Physiologic foundations of cardiopulmonary resuscitation


  • Both the cardiac and thoracic pump mechanisms play a role in infants and children during cardiopulmonary resuscitation. Thus, attention to excellent chest compression technique—with an emphasis on “push hard, push fast”—is critical to attaining sufficient cardiac output to maintain coronary and cerebral blood flow.

  • Use of any vasoconstrictor should be sufficient to raise aortic diastolic pressure during cardiopulmonary resuscitation above the critical level for resuscitation success (>15–20 mm Hg).

  • Amiodarone or lidocaine may be the most effective pharmacologic treatments for shock-resistant ventricular tachycardia or fibrillation.

  • Use of the biphasic defibrillator is an important advance in the treatment of tachyarrhythmias and has advantages in its safety profile compared with monophasic defibrillators.

  • Resuscitative care following cardiac arrest is critical to survival and includes appropriate uses of inodilators and neuroprotective strategies, including avoidance of hyperthermia.

With the development of basic cardiopulmonary resuscitation (CPR) in the early 1960s, skilled resuscitation teams both in and out of the hospital were formed. The development of CPR saved lives; previously, every victim of cardiac arrest had died. Soon thereafter, successful resuscitation of patients by basic life support measures, defibrillation, and medications became common even as long as 5 hours after commencement of CPR. Over the past decade, both survival and neurologic outcomes after in-hospital cardiac arrest have improved in adults and children. Data show that the success of CPR depends on many factors. Rapid institution of basic life support measures (i.e., bystander CPR for sudden out-of-hospital cardiac arrest and immediate electrical countershock for ventricular fibrillation [VF]) improve the chances of survival for patients experiencing sudden out-of-hospital cardiac arrest. These measures led to the growing deployment of automatic external defibrillators (AEDs) in public places. Immediate defibrillation is currently the standard of care in witnessed VF arrests. However, evidence indicates that basic life support and other measures directed at restoring energy substrates to the myocardium before countershock in patients with unwitnessed, out-of-hospital arrest may further improve outcome.

Other preexisting factors that play a role in successful resuscitation include the patient’s age, prior medical condition, presenting cardiac rhythm, and the etiology of cardiac arrest. In 2008, a multi-institutional prospective study was published that examined these preexisting factors and further described in two additional studies the clinical characteristics, hospital course, and outcomes of a cohort of children after in-hospital or out-of-hospital arrest. In addition to demonstrating differences in clinical characteristics, these studies offered future considerations for the care of children who had experienced cardiac arrest and postresuscitative care, including hypothermia. The low resuscitation rate in children, even when the patient does not have preexisting disease, probably results from the high incidence of asystole as the presenting rhythm. Asystole is the most common presenting rhythm in both in-hospital and out-of-hospital arrests, noted in 25% to 70% of victims. Bradycardia and pulseless electrical activity (PEA) are other common rhythms. The high incidence of asystole in children who experience cardiac arrest can be explained by systemic disturbances—such as hypoxia, acidosis, sepsis, and hypovolemia—that commonly precede the arrest. Although ventricular arrhythmias usually are reported to be infrequent (range, 1.3%–3.8%), out-of-hospital series report VF in 10% to 19% of victims younger than 20 years. , These series, along with the observation that the frequency of witnessed arrest is much lower than in adults, suggests that ventricular rhythms may be more common than usually estimated and that delay in resuscitation results in progression of nonperfusing rhythms to asystole. Increasing availability of AEDs may be contributing to the increased recognition of ventricular arrhythmias in out-of-hospital pediatric cardiac arrest. In specialized cardiac intensive care units (ICUs), ventricular arrhythmias account for as many as 30% of the arrests. , ,

In their original work on CPR, Kouwenhoven et al. proposed that blood flow during closed-chest compressions resulted from squeezing of the heart between the sternum and vertebral column, now termed the cardiac blood flow mechanism . In fact, the precise mechanism by which forward circulatory flow is generated during closed-chest cardiac massage has major implications for current approaches to CPR. Other methods—such as vest CPR, active compression-decompression CPR (ACD-CPR) both without and with an impedance threshold valve (ITV), and interposed abdominal compressions with CPR (IAC-CPR)—take into account advances in our understanding of the mechanism of blood flow during resuscitation.

The pharmacology of resuscitation remains controversial; these controversies have led to major changes in the guidelines for CPR. Use of sodium bicarbonate, calcium chloride, and glucose remains unresolved at this time. The role of high-dose epinephrine has been minimized because of concerns over postresuscitation deleterious effects on myocardial performance and poor outcomes. Evidence for a role for vasopressin, with a relatively pure vasoconstrictor effect, is accumulating. Data have been accumulating for the use of amiodarone or lidocaine as the antiarrhythmics of choice for ventricular ectopy in persons in cardiac arrest. Research is ongoing into alternative vasoconstrictors and the use of pharmacologic cocktails that may include β-blockers, antiarrhythmic agents, antioxidants, nitroglycerin, and a vasoconstrictor in attempts to improve the resuscitation outcome and postresuscitation cardiac function.

Developments in the use of direct current countershock have occurred. Biphasic defibrillators are now widely in use and appear to improve the success of defibrillation at lower delivered energies. It is hoped that they decrease myocardial injury. As noted, the role of “shock first” is being reassessed because the success of electrical countershock in restoring spontaneous circulation declines rapidly after 3 to 4 minutes have elapsed.

Although postresuscitation cerebral preservation has become an important area of focus, therapeutic hypothermia has not been found to improve neurologic outcome after pediatric cardiac arrest. ,

This chapter discusses the physiologic foundations of CPR. In the first section, the possible mechanisms of blood flow by the thoracic and cardiac pump mechanisms are discussed, including how the specific chest geometry of children and infants helps decide which of these mechanisms applies. Newer CPR techniques, which consider the physiologic mechanisms discussed in the first section, are then discussed. Controversies and advances in pharmacologic management during CPR and current guidelines for use of drugs for resuscitation are addressed. New developments in the use of countershock—including the timing of shocks, energy used, and type of current delivery system used (biphasic or monophasic)—are discussed. Finally, the role of therapeutic hypothermia is reviewed.

Mechanisms of blood flow

Cardiac versus thoracic pump mechanism

The cardiac pump hypothesis holds that blood flow is generated during closed-chest compressions when the heart is squeezed between the sternum and vertebral column. This mechanism of flow implies that ventricular compression causes closure of the atrioventricular valves and that ejection of blood reduces ventricular volume. During chest relaxation, ventricular pressure falls below atrial pressure, allowing the atrioventricular valves to open and the ventricles to fill. This sequence of events resembles the normal cardiac cycle and occurs during cardiac compression when open-chest CPR is used.

Numerous clinical observations have conflicted with the cardiac pump hypothesis of blood flow. In 1964, Mackenzie et al. found that closed-chest CPR produced similar elevations in arterial and venous intravascular pressures, the result of a generalized increase in intrathoracic pressure. In 1976, Criley et al. made the dramatic observation that several patients in whom VF developed during cardiac catheterization produced enough blood flow to maintain consciousness by repetitive coughing. The production of blood flow by increasing thoracic pressure without direct cardiac compression describes the thoracic pump mechanism of blood flow during CPR.

During normal cardiac function, the lowest pressure in the vascular circuit occurs on the atrial side of the atrioventricular valves. This low-pressure compartment is the downstream pressure for the systemic circulation, which allows venous return to the heart. Angiographic studies show that blood passes from the vena cava through the right heart into the pulmonary artery and from the pulmonary veins through the left heart into the aorta during a single chest compression.

Echocardiographic studies show that, unlike normal cardiac activity or during open-chest CPR, during closed-chest CPR in both dogs and humans, the atrioventricular valves are open during blood ejection and aortic diameter decreases rather than increases during blood ejection. , These findings during closed-chest CPR support the thoracic pump theory and argue that the heart is a passive conduit for blood flow ( Fig. 38.1 ).

• Fig. 38.1

Possible mechanisms for blood flow during cardiopulmonary resuscitation include direct cardiac compression (left) and the thoracic pump (right). With direct cardiac compression, an increase in chest compression rate causes an increase in blood flow by squeezing the heart between the vertebral column and sternum. With the thoracic pump mechanism, factors that increase pleural pressure cause an increase in pressure within the heart chambers and, ultimately, an increase in blood flow.

(Modified from Schleien CL et al. Controversial issues in cardiopulmonary resuscitation, Anesthesiology. 1989;71:135.)

Initial measurements of hemodynamic data during chest compression for CPR found the generation of almost equal pressures in the left ventricle, aorta, right atrium, pulmonary artery, and esophagus. The finding that all intrathoracic vascular pressures are equal implies that suprathoracic arterial pressures must be higher than suprathoracic venous pressures. The unequal transmission of intrathoracic pressure to the suprathoracic vasculature establishes the gradient necessary for blood flow. The transmission of intrathoracic pressure to the suprathoracic veins may be modulated by venous valves. The presence of these jugular venous valves has been demonstrated in animals and humans undergoing CPR. An ultrasonography study of healthy children confirmed the presence of these valves in 84% of 239 jugular veins studied. The valves were bilateral in 74% of children. Transmission of intrathoracic pressure to the intracranial vault during CPR indicates that any such valve function is partial. Pathologic studies have also identified valves in the subclavian vein in the large majority of cadavers studied (87%). The absence of these valves in some patients is postulated to lead to failure of closed-chest CPR.

Subsequent hemodynamic and echocardiographic studies found different results. Deshmukh et al. demonstrated in a porcine model that mitral valve function persisted throughout resuscitation in 17 of 22 animals and that in successfully resuscitated animals, maximal aortic pressure exceeded that in the right atrium throughout the resuscitation. In another porcine model of resuscitation, Hackl et al. manipulated the compressive force and depth of resuscitation by using a mechanical resuscitator. The frequency of mitral valve closure during compressive systole was directly proportional to the force and depth of chest compression. When the depth of compression reached 25% of the anteroposterior diameter, valve closure occurred in 95% of cycles. They concluded that the mechanism of blood flow was dependent on the force and depth of compression. In a study of CPR using transesophageal Doppler echocardiography in adults, Porter et al. demonstrated mitral valve closure in compressive systole in the majority of patients (12 of 17) but not all patients. Peak mitral flow occurred in diastole and was significantly higher in the group with mitral valve closure. Peak mitral flow occurred during compressive systole in those without valve closure. Left ventricular (LV) fractional shortening correlated with change in anteroposterior chest wall diameter and not mitral valve flow. These authors concluded that nonuniform increased intrathoracic pressure plays a role in determining whether valve closure occurs during chest compressions. As noted, a decrease in aortic dimension during CPR has been demo nstrated by echocardiography and taken as evidence for the thoracic pump mechanism of blood flow. Hwang et al. readdressed this issue using transesophageal echocardiography. They studied the aortic dimension of the proximal and distal thoracic aorta and noted a decrease in the aortic dimension in the distal aorta directly inferior to the zone of direct compression and an increase in the dimension of the proximal aorta. They also noted mitral valve closure in all subjects and a decrease in LV volume of almost 50% at end compression. These findings were believed to be most consistent with the cardiac pump mechanism of blood flow. Kim et al. also used transesophageal echocardiography to explore the role of the left ventricle during nontraumatic arrests. They noted that during the compression phase of CPR, there was anterograde flow from the ventricle to the aorta and retrograde flow toward the mitral valve. The mitral valve remained closed during compression and open during relaxation, while the aortic valve remained open during compression and closed during relaxation, which they concluded to be consistent with the cardiac pump mechanism.

The cardiac pump mechanism appears to predominate during closed-chest CPR in specific clinical situations. As noted, increasing the applied force during chest compressions increases the likelihood of direct cardiac compression. A smaller chest size may allow for more direct cardiac compression. , Adult dogs with small chests have better hemodynamics during closed-chest CPR than do dogs with large chests. Because the infant chest is smaller and more compliant than the adult chest, direct compression of the heart during CPR is more likely to occur. Blood flow during closed-chest CPR in a piglet model of cardiac arrest is higher than that achieved in adult models. A study of 20 randomized swine to either a patient-centric blood pressure targeted approach with titration of compression depth to a systolic blood pressure of 100 mm Hg and vasopressors to a coronary perfusion pressure greater than 20 mm Hg or current usual practice, the blood pressure targeted group demonstrated improved 24-hour survival (8 of 10 vs 0 of 10 survival; P = .001). This study suggests that physiologic targets rather than absolute depths by age may in fact confer better outcomes.

Rate and duty cycle

In 2015, the American Heart Association (AHA) recommended a rate of chest compressions of at least 100 per minute. At faster rates, blood flow is enhanced whether the thoracic pump mechanism or cardiac pump mechanism is invoked. Duty cycle is defined as the ratio of the duration of the compression phase to the entire compression-relaxation cycle, expressed as a percent. For example, at a rate of 30 compressions/min, a 1.2-second compression time produces a 60% duty cycle. If blood flow is generated by direct cardiac compression, then the stroke volume is determined primarily by the force of compression. Prolonging the compression (increasing the duty cycle) beyond the time necessary for full ventricular ejection should have no additional effect on stroke volume. Increasing the rate of compressions should increase cardiac output because a fixed, relatively small volume of blood is ejected with each cardiac compression. In contrast, if blood flow is produced by the thoracic pump mechanism, the volume of blood to be ejected comes from a large reservoir of blood contained within the capacitance vessels in the chest. With the thoracic pump mechanism, flow is enhanced by increasing either the force of compression or the duty cycle but is not affected by changes in compression rate over a wide range of rates. Additionally, the “push hard, push fast” recommendation is based on the maintenance of a higher compression rate with a higher force of compression. Allowing total recoil of the chest allows for full blood return during the relaxation phase of the cycle.

It appears from experimental animal data that both the thoracic pump and cardiac pump mechanisms can effectively generate blood flow during closed-chest CPR. Differences between various studies may be attributed to differences in animal models or compression techniques. Important differences in animal models include chest wall geometry, compliance and elastic recoil, compliance of the diaphragm, and intraabdominal pressure. Differences in technique include the magnitude of sternal displacement; compression force; and momentum of chest compression, compression rate, and duty cycle. Experimental and clinical data support both mechanisms of blood flow during CPR in human infants.

Results of several studies in dogs demonstrated a benefit of a compression rate of 120 per minute compared with slower rates during conventional CPR. , In studies of piglets, puppies, and humans, no differences were found comparing different rates of compression during conventional CPR. , In a study of piglet CPR, duty cycle was the major determinant of cerebral perfusion pressure (CPP). The duty cycle at which venous return became limited varied with age. A longer duty cycle was more effective in younger piglets. In a more recent study of 22 pigs randomized to head-up tilt versus supine CPR using an automated CPR device plus an impedance threshold device after VF arrest, CPP and ICP improved substantially with head-up tilt position. This suggests that gravity has an effect on the venous circulation with high-quality CPR and head tilted up positioning on cerebral perfusion. The discrepant importance of rate and duty cycle in various models (by different investigators) is confusing; however, increasing the rate of compressions during conventional CPR to 100 per minute satisfies both those who prefer the faster rates and those who support a longer duty cycle.

Appropriate chest compression rate, depth, and fraction are still being investigated. Niles et al. characterized these in-hospital CPR metrics and compliance to the 2015 AHA guidelines by evaluating the pediRES-Q database. They found that meeting all three of these AHA-derived CPR targets was extremely difficult to accomplish. Their study was not powered to address short- or long-term survival but was an important step in evaluating the optimal cutoffs for survival.

Chest geometry

Chest geometry plays an important role in the ability of extrathoracic compressions to generate intrathoracic pressure. Shape, compliance, and deformability, which change greatly with age, are the chest characteristics that have the greatest impact during CPR.

The change in cross-sectional area of the chest during anterior to posterior delivered compressions is related to its shape ( Fig. 38.2 ). The ratio of the chest anteroposterior diameter to the lateral diameter is referred to as the thoracic index . A keel-shaped chest, as seen in an adult dog, has a greater anteroposterior diameter and, thus, a thoracic index greater than 1. A flat chest, as in a thin human, has a greater lateral diameter and, thus, a thoracic index less than 1. A circular chest has a thoracic index equal to 1. A circle has a larger cross-sectional area than either of these elliptical chests. As an anteroposterior compression flattens a circle, the cross-sectional area decreases and compresses its contents. In contrast, as an anteroposterior compression is applied to the keel-shaped chest, the cross-sectional area increases as a circular shape is approached. The cross-sectional area of the keel-shaped chest does not decrease until the chest compression continues past the circular shape to flatten the chest. This implies a threshold past which the compression must proceed before intrathoracic contents are decreased and squeezed. Thus, the rounder, flatter chests of small dogs and pigs may require less chest displacement than the keel-shaped chests of adult dogs to generate thoracic ejection of blood. This dynamic has been demonstrated in small dogs having round chests compared with adult dogs having keel-shaped chests.

• Fig. 38.2

Changes in area of ellipses with constant circumference. Each ellipse is labeled with the anteroposterior (ap) and lateral (lat) radii, and a 20% anteroposterior compression is applied. Indicated change in area equals relaxed area – compressed area. (A) Initial anteroposterior/lateral ratio = 0.7, and compression leads to positive ejection because relaxed area – compressed area is negative. (B) Initial anteroposterior/lateral ratio = 1.4, and compression toward a circular shape results in an increase in area.

(Modified from Dean JM, Koehler RC, Schleien CL, et al. Age-related changes in chest geometry during cardiopulmonary resuscitation. J Appl Physiol . 1987;62[6]:2212–2219.)

As humans age, the cartilage of the rib cage calcifies and chest wall compliance decreases. Older patients may require greater compression force to generate the same sternal displacement. A 3-month-old piglet requires a much greater compression force for anteroposterior displacement than its 1-month-old counterpart. Direct cardiac compression is more likely to occur in the more compliant chest of younger animals. Cerebral and myocardial blood flow during closed-chest CPR was much higher in infant piglets than in adults ( Figs. 38.3 and 38.4 ). This finding supports the cardiac pump mechanism of blood flow in infants because the level of organ blood flow achieved during closed-chest CPR in piglets approaches the level achieved during open-chest cardiac massage in adults.

• Fig. 38.3

Total cerebral blood flow, cerebral fractional oxygen (O 2 ) extraction, and cerebral O 2 uptake before cardiac arrest and during 50 minutes of cardiopulmonary resuscitation in the groups with and without epinephrine.

• Fig. 38.4

Top, Total myocardial blood flow during cardiopulmonary resuscitation (CPR) in piglets with and without epinephrine. Asterisk indicates significant difference between groups at 5 and 20 minutes. Bottom, Blood flow to right ventricular free wall (RV, circle ), left ventricular free wall (LV, squares ), and interventricular septum (triangles) in the groups with and without epinephrine. Standard error bars are omitted for clarity, but the least significant difference bar ( LSD, derived from Duncan multiple-range test) is shown for comparisons among heart regions within an animal group. (Means must differ by height of bar for P < .05.) LSD for comparing means between groups is twice that shown for within-group LSD. Asterisk indicates that RV blood flow was greater than LV and septal blood flows at 5 minutes in the group without epinephrine. Flows in all three regions in the group with epinephrine were greater than those in the respective regions in the group without epinephrine at 5 and 20 minutes.

(Modified from Schleien CL, Dean JM, Koehler RC, et al. Effect of epinephrine on cerebral and myocardial perfusion in an infant animal preparation of cardiopulmonary resuscitation. Circulation . 1986;73[4]:809–817.)

Marked deformation of the chest can occur during prolonged CPR and may alter the effectiveness of CPR ( Fig. 38.5 ). Over time, the chest assumes a flatter shape, producing a larger percent decrease in cross-sectional area at the same absolute chest displacement. Progressive deformation may be beneficial if it leads to more direct cardiac compression. Unfortunately, too much deformation may decrease the recoil of the chest wall during the relaxation phase, leading to decreased cardiac filling. A progressive decrease in the effectiveness of chest compressions to produce blood flow is seen in piglets receiving conventional CPR. Permanent deformation of the chest in this model approaches 30% of the original anteroposterior diameter. Using a thoracic vest to limit deformation when performing CPR greatly decreased permanent chest deformation (3% vs. 30%) but did not attenuate the deterioration of vital organ blood flow with time.

• Fig. 38.5

Piston position during chest compression and relaxation phases of the cycle, and net piston displacement expressed as a percent of prearrest anteroposterior chest diameter (12.0 ± 0.3 cm) in piglets. Note that displacement was essentially unchanged over the 50-minute duration, but marked deformation occurred during the relaxation phase by 5 minutes and continued to further deform over the 50-minute period in the groups with or without epinephrine. CPR, Cardiopulmonary resuscitation.

(Modified from Schleien CL, Dean JM, Koehler RC, et al. Effect of epinephrine on cerebral and myocardial perfusion in an infant animal preparation of cardiopulmonary resuscitation. Circulation . 1986;73[4]:809–817.)

The characteristics of chest geometry of animals may relate to that in humans. Body weight, surface area, chest circumference, and diameter did not correlate with the magnitude of aortic pressure produced during CPR in a study of nine adults already declared dead. A direct comparison of adult and pediatric human CPR has not been performed. The higher intravascular pressures and organ blood flow during CPR in infants compared with adults may result from more effective transmission of the force of chest compression because of the higher compliance and greater deformability of the infant chest.

Effects of cardiopulmonary resuscitation on intracranial pressure

When chest compressions are applied, the increase in intrathoracic pressure is transmitted through the venous system of the head and neck to the intracranial vault, resulting in an increased intracranial pressure (ICP). Pressure is transmitted via the paravertebral veins and the cerebrospinal fluid during CPR in dogs. Large swings in ICP corresponding to chest compressions occur in children undergoing CPR. This transmission of intrathoracic pressure to the intracranial contents accounts for the low CPP by increasing the downstream pressure and cerebral blood flow during closed-chest CPR. However, in a porcine model of CPR with an impedance threshold device, it was found that CPR done in a reverse Trendelenburg position (head up at 30 degrees) reduced ICP and improved cerebral perfusion, likely because gravity improved venous drainage and, thus, reduced impedance to forward flow.

The relationship of ICP to intrathoracic pressure during CPR is linear. In dogs receiving conventional CPR, ICP increased by one-third of the rise of intrathoracic pressure in a range from 10 to 90 mm Hg. However, some modes of CPR change the intrathoracic to ICP relationship. In dogs, abdominal binding increases the transmission of pressure to the intracranial space to one-half of the rise of intrathoracic pressure. Open-chest CPR decreases the transmission of pressure and improves CPP compared with conventional CPR. Thus, increasing intrathoracic pressure may decrease cerebral blood flow because of the increase in downstream pressure, the ICP.

In this regard, ACD-CPR and ACD-ITV-CPR may have an advantage over conventional CPR. These techniques are designed to reduce intrathoracic pressure. Lindner et al. showed in a porcine model that cerebral perfusion is increased with ACD-ITV-CPR compared with standard CPR. Using an adult porcine model of hypothermic VF arrest, the same group demonstrated by micro-dialysis techniques improved lactate/pyruvate ratios and reduced glucose accumulation in the ACD-ITV group compared with standard CPR. In a pediatric porcine model of resuscitation, Voelckel et al. found that ACD-CPR with ITV provided superior cerebral blood flow compared with standard CPR.

Newer cardiopulmonary resuscitation techniques

A recent Cochrane systematic review of 11 trials including 12,944 adult patients with use of mechanical chest compression devices for CPR has not suggested that these mechanical devices are superior to conventional therapy. Similar recent reviews have had the same findings. However, pediatric trials were not included in this review and the authors did conclude that mechanical devices are a reasonable alternative where consistent, high-quality manual chest compressions are not possible or dangerous for the provider . Thus, simultaneous compression ventilation CPR, vest approaches, interposed abdominal compression CPR, and open-chest modalities are discussed. More recent approaches, including tourniquet-assisted CPR to augment myocardial perfusion, have been presented in limited animal case series and will be reserved for future editions.

Simultaneous compression ventilation cardiopulmonary resuscitation (SCV-CPR) is a technique designed to increase blood flow during conventional CPR by increasing the thoracic pump mechanism contribution to blood flow. Delivering a breath simultaneously with every compression, instead of after every fifth compression, increases intrathoracic pressure and augments blood flow produced by closed-chest CPR. Survival has been shown to be equivalent or significantly worse in both animals and humans who received SCV-CPR compared with conventional CPR. , , , No study has shown an increased survival rate with this CPR technique.

Interposed abdominal compression cardiopulmonary resuscitation (IAC-CPR) is the delivery of an abdominal compression during the relaxation phase of chest compression. An extensive review by Babbs has been published. IAC-CPR may augment conventional CPR in several ways. First, IAC-CPR may return venous blood to the chest during chest relaxation. , Second, IAC-CPR increases intrathoracic pressure and augments the duty cycle of chest compression. , Third, IAC-CPR may compress the aorta and return blood retrograde to the carotid or coronary arteries. IAC-CPR is an attractive alternative to some of the newer techniques of CPR because it requires no additional equipment for implementation. However, it does require training and manpower.

Four randomized controlled trials have compared IAC-CPR with standard CPR. The first trial, reported in 1985 by Mateer et al., was the largest and included 291 patients. IAC-CPR was applied in the field by paramedics until ambulance transport. No differences in mortality were found. The later trials involved a total of 279 hospitalized patients. The results from these trials are more positive; a meta-analysis of these studies found an increased likelihood of return of spontaneous circulation (ROSC) and intact survival to discharge with IAC-CPR versus standard CPR. , Although no intraabdominal trauma was detected in any of the 426 patients in these trials, one pediatric case report demonstrated direct pancreatic injury. More recent studies evaluating IAC-CPR to standard CPR on end-tidal carbon dioxide (ETCO 2 ) and ROSC demonstrated improvement in ETCO 2 without a difference in ROSC. This suggests that IAC-CPR may indicate improved cardiac output via this method without clinical improvement.

Application of IAC-CPR is limited by the need for training and additional manpower. Although it has not been studied in a pediatric group, with skilled personnel available, IAC-CPR could be considered for use with inpatient arrests.

Active compression-decompression cardiopulmonary resuscitation (ACD-CPR) involves a negative-pressure “pull” on the thorax during the release phase of chest compression using a hand-held suction device ( Fig. 38.6 ). This technique improves vascular pressures and minute ventilation during CPR in animals and humans. , The mechanism of benefit of this technique is attributed to enhancement of venous return by the negative intrathoracic pressure generated during the decompression phase. In addition, it reverses the chest wall deformation that accompanies standard CPR. Preliminary results in adults were promising, and a large multi-institutional study of ACD-CPR completed in Europe found that ACD-CPR was superior to standard CPR. , , In this study, a total of 750 patients were randomly assigned to receive standard CPR or ACD-CPR. In the experimental group, 5% survived to 1 year (12 patients with intact neurologic status) versus 2% (three patients with intact neurologic status) in the standard group. However, a number of other trials have not shown a difference between standard CPR and ACD-CPR, and a Cochrane Database Systematic Review concluded that there was no consistent benefit from use of this technique. The effectiveness of ACD-CPR appears to be relatively site specific. Explanations for this variability have focused on the effectiveness of training for providers and intersite variation of on-scene advanced life support techniques. Use of ACD-CPR requires significantly more physical effort than conventional CPR; this requirement may have influenced outcome. No device is cleared for clinical use in the United States at this time.

• Fig. 38.6

Device for performing active compression-decompression cardiopulmonary resuscitation. The upper part is a handle and the lower part is a suction cup.

(Courtesy AMBU Corporation.)

Use of an ITV has been evaluated in attempts to improve the outcome with ACD-CPR. , This technique involves the use of a valve placed between the ventilating bag and airway, which is designed to close when the tracheal pressure falls below atmospheric pressure, enhancing the development of negative intrathoracic pressure during ACD-CPR ( Figs. 38.7 and 38.8 ). Animal studies, including a young porcine model, showed improved organ perfusion, and brain micro-dialysis studies demonstrated decreased lactate accumulation and improved glucose utilization. , , , In a small series of patients, diastolic pressure was raised along with CPP and ETCO 2 release. These studies led to an inclusion of the technique as an acceptable alternative to standard CPR in the 2000 AHA guidelines and subsequent revised guidelines. , Plaisance et al. reported on a series of 400 patients randomly assigned to ACD-CPR with ITV or sham ITV. Survival at 24 hours was significantly improved. There was a nonsignificant trend toward improved neurologic survival, with 6 of 10 discharged patients having intact survival compared with 1 of 8 discharged survivors in the sham ITV group. In a randomized controlled study by Wolcke et al. of 610 adults in cardiac arrest in the out-of-hospital setting, use of ACD-CPR plus the ITV was associated with improved ROSC and 24-hour survival rates when compared with CPR alone. The addition of the ITV was associated with improved hemodynamics during standard CPR in one clinical study. The ultimate role of this technique, which requires specialized equipment and significant resuscitator training, remains to be determined. ,

• Fig. 38.7

Schematic diagram of impedance threshold valve. During positive-pressure ventilation, the valve is open and gas flows. During chest compression or exhalation, air moves freely through the valve. During chest decompression, airflow is impeded by the valve, decreasing intrathoracic pressure. During spontaneous ventilation, the check valve opens, allowing gas flow. CPR, Cardiopulmonary resuscitation.

(Modified from Lurie KG, Barnes TA, Zielinski TM, et al. Evaluation of a prototypic inspiratory impedance valve designed to enhance the efficiency of cardiopulmonary resuscitation. Resp Care. 2003;48[1]:52–57.)

• Fig. 38.8

Example of intratracheal pressures, a surrogate for intrathoracic pressures, in a patient undergoing cardiopulmonary resuscitation (CPR) with an automated compression device (ACD) with and without an impedance threshold device attached to a face mask. CPR was delivered at 100 compression/decompression cycles/min with a synchronized compression/ventilation ratio of 15:2. Note the absence of significant decreases in intratracheal pressures with a sham device. With the active impedance threshold device, wide fluctuations in intratracheal pressure are seen with each compression and decompression.

(Courtesy M. Lurie, MD, and Advanced Circulatory Systems, Inc.)

Vest CPR uses an inflatable bladder resembling a blood pressure cuff that is wrapped circumferentially around the chest with phased inflation to increase intrathoracic pressure. Because chest dimensions are changed minimally, direct cardiac compression is unlikely. In addition, the even distribution of the force of compression over the entire chest wall decreases the likelihood of trauma to the skeletal chest wall and its thoracic contents.

In a human study, vest CPR increased aortic systolic pressure but had little effect on aortic diastolic pressure compared with conventional CPR. Despite its late application, vest CPR improved the hemodynamics and rate of ROSC in adult patients in another study. Evidence from a case control study of 162 adults documented improvement in survival to the emergency department when vest CPR was administered by adequately trained personnel to patients in cardiac arrest in the out-of-hospital setting. The lack of metallic parts has allowed vest CPR to be used experimentally during nuclear magnetic resonance spectroscopy to study brain intracellular pH. Clinically, the use of vest CPR depends on sophisticated equipment and remains experimental at this time.

Abdominal binding

Abdominal binders and military antishock trousers have been used to augment closed-chest CPR. Both methods apply continuous compression circumferentially below the diaphragm. Three mechanisms have been proposed for augmentation of CPR by these binders. First, binding the abdomen decreases the compliance of the diaphragm and raises intrathoracic pressure. Second, blood may be moved out of the intrathoracic structures to increase circulating blood volume. Third, applying pressure to the subdiaphragmatic vasculature and increasing its resistance may increase suprathoracic blood flow. These effects increase aortic pressure and carotid blood flow in both animals and humans. , Unfortunately, as aortic pressure increases, the downstream component of CPP, namely, right atrial pressure, increases to an even greater extent, resulting in decreased CPP and myocardial blood flow. These techniques also lower the CPP by enhanced transmission of intrathoracic pressure to the intracranial vault, which raises ICP (the downstream component of CPP). Clinical studies have failed to show an increased survival when an abdominal binder or military antishock trouser suit was used to augment CPR.

Open-chest cardiopulmonary resuscitation

Use of open-chest cardiac massage has generally been replaced by closed-chest CPR. Compared with closed-chest CPR, open-chest CPR generates higher cardiac output and vital organ blood flow. During open-chest CPR, there is less elevation of intrathoracic, right atrial, and intracranial pressures, resulting in higher coronary and cerebral perfusion pressures and higher myocardial and cerebral blood flows.

Open-chest CPR is not a technique that can be applied by most health care personnel. It can be used in the operating room, ICU, or emergency department equipped with the necessary surgical and technical equipment and personnel. It is easily used in the operating room or ICU after cardiac surgery when the open chest can be easily accessed. Open-chest CPR is indicated for cardiac arrest resulting from cardiac tamponade, hypothermia, critical aortic stenosis, and ruptured aortic aneurysm. Other indications include cardiac arrest resulting from penetrating or crushed chest wall abnormalities that make closed-chest CPR impossible or ineffective. Open-chest CPR is indicated for select patients when closed-chest CPR has failed, although exactly which patients should receive this method of resuscitation under this condition is controversial. When initiated early after failure of closed-chest CPR, open-chest CPR may improve outcome. When performed after 15 minutes of closed-chest CPR, open-chest CPR significantly improves CPP and the rate of successful resuscitation.

Cardiopulmonary bypass and extracorporeal cardiopulmonary resuscitation

Because of the low rate of survival after prolonged CPR, more aggressive methods have been suggested to improve its success: cardiopulmonary bypass (CPB) and extracorporeal membrane oxygenation CPR (E-CPR).

CPB is one of the most effective ways to restore circulation after cardiac arrest. Animal studies show that CPB increases survival at 72 hours, increases recovery of consciousness, and preserves the myocardium better than does conventional CPR. In dogs, CPB resulted in better neurologic outcome than conventional CPR after a 4-minute ischemic period. However, neurologic outcome was dismal in both groups when the ischemic period lasted 12 minutes. Some 90% of dogs survived 24 hours after 15 to 20 minutes of cardiac arrest, but only 10% survived when the arrest time was prolonged to 30 minutes when CPB was used for stabilization during defibrillation. CPB decreased myocardial infarct size in a model involving coronary artery occlusion compared with conventional CPR. In all animal models, CPB improves the success of resuscitation compared with conventional CPR.

Human experience with CPB for cardiac arrest outside the operating room is growing. In the first major series of pediatric patients undergoing E-CPR, reported by Morris et al., 64 children underwent 66 extracorporeal membrane oxygenation (ECMO) runs initiated during active resuscitation with chest compressions or internal cardiac massage. Of these patients, 33 (50%) were decannulated and survived for more than 24 hours, 21 (33%) survived to hospital discharge, and 16 (26%) reportedly had no major changes in neurologic outcome. The average duration of CPR before cannulation in the survivors was 50 minutes. Of the 6 surviving children who required more than 60 minutes of CPR before ECMO, 3 had no apparent change in neurologic status. During the same period, 73 children underwent standard CPR; 10 received CPR for more than 30 minutes, with no survivors. Duncan et al. reported a series of 18 pediatric cardiac surgical patients at the Boston Children’s Hospital who received ECMO during active chest compressions. Of the first 7 patients, only 29% survived. This led to the development of a rapid ECMO deployment strategy in which an ECMO pump is kept saline-primed in the ICU at all times, allowing initiation of extracorporeal support within 15 minutes. Precannulation support times dropped from an average of 90 minutes but still remained high at an average of 50 minutes. Of the remaining 11 patients, 10 were decannulated successfully, with 6 long-term survivors, 5 of whom were in New York Heart Association class I. This rapid deployment strategy likely will become more commonplace in large pediatric centers.

Many subsequent pediatric and adult studies have shown both the feasibility and varied success of E-CPR. Clear indications for its use include witnessed arrest in a biventricular circulation, while contraindications include inability to provide effective CPR. There is uncertainty for the role of E-CPR in prolonged conventional resuscitation, as there are no guidelines on when to initiate or which patient population it would be best suited for. This lack of criteria may explain the wide range of success of E-CPR, with survival ranging between 33% and 100%.

Subgroup analysis from the Therapeutic Hypothermia after Cardiac Arrest In-Hospital trial demonstrated, in a univariate analysis of 56 children receiving open-chest CPR, that approximately one-half survived with good neurobehavioral outcomes at 1 year from index hospitalization. On multivariate analysis, the use of ECMO or other extracorporeal therapies demonstrated worse survival at 12 months and worsened neurobehavioral outcomes. However, this trial was designed to evaluate hypothermia as an intervention and many potential confounders may have biased these findings. Conversely, Lasa et al. compared 591 pediatric patients who received E-CPR versus 3165 that received CPR alone and showed that survival to hospital discharge and survival with a favorable neurologic outcome was more favorable with E-CPR compared with CPR alone. Bembea et al. linked data from the Extracorporeal Life Support Organization and AHA Get with the Guidelines–Resuscitation registries to determine risk factors related to unfavorable outcomes with E-CPR among 593 children. In this study, they found that odds of death were increased with a noncardiac diagnosis and preexisting renal insufficiency and that for each additional 5 minutes of CPR prior to ECMO initiation, the odds risk of death increased by 1.04.

Under the 2015 AHA guidelines, centers should consider E-CPR for in-hospital cardiac arrest refractory to standard resuscitation attempts if the condition leading to cardiac arrest is reversible or amenable to heart transplantation, if excellent conventional CPR has been performed after no more than several minutes of no-flow cardiac arrest, and if the institution is able to rapidly perform ECMO. Long-term survival has been reported even after more than 150 minutes of CPR in select patients.

Data are emerging involving the role of ECMO in persons with refractory VF. These data have solely been in the form of case reports but may represent a future direction for the care of patients with VF. In 2006, Samson et al. reported successful treatment of in-hospital VF in children with ECMO after cardiac arrest who had an initial rhythm of VF and immediate initiation of CPR.

CPB and ECMO require a great deal of technical support and sophistication. In units with preprimed circuits on standby, CPB can be implemented quickly and with moderate success in a population of children who would otherwise almost certainly die. The success with some patients undergoing very long CPR times followed by ECMO use is encouraging and suggests the possibility of reversible myocardial injury as a cause of resuscitation failure in a subset of patients.

Transcutaneous cardiac pacing

TCP is used as a method for noninvasive pacing of the ventricles for a relatively short period. Emergency cardiac pacing is successful in resuscitation only if it is initiated soon after the onset of arrest. In the absence of in situ pacing wires or an indwelling transvenous or esophageal pacing catheter, TCP is the preferred method for temporary electrical cardiac pacing. Since 1992, the AHA Advanced Cardiac Life Support (ACLS) guidelines have recommended the early use of an external pacemaker in patients with symptomatic bradycardia or asystole.

Since Zoll established TCP in 1952 as a clinically useful method of pacing adult patients during ventricular standstill (Stokes-Adams attacks) and bradycardia-associated hypotension, numerous anecdotal reports have supported its use for bradycardic or asystolic arrests. Zoll et al. reported successful in-hospital resuscitation of 12 of 16 patients with hypotensive bradycardia or asystole if TCP was initiated within 5 minutes of the arrest. In contrast, if TCP was started between 5 and 30 minutes after the arrest, only 8 of 44 patients with either of these rhythms could be resuscitated. In two controlled clinical trials of prehospital TCP, no differences in the survival rate or success of resuscitation were observed in paced and nonpaced patients who had asystole or PEA. , In patients with symptomatic bradycardia, TCP improved resuscitation and the survival rate.

To date, the efficacy of TCP in resuscitation of children has not been studied. Beland et al. showed that effective TCP could be achieved in hemodynamically stable children during induction of anesthesia for heart surgery. They were successful in 53 of 56 pacing trials, and the patients experienced no complications.

TCP is indicated for patients whose primary problem is impulse formation or conduction and who have preserved myocardial function. TCP is most effective in patients with sinus bradycardia or high-grade atrioventricular block with slow ventricular response who also have a stroke volume sufficient to generate a pulse. TCP is not indicated for patients in prolonged arrest because, in this situation, TCP usually results in electrical but not mechanical cardiac capture and its use may delay or interfere with other resuscitative efforts.

To set up pacing, one electrode is placed anteriorly at the left sternal border and the other posteriorly just below the left scapula. Smaller electrodes are available for infants and children; adult-sized electrodes can be used in children weighing more than 15 kg. Electrocardiographic leads should be connected to the pacemaker, the demand or asynchronous mode selected, and an age-appropriate heart rate used. The stimulus output should be set at zero when the pacemaker is turned on and then increased gradually until electrical capture is seen on the monitor. The output required for a hemodynamically unstable rhythm is higher than that for a stable rhythm in children in whom the mean stimulus required for capture was between 52 and 65 mA. After electrical capture is achieved, one must ascertain whether an effective arterial pulse is generated. If pulses are not adequate, other resuscitative efforts should be used.

The most serious complication of TCP is induction of a ventricular arrhythmia. Fortunately, this complication is rare and may be prevented by pacing only in the demand mode. Mild transient erythema beneath the electrodes is common. Skeletal muscle contraction can be minimized by using large electrodes, a 40-ms pulse duration, and the smallest stimulus required for capture. Sedatives or analgesics may be necessary in the patient who is awake. If defibrillation or cardioversion is necessary, one must allow a distance of 2 to 3 cm between the electrode and paddles to prevent arcing of the current.


Adrenergic agonists

In 1963, only 3 years after the original description of closed-chest CPR, Pearson and Redding described the use of adrenergic agonists for resuscitation. They subsequently showed that early administration of epinephrine in a canine model of cardiac arrest improved the success rate of CPR. They also demonstrated that the increase in aortic diastolic pressure by administration of α-adrenergic agonists was responsible for the improved success of resuscitation. They theorized that vasopressors such as epinephrine were of value because the drug increased peripheral vascular tone, not because of a direct effect on the heart.

Yakaitis et al. investigated the relative importance of α- and β-adrenergic agonist actions during resuscitation. Only 27% of dogs that received a pure β-adrenergic receptor agonist along with an α-adrenergic antagonist were resuscitated successfully compared with all of the dogs that received a pure α-adrenergic agonist and a β-adrenergic antagonist. Later studies reconfirmed this finding. Michael et al. demonstrated that the α-adrenergic effects of epinephrine result in intense vasoconstriction of the resistance vessels of all organs of the body, except those supplying the heart and brain. Because of the widespread vasoconstriction in nonvital organs, adequate perfusion pressure—and, thus, blood flow to the heart and brain—can be achieved despite the fact that cardiac output is very low during CPR.

The increase in aortic diastolic pressure associated with epinephrine administration during CPR is critical for maintaining coronary blood flow and enhancing the success of resuscitation. Even though the contractile state of the myocardium is increased by use of β-adrenergic agonists in the spontaneously beating heart, β-adrenergic agonists actually may decrease myocardial blood flow during CPR by increasing intramyocardial wall pressure and vascular resistance. This decrease in myocardial blood flow could redistribute intramyocardial blood flow away from the subendocardium, increasing the likelihood of ischemic injury to this region. Moreover, evidence indicates that left ventricular end-diastolic pressure (LVEDP) rises with epinephrine use, reducing the overall impact of the vasoconstrictor effects of epinephrine on CPP. Tang et al. showed elevated LVEDP and decreased measures of diastolic performance in epinephrine-resuscitated rats after induced VF compared with phenylephrine-resuscitated animals or epinephrine-resuscitated animals who also received a β-blocker. Similar data were found by McNamara, who used a rat pup model of asphyxial arrest. LVEDP was increased and diastolic function indices were decreased with epinephrine compared with either saline solution alone or epinephrine combined with verapamil. These data imply that excessive β-adrenergic effects prevent the intracellular calcium reuptake during diastole that is required for myocardial relaxation. By its inotropic and chronotropic effects, β-adrenergic stimulation increases myocardial oxygen demand, which increases the risk of ischemic injury when superimposed on low coronary blood flow. This combination of increased oxygen demand by β-adrenergic agonists and decreased oxygen supply may damage an already ischemic heart, raising the question of whether a pure α-adrenergic agonist would be better than epinephrine, with its significant β-adrenergic effects ( Box 38.1 ). The effects on energy utilization and oxygen supply not only have implications for the success of the initial resuscitation but also for the postresuscitation function of the myocardium.

• BOX 38.1

α-Adrenergic vs β-Adrenergic Agonist Effects

α-Adrenergic effects

  • Vasoconstrict peripheral vessels

  • Maintain aortic diastolic pressure

  • Improve coronary blood flow

  • No metabolic stimulatory effect

β-Adrenergic effects

  • Vasodilate peripheral vessels

  • Decrease aortic diastolic pressure

  • Increase cellular metabolic rate

  • Positive inotrope

  • Increase intensity of ventricular fibrillation

  • Increase heart rate and/or dysrhythmias following resuscitation

In an effort to address the balance of the risks and benefits of epinephrine, a large randomized controlled trial was conducted among five National Health Service ambulance companies in the United Kingdom. The PARAMEDIC2 trial enrolled 8014 adult patients with out-of-hospital cardiac arrest and randomized them to either standard-dose epinephrine or saline placebo. Overall survival was improved in the epinephrine group (3.2% vs. 2.4%), but there was no difference in rates of survival with favorable neurologic outcomes. This finding is likely due to the much higher rates of severe neurologic impairment in the epinephrine survivors and is similar to results found in observational studies. A meta-analysis that included this study showed improved survival to admission with epinephrine but no improvement with survival to discharge or good neurologic outcome. Timing of epinephrine administration may have an impact. Recent analyses of large resuscitation databases have shown that survival is greatest with early administration of epinephrine in both adult and pediatric patients. Another interesting finding from analyzing these large databases is that longer time periods between epinephrine doses than what is currently recommended may be associated with better survival to hospital discharge. However, as this was a retrospective analysis, it warrants further investigation.

A number of studies have attempted to compare α-adrenergic agonists to epinephrine during CPR. Phenylephrine and methoxamine are two pure α-adrenergic agonists that have been used in animal models of CPR with success equal to that of epinephrine. , , More recently, vasopressin (discussed in depth later) has been studied as a noncatecholamine vasoconstrictor in the management of patients who experience cardiac arrest. These agents cause peripheral vasoconstriction and increase aortic diastolic pressure, resulting in improved myocardial and cerebral blood flows. This effect results in a higher oxygen supply/demand ratio in the ischemic heart and a theoretical advantage over the combined α- and β-adrenergic agonist effects of epinephrine. These agonists, as well as vasopressors such as vasopressin, have been used successfully for resuscitation. , , These drugs maintain blood flow to the heart during CPR with similar performance to epinephrine. In an animal model of VF cardiac arrest, a resuscitation rate of 75% was reported for both epinephrine- and phenylephrine-treated groups. In this study, the ratio of endocardial to epicardial blood flow was lower in the group treated with epinephrine, suggesting the presence of subendocardial ischemia. However, studies of this kind are difficult to interpret because of the inability to measure the degree of α-receptor activation by the different vasopressors. The higher subendocardial blood flow in the phenylephrine group may have been the result of less α-receptor activation. Moreover, some investigators have questioned the merits of using a pure α-adrenergic agonist during CPR. Although the inotropic and chronotropic effects of β-adrenergic agonists may have deleterious hemodynamic effects during CPR administered for VF, increases in both heart rate and contractility increase cardiac output when spontaneous coordinated ventricular contractions are achieved.

Cerebral blood flow during CPR, like coronary blood flow, depends on peripheral vasoconstriction and is enhanced by use of α-adrenergic agonists. This action produces selective vasoconstriction of noncerebral peripheral vessels to areas of the head and scalp without causing cerebral vasoconstriction. As with myocardial blood flow, pure α-agonist agents are as effective as epinephrine in generating and sustaining cerebral blood flow during CPR in adult animal models and in infant models. , No difference in neurologic deficits was found at 24 hours after cardiac arrest between animals receiving either epinephrine or phenylephrine during CPR. Animal studies have shown transient improvements in cerebral oxygenation, and similarly in a prospective study of 36 adult patients, a bolus of epinephrine showed a small increase in cerebral oxygenation within the first 5 minutes after administration.

Analogous to the heart, β-adrenergic agonists could increase cerebral oxygen uptake if a sufficient amount of drug crosses the blood-brain barrier during or after resuscitation. In addition, adrenergic agonists may vasoconstrict or dilate cerebral vessels depending on the balance between α- and β-adrenergic receptors. Epinephrine and phenylephrine had similar effects on cerebral blood flow and metabolism, maintaining normal cerebral oxygen uptake for 20 minutes of CPR in dogs. This finding implies that cerebral blood flow was high enough to maintain adequate cerebral metabolism and that β-receptor stimulation did not increase cerebral oxygen uptake, despite the fact that the combined effects of brain ischemia and CPR can increase the permeability of the blood-brain barrier to drugs used during CPR or when enzymatic barriers to vasopressors (e.g., by monoamine oxidase) are overwhelmed during tissue hypoxia. Mechanical disruption of the barrier could occur during chest compressions by large fluctuations in cerebral venous and arterial pressures or as a result of hyperemia, the large increase in cerebral blood flow that occurs during the early reperfusion period when the cerebral vascular bed is maximally dilated following resuscitation, particularly if systemic hypertension occurs. No blood-brain barrier permeability changes during CPR immediately after resuscitation or 4 hours after resuscitation were found in adult dogs. However, after 8 minutes of cardiac arrest and 6 minutes of CPR in piglets, the blood-brain barrier was permeable to the small neutral amino acid α-aminoisobutyric acid 4 hours after cardiac arrest ( Fig. 38.9 ). , The increase in permeability could be prevented by prearrest administration of conjugated superoxide dismutase and catalase, indicating a role of oxygen free radicals in the pathogenesis of this injury to the blood-brain barrier ( Fig. 38.10 ). These endothelial membrane changes frequently were associated with the presence of intravascular polymorphonuclear and monocytic leukocytes. Whether leukocytes disrupt the blood-brain barrier by release of toxic substances, such as oxygen free radicals or proteases, or appear in the postischemic microvessels as an epiphenomenon of a more important derangement is unknown ( Fig. 38.11 ).

• Fig. 38.9

Transfer coefficient (Ki) of α-aminoisobutyric acid for pons, diencephalon (DIE), and middle cerebral (MCA) artery regions. Control group: 8 minutes ischemia and 10 minutes cardiopulmonary resuscitation (CPR); 8 minutes ischemia and 40 minutes cardiopulmonary resuscitation; 3 minutes after resuscitation; 4 hours after resuscitation. Group 5 in each region, * P < .05, different from group 1 by one-way analysis of variance and Dunnett test for all three regions.

(Modified from Schleien CL, Koehler RC, Schaffner DH, et al. Blood-brain barrier disruption after cardiopulmonary resuscitation in immature swine. Stroke. 1991;22:477.)

• Fig. 38.10

(A) Transfer coefficient of α-aminoisobutyric acid (AIB) from plasma to brain in hippocampus, caudate nucleus, and primary supply and border regions of cerebral arteries in nonischemic time controls (group 1, n = 5), ischemia group treated with polyethylene glycol (PEG; group 2, n = 8), and ischemia group treated with PEG-superoxide dismutase and PEG-catalase (group 3, n = 8). Error bars represent standard error of the mean (SEM). * P < .05 between groups 2 and 3 by Mann-Whitney U test. (B) Transfer coefficient of AIB from plasma to brain in caudal brain regions in nonischemic time controls (group 1, n = 5), ischemia group treated with PEG (group 2, n = 8), and ischemia group treated with PEG-superoxide dismutase and PEG-catalase (group 3, i= 8). Error bars represent SEM. * P < .05 between groups 2 and 3 by Mann-Whitney U test.

(Modified from Schleien CL, Eberle B, Schaffner DH, et al. Reduced blood-brain barrier permeability after cardiac arrest by conjugated superoxide dismutase and catalase in piglets. Stroke . 1994;25[9]:1830–1834.)

• Fig. 38.11

Transmission electron micrograph of an infant piglet brain 4 hours after 8 minutes of cardiac arrest and 6 minutes of cardiopulmonary resuscitation (magnification ×5000). An intravascular leukocyte, which has the morphologic features of a monocyte, is adherent to the endothelial surface of a venule and appears to be occluding the lumen. The luminal surface of the endothelial cell contains membrane blebs and discontinuities.

(From Caceres MJ, Schleien CL, Kuluz JW, et al. Early endothelial damage and leukocyte accumulation in piglet brains following cardiac arrest. Acta Neuropathol. 1995;90[6]:582.)


The role of vasopressin as a noncatecholamine vasoconstrictor in the management of patients who experience cardiac arrest has received a great deal of interest. Work by Lindner in Europe and Landry in the United States had established sufficient evidence of efficacy for its use to be included in the 2010 AHA Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. , Subsequent adult studies comparing standard-dose epinephrine with vasopressin alone or in combination with standard-dose epinephrine have showed that vasopressin offered no advantage in ROSC or survival to discharge. It has subsequently been removed from the 2015 AHA guidelines. However, increasing evidence indicates that vasopressin is a useful agent in the management of shock of multiple etiologies. Therefore, it may have a role in postresuscitation management.

Arginine vasopressin is a short peptide hormone secreted by the posterior pituitary gland in response to changes in tonicity and in effective intravascular volume, signaled primarily via baroreceptor unloading in the aorta. Severe shock is the most potent stimulus to vasopressin secretion. Serum levels 20- to 200-fold higher than normal may be found immediately after cardiac arrest and in other severe shock states. Despite these observations, lower than expected vasopressin levels have been found in some patients with profound shock, and patients dying of cardiac arrest have been found to have significantly lower vasopressin levels than do survivors. The cause of lower than expected vasopressin levels in some patients is unclear. Depletion of vasopressin stores may be a potential mechanism. Dogs subjected to profound hemorrhagic shock have an early massive elevation of vasopressin levels immediately after the event, followed by a depression below expected levels within 1 hour of the insult. Severe depletion of vasopressin stores from the posterior hypophysis was noted. These animals developed a catecholamine-refractory vasodilatory shock that responded dramatically to low doses of vasopressin. These observations have led to an exploration of the use of vasopressin in both cardiac arrest and shock states.

Vasopressin is an extremely potent vasoconstrictor. Its effects on vascular tone are primarily mediated through interaction with a specific G protein–coupled receptor referred to as the V 1a -receptor , which is distributed widely throughout vascular beds. Of note, the V 1 a-receptor is linked to the same second messenger system as the α-adrenergic receptor that mediates vasoconstriction through alteration of intracellular calcium levels. However, in the pulmonary circulation, vasopressin activation of V 1 -receptors mediates the release of nitric oxide and causes pulmonary vasodilatation. Vasopressin also interacts with its V 2 -receptor, which regulates aquaporin expression on the renal collecting duct epithelium. Stimulation of the V 2 -receptor occurs at substantially lower levels than those required to activate the V 1 a-receptor.

Vasopressin use during resuscitation has been studied in animals and humans. In an adult porcine model of VF, vasopressin at a dose of 0.8 μg/kg was found to be superior to the maximally effective dose of epinephrine 200 μg/kg in restoring LV myocardial blood flow, increasing diastolic CPP and total cerebral blood flow as well as rates of ROSC. Moreover, the duration of the effect was sustained for 4 minutes, compared with 1.5 minutes for epinephrine. Adverse effects noted in the postresuscitation phase included decreased renal and adrenal blood flows and reduced cardiac output. ,

In a pediatric porcine model of cardiac arrest, vasopressin at a dose of 0.8 μg/kg was not as effective as epinephrine at 200 μg/kg in restoring LV myocardial blood flow or achieving ROSC. Only 1 of 6 animals achieved ROSC compared with 6 of 6 in the epinephrine group. A combination group that received both epinephrine at 45 μg/kg and vasopressin at 0.8 μg/kg fared better with ROSC in 4 of 6 animals. Possible explanations for the difference between adult and juvenile animals include different dose-response curves for the two drugs, failure of maturation of vasopressin receptors, a different distribution of vasopressin receptors, or the different experimental model.

In an initial small, randomized clinical trial of vasopressin compared with epinephrine for refractory VF, the rate of achieving ROSC was higher in the vasopressin group. A large multicenter randomized trial of vasopressin for cardiac arrest in adults has been reported. More than 1200 patients were randomly assigned in the field to receive 2 doses of either 40 international units (IU) of vasopressin or 1 mg of epinephrine followed by additional treatment with epinephrine, if necessary. In patients in whom ROSC was not achieved after two doses of medication, a third dose of medication such as epinephrine could be added at the resuscitating physician’s discretion. Initial dose vasopressin was equivalent to epinephrine in achieving survival to both hospital admission and discharge in patients with either PEA or VF. In patients with asystole, vasopressin was superior to epinephrine in achieving both survival to admission and discharge, although intact neurologic outcome was not improved. In the group receiving a third dose of medication such as epinephrine, survival was greater in the vasopressin group. In a study of 200 patients with in-hospital cardiac arrest, patients were randomly assigned to receive either 1 mg of epinephrine or 40 IU of vasopressin. Again, no statistical difference in survival to 1 hour or to hospital discharge was found between groups or subgroups. The results of these studies led to the classification of evidence for vasopressin for use in adults as indeterminate in the 2010 AHA guidelines. Subsequently, a meta-analysis of 10 randomized trials, including a total of 6120 adult patients, showed that the use of vasopressin was neither beneficial nor harmful in an unselected patient population in terms of ROSC, survival to hospital admission or discharge, or favorable neurologic outcome. However, chance of ROSC was significantly higher in in-hospital cardiac arrest patients when vasopressin was used.

The published experience with vasopressin in children who experience cardiac arrest is limited. The first case series of vasopressin use in CPR reported the outcome of four children with six prolonged refractory cardiac arrests that were unresponsive to standard resuscitation efforts. Each child received one or more bolus doses of vasopressin (0.4 U/kg) as rescue therapy. In all children, the initial rhythm was a form of PEA that deteriorated to asystole in four of six events. Three children had ROSC for more than 60 minutes, including one child with asystole. Two children survived for more than 24 hours and one survived to hospital discharge.

A review of a national registry of in-hospital CPR showed that patients who received vasopressin had a lower incidence of ROSC greater than 20 minutes (22 [34%] of 64) than patients who did not receive the medication (675 [55%] of 1229). The association of poor outcome with vasopressin persisted even with multivariate analysis with logistic regression to attempt to control for other factors that might affect ROSC. The effect of vasopressin in pediatric arrest that was refractory to an initial epinephrine dose was evaluated in a pilot study. Patients were given vasopressin after an initial dose of epinephrine and were compared with a retrospective matched cohort of patients who experienced cardiopulmonary arrest that required greater than two doses of a vasopressor, not including vasopressin. Ten patients were enrolled; while there was an increased 24-hour survival, there was no difference in ROSC, survival to hospital discharge, or favorable neurologic status at discharge.

The current evidence examines the use of vasopressin only as a potential alternative when standard therapies, such as epinephrine, fail to cause ROSC. Unfortunately, variables such as dosing, timing of vasopressin infusion, or pediatric risk of mortality scores have not been controlled for in these studies. No double-blinded, randomized controlled studies have been performed; thus, no firm recommendations are available concerning the use of vasopressin for CPR in infants and children.

The current recommended dose of vasopressin for adults in cardiac arrest is 40 IU. No data comparing this dose to other doses are available, and concern exists regarding postresuscitation complications related to this dose. We have selected 0.5 IU/kg as the standard for cardiac arrest. Further data are needed before more definitive dosing recommendations can be made.

Use of vasopressin in postresuscitation management may be considered. A relative vasopressin deficiency has been noted in a number of shock states, including hemorrhage, sepsis, and post-CPB, as well as in patients who have unsuccessful resuscitations. In these settings, shock may be refractory to catecholamines. These patients may respond to a vasopressin infusion, allowing the weaning of high-dose catecholamines. Although a role in the postresuscitation setting has not been demonstrated based on the data related to refractory shock, consideration of the use of vasopressin for refractory hypotension may be appropriate.

Additionally, the additive effects of combination drug therapy for adults with cardiac arrest may be most beneficial. In a recent randomized, double-blind placebo control study investigating vasopressin plus epinephrine or saline placebo plus epinephrine with or without methylprednisolone, the group receiving combination therapy with vasopressin-epinephrine-methylprednisolone with CPR resulted in improved survival to hospital discharge with improved neurologic status. ,

High-dose epinephrine

The physiologic responses of animals and humans to higher doses of epinephrine include higher cerebral blood flow, increased myocardial and submyocardial blood flow, improved oxygen delivery relative to oxygen consumption, and less depletion of myocardial adenosine triphosphate (ATP) stores with more rapid repletion of phosphocreatine. , Contrary results, with increased myocardial oxygen consumption and decreased myocardial blood flow, have been demonstrated during CPR following VF cardiac arrest. , In a piglet model, high-dose epinephrine (HDE) produced lower myocardial blood flow than standard-dose epinephrine (SDE). In neonatal lambs following asphyxia-induced bradycardia, HDE resulted in a higher heart rate but lower stroke volume and cardiac output. Additionally, prolonged peripheral vasoconstriction and excessive doses of epinephrine may delay or impair reperfusion of systemic organs, particularly the kidneys and gastrointestinal tract.

Studies regarding survival of patients who were given HDE have been contradictory. In out-of-hospital patients who experienced cardiac arrest, HDE produced higher aortic diastolic pressure during CPR and increased the rate of ROSC compared with standard doses of epinephrine. Gonzalez et al. demonstrated a dose-dependent increase in aortic blood pressure by epinephrine in patients who failed to respond to prolonged resuscitative efforts. Paradis et al. showed that HDE increased aortic diastolic pressure and improved the rate of successful resuscitation in patients in whom ACLS protocols had failed. This group also reported on a series of 20 children treated with HDE and compared them with 20 historic control subjects consisting of children with cardiac arrest treated with SDE. They reported that 14 of the children in the HDE group had ROSC, 8 survived to hospital discharge, and 3 were neurologically intact. There were no survivors in the SDE comparison group. Other centers have claimed that higher-than-standard doses of epinephrine during CPR in children improve the hemodynamics and increase the success of CPR. However, no one has provided any valid data suggesting that HDE improves survival beyond the immediate postresuscitation period. , , , Based on these studies, the 1992 AHA guidelines for pediatric advance life support recommended HDE if an initial SDE failed to resuscitate the child.

Three large multicenter studies were subsequently published that dampened enthusiasm for the use of HDE. Stiell et al. studied 650 cardiac arrest adult patients who were randomly assigned to receive either an SDE or HDE (7 mg) protocol. No differences were observed between the groups with regard to 1-hour survival (23% vs. 18%), rate of hospital discharge (5% vs. 3%), or neurologic outcome. Brown et al. reported on 1280 cardiac arrest adult patients who received either SDE (0.02 mg/kg) or HDE (0.2 mg/kg). Again, no differences in ROSC, short-term survival, survival to hospital discharge, or neurologic outcome were observed between the two groups of patients. In a study of 816 adults, Callaham et al. reported a higher ROSC in the HDE group. However, there were no differences in the rate of hospital discharge or ultimate survival of these patients. In addition to these studies, a specific pediatric animal study was published that failed to demonstrate a clear survival benefit for HDE, although the occurrence of ROSC appeared to be greater. The 2000 AHA guidelines changed the recommendation for HDE to an option for second and subsequent doses of epinephrine.

A prospective, randomized, double-blind clinical trial of HDE in 68 pediatric inpatients was reported by Perondi et al. ROSC for more than 20 minutes was achieved in 15 of 34 patients who received HDE but in only 8 of 34 patients who received SDE ( P = .07). However, survival to 24 hours occurred in only two of the HDE group versus seven of the SDE group ( P = .05). In the group that experienced an asphyxial arrest, none of 12 treated with HDE were alive at 24 hours, whereas 7 of 18 patients in the SDE group survived. Four survived to hospital discharge, and two patients were neurologically normal. A meta-analysis of epinephrine use in adult cardiac arrest showed no difference in survival to discharge or neurologic outcome when using high dose over standard dose. These studies reinforced concerns that HDE may account for some of the adverse effects that occur after resuscitation and is the basis of the 2015 AHA guidelines’ recommendation against the use of HDE during CPR. , ,


Atropine is a parasympatholytic agent that acts by blocking cholinergic stimulation of the muscarinic receptors of the heart, which usually results in an increase in the sinus rate and shortening of the atrioventricular node conduction time. Atropine may also activate latent ectopic pacemakers. It has little effect on systemic vascular resistance, myocardial perfusion pressure, or contractility.

Atropine is indicated for treatment of asystole, PEA, bradycardia associated with hypotension, second- and third-degree heart block, and slow idioventricular rhythms. In children who present in cardiac arrest, sinus bradycardia and asystole are the most common initial rhythms, which make atropine useful as a first-line drug. Atropine is particularly effective in clinical conditions associated with excessive parasympathetic tone.

The recommended dose of atropine is 0.02 mg/kg, with a minimum dose of 0.1 mg and a maximum dose of 0.5 mg/dose. Smaller doses than 0.1 mg, even in small infants, may result paradoxically in bradycardia because of a central stimulatory effect on the medullary vagal nuclei by a dose that is too low to provide anticholinergic effects on the heart, although this phenomenon has come under debate. , Atropine may be given by any route, including intravenous, endotracheal, interosseous, intramuscular, and subcutaneous. Its onset of action occurs within 30 seconds, and its peak effect occurs between 1 and 2 minutes after an intravenous dose. The recommended adult dose is 0.5 mg every 5 minutes until the desired heart rate is obtained up to a maximum of 3 mg. For asystole, 1 mg is given intravenously and repeated every 5 minutes if asystole persists. Full vagal blockade usually is obtained with a dose of 2 mg in adults.

Because of its parasympatholytic effects, atropine should not be used in patients in whom tachycardia is undesirable. In patients after myocardial infarction or ischemia with persistent bradycardia, atropine should be used in the lowest dose possible to increase heart rate. Using the lowest possible dose will limit tachycardia, a potent contributor to increased myocardial oxygen consumption, which could lead to VF. In addition, atropine should not be used in patients with pulmonary or systemic outflow tract obstruction or idiopathic hypertrophic subaortic stenosis because tachycardia decreases ventricular filling and lowers cardiac output in this setting.

Sodium bicarbonate

The administration of sodium bicarbonate results in an acid-base reaction in which bicarbonate combines with hydrogen to form carbonic acid, which dissociates into water and carbon dioxide. Because of the generation of carbon dioxide, adequate alveolar ventilation must be present to achieve the normal buffering action of bicarbonate. Use of sodium bicarbonate during CPR remains controversial because of its potential adverse effects and the lack of evidence showing any benefit from its use during CPR. ,

Sodium bicarbonate is indicated for correction of significant metabolic acidosis, especially when signs of cardiovascular compromise are present. Acidosis itself may have a number of negative effects on the circulation, including depression of myocardial function by prolonging diastolic depolarization, depressing spontaneous cardiac activity, decreasing the electrical threshold for VF, decreasing the inotropic state of the myocardium, and reducing the cardiac response to catecholamines. Acidosis also decreases systemic vascular resistance and attenuates the vasoconstrictive response of peripheral vessels to catecholamines. This effect is contrary to the desired effect during CPR. In addition, particularly in patients with a reactive pulmonary vascular bed, pulmonary vascular resistance is inversely related to pH. Rudolph and Yuan observed a twofold increase in pulmonary vascular resistance in calves when pH was lowered from 7.4 to 7.2 under normoxic conditions. Therefore, correction of even mild acidosis may be helpful in resuscitating patients who have the potential for increased right-to-left shunting through a cardiac septal defect, patent ductus arteriosus, or aortic-to-pulmonary shunt during periods of elevated pulmonary vascular resistance.

Multiple adverse effects of bicarbonate administration include metabolic alkalosis, hypercapnia, hypernatremia, and hyperosmolality. All of these adverse effects are associated with a high mortality rate. Alkalosis causes a leftward shift of the oxyhemoglobin dissociation curve, thus, impairing release of oxygen from hemoglobin to tissues at a time when oxygen delivery already may be low. Alkalosis can result in hypokalemia, by enhancing potassium influx into cells, and ionic hypocalcemia, by increasing protein binding of ionized calcium. Hypernatremia and hyperosmolality may decrease tissue perfusion by increasing interstitial edema in microvascular beds. The marked hypercapnic acidosis that occurs during CPR on the venous side of the circulation, including the coronary sinus, may be worsened by administration of bicarbonate. , Myocardial acidosis during cardiac arrest is associated with decreased myocardial contractility. The mean venoarterial partial pressure of carbon dioxide difference was 24 ± 15 mm Hg in 5 patients during CPR and actually increased from 16 to 69 mm Hg in 1 patient after administration of bicarbonate. Another group showed a mean difference of 42 mm Hg between partial pressure of carbon dioxide in mixed venous blood and partial pressure of arterial carbon dioxide (Pa co 2 ) during CPR. Paradoxical intracellular acidosis after bicarbonate administration is possible because of rapid entry of carbon dioxide into cells with a slow egress of hydrogen ion out of cells. Paradoxical intracellular acidosis in the central nervous system after bicarbonate administration has been proposed but not definitively shown. In neonatal rabbits recovering from hypoxic acidosis, bicarbonate administration increased both arterial pH and intracellular brain pH as measured by nuclear magnetic resonance spectroscopy. In another study, intracellular brain ATP concentration in rats did not change during severe intracellular acidosis in the brain produced by extreme hypercapnia. The rats who maintained ATP concentration even in the face of severe brain acidosis had no functional or histologic differences from normal control subjects. Using nuclear magnetic resonance spectroscopy of the brain in dogs during cardiac arrest and CPR, intracellular brain pH decreased to 6.29, with total depletion of brain ATP after 6 minutes of cardiac arrest. Following effective CPR, ATP levels rose to 86% of prearrest levels and to normal by 35 minutes of CPR despite ongoing peripheral arterial acidosis ( Fig. 38.12 ). However, cerebral pH decreased in parallel with blood pH when CPR was started immediately after arrest. Bicarbonate administration ameliorated and did not worsen the cerebral acidosis, indicating that the blood-brain pH gradient is maintained during CPR.

Jun 26, 2021 | Posted by in CRITICAL CARE | Comments Off on Physiologic foundations of cardiopulmonary resuscitation

Full access? Get Clinical Tree

Get Clinical Tree app for offline access