:549.
• Booth CM, Boone RH, Tomlinson G, et al. Is this patient dead, vegetative, or severely neurologically impaired? Assessing outcome for comatose survivors of cardiac arrest. JAMA. 2004;291:870.