Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

PART FOUR – Associated Problems in Pediatric Anesthesia

Chapter 33 – Pediatric Cardiopulmonary Resuscitation

  1. Blaine Easley,Charles L. Schleien,
    Donald H. Shaffner



Incidence of Cardiac Arrest During Anesthesia, 1110



Etiology of Cardiac Arrest During Anesthesia, 1112 



The Relationship of the Etiology of a Cardiac Arrest to the Anesthesia Care,1114



Outcome From Cardiac Arrest During Anesthesia, 1114



Prevention of Cardiac Arrest During Anesthesia, 1116



Physiology of Cardiopulmonary Resuscitation, 1117 



Recognition of the Need for Cardiopulmonary Resuscitation,1117



Reestablishment of Ventilation, 1117



Reestablishment of Circulation, 1119



Maintenance of Circulation, 1122



Methods of Cardiopulmonary Resuscitation, 1123 



Conventional Cardiopulmonary Resuscitation, 1123



Conventional Cardiopulmonary Resuscitation in the Prone Position, 1123



Simultaneous Compression–Ventilation Cardiopulmonary Resuscitation, 1124



Abdominal Binding, 1124



Abdominal Compression, 1124



Vest Cardiopulmonary Resuscitation, 1125



High-Impulse Cardiopulmonary Resuscitation, 1125



Active Compression–Decompression Cardiopulmonary Resuscitation, 1125



Phased Chest Abdominal Compression–Decompression Cardiopulmonary Resuscitation,1125



Open-Chest Cardiopulmonary Resuscitation, 1126



Cardiopulmonary Bypass, 1126



Vascular Access for Drug and Fluid Administration, 1126 



Peripheral and Central Vascular Access,1126



Intraosseous Access, 1127



Intratracheal Access, 1128



Fluid Resuscitation, 1128



Electrical Interventions in Arrest, 1128 



Pacing, 1128



Cardioversion, 1131



Defibrillation, 1133



Automated External Defibrillation,1137



Pharmacology of Resuscitation, 1137 



Antiarrhythmic Agents, 1137



Vasoactive Drugs, 1142



Other Resuscitation Medications,1145



Postresuscitation Care, 1147



Summary, 1147

In the late 1950s, children with cardiac arrest during anesthesia received 1.5 minutes of knee-to-chest “artificial respiration” ( Rainer, 1957 ), followed by a thoracotomy for internal cardiac massage. In 1958, closed-chest compressions were successfully performed on a 2-year-old child ( Sladen, 1984 ). The resuscitation of this child, along with several successful resuscitations of subsequent patients (many undergoing anesthesia), led to the reporting of closed-chest compressions ( Kouwenhoven et al., 1960 ). The high rate of successful resuscitation after cardiac arrest in the operating room (42%) ( Jude et al., 1961 ) helped establish closed-chest compression as the standard for cardiopulmonary resuscitation (CPR).

Despite the success rate of resuscitation during anesthesia, the potential for disaster and the increased likelihood of arrests in young children require that pediatric anesthesiologists have a complete understanding of the physiology and pharmacology of CPR. “No more depressing shadow can darken an operating room than that occasioned by the death of a child” ( Leigh and Belton, 1949 ).


Anesthesia-related cardiac arrest and anesthesia-related mortality are the commonly used terms that suggest a general association between the delivery of anesthesia and the most devastating events that can result from the care of a child by an anesthesiologist in the operating room or in the perioperative period. To better understand and, it is hoped, prevent these events, it is necessary to know both the likelihood of their occurrence and their associated risk factors. The likelihood (or frequency) of a cardiac arrest is usually described as either a percentage of occurrence within a subgroup (children) compared with the main group (all patients) or an incidence of occurrence expressed as a ratio of a total number of anesthetic procedures for a group. The presentation of the likelihood of cardiac arrest as a percentage can be misleading because the groups being compared often do not have the same denominator of anesthetics performed. For example, the impact of a 20% anesthesia-related cardiac arrest rate for children depends upon the percentage of the anesthetics at an institution that were performed in children. Incidence data are more informative but less available than percentage-based data because most data sets collect events and are unlikely to have the denominators that include the number of uneventful anesthetics for each group. A compilation of data regarding the incidence of cardiac arrest and mortality from the anesthesia literature is available in Table 33-1 .

TABLE 33-1   -- Incidence of cardiac arrests and mortality by age (per 10,000 anesthetic procedures)[*]


Anesthesia Related

All Causes









No.of Studies



No.of Studies



No.of Studies



No.of Studies
















9.2 to 19



0 to 9.3



16 to 24



12 to 165




0 to 4.3



0 to 3.2



4.4 to 6.0



1.8 to 59


All pediatric groups


0.4 to 5.5



0 to 4.7



2.7 to 14



3.8 to 88


All ages


0.1 to 7.8



0.1 to 7.9



1.5 to 27



0.9 to 189



The incidence of cardiac arrest and mortality for all ages includes studies that combine pediatric and adult patients.


Multiple risk factors for “anesthesia-related” cardiac arrest have been suggested and include the patient's age, their condition or “physical status,” and the emergency nature of the procedure. Age is a risk factor for an increased incidence of both cardiac arrest and mortality (see Table 33-1 ). The data in the table for neonates are incomplete because incidence data were not available for “anesthesia-related” events, but one study was available for “all causes” of intraoperative events in neonates. Neonates and infants appear to be at the greatest risk for both cardiac arrest and mortality. Children have the lowest incidence compared with neonates and infants. The table indicates that the group “All pediatric groups” has a lower incidence than the group “Children” because “All pediatric groups” includes two studies with very large denominators and low incidence rates that did not break down all pediatric cases into children and infant age subgroups. Pediatric and adult studies often have similar incidences of anesthesia-related cardiac arrest because both have high-risk groups at the extremes of age. The incidence of cardiac arrest is low for both older children and young adults, but the risk of cardiac arrest increases for the youngest children and the eldest adults (see Table 33-1 ).

Review of the effect of the physical condition of the patient shows that the American Society of Anesthesiologists (ASA) physical status correlates with the incidence of cardiac arrest and mortality ( Table 33-2 ). ASA physical status as an indicator of the patient's physical condition represents another risk for both cardiac arrest and mortality. Unfortunately, there is a lack of incidence data for pediatrics, with only one cardiac arrest study and no mortality studies; the table represents the effect of ASA physical status for patients of all ages. “ASA physical status V” patients are not included in most reports of “anesthesia-related” events because by definition they have a low likelihood of survival, making it difficult to determine if events are a result of their condition or related to anesthesia. ASA physical status IV and V patients have a 30 to 300 times greater risk of cardiac arrest than ASA I or II patients ( Rackow et al., 1961 ; Newland et al., 2002 ). Prematurity, congenital heart disease, and congenital defects are common pediatric comorbidities that increase the risk for children ( Morray, 2000) .

TABLE 33-2   -- Incidence of cardiac arrests or mortality by American Society of Anesthesiology physical status for all ages (per 10,000 anesthetic procedures)


Anesthesia Related

All Causes






ASA Class



No.of Studies



No.of Studies



No.of Studies



No.of Studies



0.2 to 0.9



0 to 1.1










0.7 to 7.1



0.4 to 0.9










6.5 to 13



3.3 to 29










19 to 25



19 to 75






















No data available.


The designation of an emergency status to patient's procedure is a third risk factor for both cardiac arrest and mortality. Emergency surgery is associated with a six times increased incidence of anesthetic cardiac arrest (6.5 versus 1.1 per 10,000, P = 0.0001) ( Keenan and Boyan, 1985 ). In addition to a higher incidence of arrests during an emergent procedure, a poorer outcome is also likely ( Vacanti et al., 1970 ; Marx et al., 1973 ; Olsson and Hallen, 1988 ; Morray, 2000 ; Biboulet et al., 2001 ; Newland et al., 2002 ; Sprung et al., 2003 ). It is not clear that the emergent procedures have increased risk because of the patient's condition, the lack of optimal personnel, or both. The extremes of age, ASA physical status, and emergency status are the three most commonly reported risk factors for cardiac arrest and mortality in the operating room.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


The etiologies of cardiac arrest during anesthesia are typically grouped by either the organ systems involved or the interventions applied. The Pediatric Perioperative Cardiac Arrests (POCA) Registry uses a classification system that involves both interventions and organ systems, thus grouping arrests as being either “medication,” “cardiovascular,” “respiratory,” or “equipment” related ( Morray, 2000) . Some etiologies may be difficult to classify because they could fit into several grouping schemes. For example, a succinylcholine-induced dysrhythmia may be classified as either a “medication” or a “cardiovascular” cause of cardiac arrest. A set of guidelines for reporting cardiac arrest data in children, known as the pediatric Utstein guidelines, suggests an organ system-based classification for etiologies ( Zaritsky et al., 1995 ). The Utstein guidelines use three groups, consisting of cardiac, pulmonary, and cardiopulmonary, for the comparison of etiologies of cardiac arrest in children. The Utstein guidelines have not yet been incorporated into much of the anesthesia cardiac arrest literature. A summary from the anesthesia literature of etiologies of cardiac arrest is listed in Box 33-1 (grouped as commonly reported in the anesthesia literature).

BOX 33-1 

Etiologies of Cardiac Arrest During Anesthesia

Medication-Related Causes

Anesthetic overdose (including relative overdose)

Inhalational agent, intravenous agent

Succinylcholine-induced dysrhythmia

Neostigmine-induced dysrhythmia

Medication “swaps”

Drug Reactions

Intravenous administration of local anesthetic intended for the caudal space

High spinal anesthesia[*]

Local anesthesia toxicity

Inadequate reversal of a paralytic (PACU)

Opioid-induced respiratory depression (PACU)

Cardiovascular-Related Causes

Intravascular volume/hemorrhage (inadequate or inappropriate volume administration[*])

Hyperkalemia (succinylcholine, transfusion, reperfusion, myopathy, or renal insufficiency)

Hypocalcemia (citrate intoxication from rapid blood administration)


Vagal episodes (traction, pressures, or insufflation involving the abdomen, eye, neck, or heart)

Central catheter (dysrhythmia, hemorrhage/tamponade)

Anaphylaxis (latex, contrast, drugs, or dextran)

Embolism (air, clot, or fat)

Temperature (malignant hyperthermia, hypothermia)

Myocardial ischemia


Adrenal insufficiency

Respiratory-Related Causes

Inadequate ventilation and oxygenation[*]

“Loss of the airway” (laryngospasm/bronchospasm, a difficult-to-manage anatomy, or a misplaced, kinked, plugged, or inadvertently removed endotracheal tube)

Residual neuromuscular weakness



PACU, causes of arrest likely to occur in the postanesthesia care unit.

*  Causes of cardiac arrest that have low resuscitation and high mortality rates.

“Medication-related” causes are some of the most frequent causes of cardiac arrest related to anesthesia in children and adults, representing approximately 35% (range, 4% to 54%) of the arrests ( Rackow et al., 1961 ; Salem et al., 1975 ; Keenan and Boyan, 1985 ; Olsson and Hallen, 1988 ; Morgan et al., 1993 ; Morray, 2000 ; Biboulet et al., 2001 ; Newland et al., 2002 ; Kawashima et al., 2003 ; Sprung et al., 2003 ). For general anesthetics, the frequently reported causes of cardiac arrest are inhalational agent overdose, intravenous agent overdose, succinylcholine-induced dysrhythmia, neostigmine-induced dysrhythmia, and medication “swaps.” The intravenous administration of local anesthetic intended for the caudal space, high spinal anesthesia, and local anesthesia toxicity is a commonly reported cause of arrest during regional anesthesia. The inadequate reversal of a paralytic agent and opioid-induced respiratory depression are medication-related causes of cardiac arrest that present in the postoperative period.

“Cardiovascular-related” causes of cardiac arrest represent approximately 20% (range, 0% to 45%) of cardiac arrest related to anesthesia in children and adults ( Rackow et al., 1961 ; Salem et al., 1975 ;Keenan and Boyan, 1985 ; Olsson and Hallen, 1988 ; Morgan et al., 1993 ; Morray, 2000 ; Biboulet et al., 2001 ; Newland et al., 2002 ; Kawashima et al., 2003 ; Sprung et al., 2003 ). Arrests due to intravascular volume are the most frequently reported in this group and may include inadequate volume administration, excessive hemorrhage, or inappropriate volume/transfusion administration. The other causes of cardiovascular collapse in this group may or may not involve dysrhythmias. Dysrhythmias may be caused by hyperkalemia seen with succinylcholine, transfusion, reperfusion, myopathy, or renal insufficiency. Dysrhythmia or cardiovascular collapse (asystole) may have a vagal etiology due to traction, pressure, or insufflation involving the abdomen, eye, neck, or heart. Cardiovascular collapse can occur with anaphylaxis from exposure to latex, contrast, drugs, or dextran. Venous air embolism is another important cause of cardiovascular collapse and cardiac arrest under anesthesia. Malignant hyperthermia is an infrequently reported cause of cardiac arrest in this group.

“Respiratory-related” causes involve approximately 31% (range, 11% to 51%) of cardiac arrests related to anesthesia in children and adults ( Rackow et al., 1961 ; Salem et al., 1975 ; Keenan and Boyan, 1985 ; Olsson and Hallen, 1988 ; Morgan et al., 1993 ; Morray, 2000 ; Biboulet et al., 2001 ; Newland et al., 2002 ; Kawashima et al., 2003 ; Sprung et al., 2003 ). Inadequate ventilation and oxygenation are broad categories often expressed in this group of causes of cardiac arrest. “Loss of the airway” may involve laryngospasm/bronchospasm, a difficult-to-manage anatomy, or a misplaced, kinked, plugged, or inadvertently removed endotracheal tube. Aspiration remains a cause of respiratory-related cardiac arrest but appears to be decreasing in occurrence.

“Equipment-related” causes involve approximately 4% (range, 0% to 20%) of cardiac arrest related to anesthesia in children and adults ( Rackow et al., 1961 ; Salem et al., 1975 ; Keenan and Boyan, 1985 ;Olsson and Hallen, 1988 ; Morgan et al., 1993 ; Morray, 2000 ; Biboulet et al., 2001 ; Newland et al., 2002 ; Kawashima et al., 2003 ; Sprung et al., 2003 ). The categories of equipment-related cardiac arrest that are most frequently described involve central line-induced bleeding or dysrhythmias and breathing circuit disconnection. Other groups of cardiac arrest are reported in some studies and include multiple events (3%) ( Morray, 2000) , inadequate vigilance (6%) ( Kawashima et al., 2003 ), or an unclear etiology (9%; range, 1% to 18%) ( Olsson and Hallen, 1988 ; Morray, 2000 ; Biboulet et al., 2001 ).


The determination that a cardiac arrest is “anesthesia related” is subjective, as is the extent to which a cardiac arrest is related to the anesthesia care. Patient-related factors, procedure-related factors, and anesthesia care-related factors are the three most important determinants of the etiology of cardiac arrest in the operating room. Trying to determine the extent of the contribution by anesthesia care has evolved terms such as anesthesia-associated cardiac arrest and anesthesia-attributable cardiac arrest. The determination of anesthesia contribution is complicated by the contribution of the patient and procedure factors. As an example, to what extent is the anesthesia care involved in an arrest related to surgical bleeding in a coagulopathic patient? Is failing to keep up with major hemorrhage or to correct a coagulopathy related to the procedure or patient compared with the anesthesia care? To avoid some of these subjective determinations, many studies use the term anesthesia related to describe a cardiac arrest after an anesthesiologist has been involved in the care of the patient.

Non-anesthesia-related cardiac arrest is most often due to the patient's underlying condition or the procedure being performed. Trauma, exsanguination, and failure to wean from cardiopulmonary bypass are three of the most often reported causes of non-anesthesia-related cardiac arrest. Myocardial infarction, pulmonary embolus, sepsis, and ruptured aneurysm are other, less frequently observed, patient-related causes of cardiac arrest. Procedure-related causes include technical problems, caval compression, vagal asystole related to traction or insufflation, and complications related to transplantation.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


The terminology used to describe outcome from cardiac arrest can be confusing as many different end points are reported. The term survival is commonly used to describe a positive outcome for a patient who sustains a cardiac arrest. The term survival is imprecise as to the duration or quality of the patient's outcome. A patient may “survive” initial resuscitation attempts but subsequently die in the intensive care unit from persistent hemodynamic instability or devastating neurologic injury. The term return of spontaneous circulation (ROSC) describes the immediate success of the resuscitation effort (initial survival). ROSC means that the native heartbeat and blood pressure are adequate for a period of at least 20 minutes. Although ROSC indicates a successful reversal of cardiac arrest, it may not be a meaningful indicator if many patients subsequently die in the intensive care unit. The number of patients with ROSC after cardiac arrest is usually much greater than the number of patients who have a longer, more meaningful, period of survival such as survival to discharge from the hospital. Although “survival to discharge from the hospital” indicates a longer survival period than ROSC, surviving a longer time period does not address the quality of that outcome. An assessment of the quality of survival should acknowledge the presence either of a new neurologic deficit or of the patient's return to baseline neurologic status. These terms are found in some of the descriptions in the anesthesia literature of the outcome of children who sustain cardiac arrest ( Table 33-3 ).

TABLE 33-3   -- Outcome from cardiac arrest


Anesthesia Related

All Causes









No.of Studies



No.of Studies



No.of Studies



No.of Studies

Return of spontaneous circulation


83 to 100



78 to 95



32 to 70



47 to 65




0 to 100



8 to 55



60 to 88



35 to 72


Survival to discharge


0 to 100



45 to 92



12 to 40



28 to 41


New neurologic deficits


0 to 16



4 to 7



12 to 15



2 to 7


Return to baseline neurologic status


21 to 92



44 to 85



20 to 25



30 to 35




It is presumed that the duration and quality of survival from arrest in the operating room should be good because the personnel who witness the arrests and provide the resuscitation are trained and prepared. A review of the anesthesia literature reveals that cardiac arrest can be reversed in approximately 90% of the anesthesia-related episodes. The likelihood of ROSC drops from about 90% to 50% to 60% if the causes of the arrest include those that are not related to anesthesia. The patient's survival to hospital discharge after an anesthetic-related cardiac arrest appears to be approximately 65% to 68% (the range for pediatric studies of this variable is very large). This survival to discharge falls to 30% if nonanesthesia causes of cardiac arrest are included. Comparing these data with descriptions from the nonanesthesia literature reveals that studies of “in-hospital” cardiac arrests in children show a survival to discharge percentage of 23% (range, 8% to 42%) ( Gillis et al., 1986 ; Von Seggern et al., 1986 ;Davies et al., 1987 ; Carpenter and Stenmark, 1997 ; Para et al., 2000; Suominen et al., 2000 ; Reis et al., 2002 ). In-hospital cardiac arrests are comparable to perioperative arrests because both include the presence of trained witnesses. This 23% survival to discharge rate is comparable to the 30% rate for “all causes” and much lower than the 65% rate for “anesthesia-related” causes of cardiac arrest in the operating room. The vigilance and training of the witnesses are expected to be higher when anesthesiologists are involved. The presence of anesthesiologists may account, in part, for the better survival outcomes in anesthesia-related cardiac arrests.

Outcome studies should include a determination of the presence of new neurologic injuries due to the hypoperfusion of the brain during the period of cardiac arrest. Pediatric studies from the nonanesthesia literature of “in-hospital” cardiac arrests show a 71% favorable neurologic outcome for the survivors (range, 45% to 90%) ( Gillis et al., 1986 ; Davies et al., 1987 ; Carpenter and Stenmark, 1997 ; Para et al., 2000; Suominen et al., 2000 ; Reis et al., 2002 ). The anesthesia literature shows that 57% of the children survive and return to their baseline neurologic status, whereas 5% survive with a new neurologic deficit (see Table 33-3 ). This frequency of 57% at baseline of the 62% total for survivors gives a rate of 92% with a favorable neurologic outcome in survivors of an “anesthesia-related” cardiac arrest. This percentage for pediatric survivors falls to 22% with return to baseline of 36% total survivors or a 61% favorable neurologic outcome when “all causes” of cardiac arrest are included. The 71% favorable neurologic outcome for in-hospital cardiac arrests is comparable to the 61% rate for “all causes” and lower than the 92% rate for “anesthesia-related” causes of cardiac arrest in the operating room. Note that the number of studies and patients for these estimates are small and the ranges are large. These data indicate that both the duration and quality of survival are favorable for children who arrest from anesthesia-related causes.

There are many potential explanations for the increased resuscitation rate (ROSC, approximately 92% versus 51%) from “anesthesia-related” cardiac arrest. Factors such as the resuscitation skills of the anesthesiologist, the preparation by the anesthesiologist for emergencies, the reversibility of the causes of cardiac arrest in the operating room, and the increased monitoring during anesthesia to provide early recognition of problems may contribute to improved resuscitation rates during anesthesia care. The survival rate after cardiac arrest is affected by many factors, some of which are the same that predispose a patient to cardiac arrest. The risk factors for cardiac arrest, consisting of the patient's age, his or her ASA physical status, and emergency procedures, are also determinants of mortality (see Tables 33-1 and 33-2 [1] [2]). The etiology of the cardiac arrest also impacts the likelihood of both resuscitation and survival. The mortality is increased if the etiology of the cardiac arrest is hemorrhage or is associated with protracted hypotension (both P < 0.001) ( Girardi and Barie, 1995 ; Newland et al., 2002 ; Sprung et al., 2003 ).

Resuscitation-related factors also have an effect on outcome. The rhythm during resuscitation and the duration of the resuscitation attempts have been related to the outcome of the patient. “No-flow” and “low-flow” states occur during the arrest and resuscitation process. A no-flow state occurs when a patient is in cardiac arrest before receiving resuscitation efforts. A low-flow state occurs when a patient is arrested and receiving resuscitation that is unable to provide adequate circulation. The longer the patient is in a “no-flow” state or a “low-flow” state, the worse the outcome is likely to be.

Asystole is a rhythm that, if present during resuscitation, has been associated with a decreased incidence of both ROSC and survival to discharge for children with cardiac arrest outside of the operating room. Usually, asystole is due to a prolonged period of hypoxia or ischemia of the myocardium and represents a terminal rhythm. The prolonged hypoxia causes the myocardium to be more resistant to resuscitation efforts. The prolonged global hypoxia leading to asystole usually also causes severe neurologic injury. If the heart can be resuscitated, there is still the likelihood of a poor outcome. Asystole in the operating room is much more likely to be associated with a good outcome. In the operating room, the continuous monitoring of the patient decreases the risk of prolonged periods of hypoxia or ischemia. Instead of a terminal rhythm, asystole in the operating room is often an initial rhythm caused by a vagal stimulation. As an initial rhythm, asystole is much more likely to be reversed. Usually, discontinuation of the vagal stimulus and support of the heart rate chemically are effective resuscitation measures. Unlike non-operating room arrests, asystole is a commonly reported rhythm with anesthesia-related cardiac arrest and is associated with a good prognosis ( Sprung et al., 2003 ).

The duration of the resuscitative efforts also has an effect on the patient's outcome. Prolonged duration of CPR increases the possibility of low-flow intervals resulting in myocardial and cerebral injury. The need for CPR for longer than 15 minutes has been determined to be a predictor of mortality in anesthesia-related cardiac arrests (P < 0.001) ( Girardi and Barie, 1995 ). The interpretation of these data is complicated by the possibility of successful outcomes even after very prolonged periods of resuscitation efforts. Up to 3 hours of CPR has been reported in anesthetic-related cardiac arrests with eventual resuscitation and a good outcome ( Cleveland et al., 1971 ; Lee et al., 1994 ). The rapid initiation of resuscitative efforts by the anesthesia team can result in effective, even if prolonged, CPR. In summary, the etiology of the arrest, the rhythm disturbance, and the duration of CPR have an impact on the outcome from cardiac arrest in the operating room.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


“Anesthesia-related” cardiac arrest is preventable 53% of the time, and “anesthesia-related” mortality is preventable 22% of the time ( Kawashima et al., 2003 ). Human error may be the most important factor in deaths attributable to anesthesia and manifests not as a fundamental ignorance but as a failure in the application of existing knowledge ( Olsson and Hallen, 1988 ). Poor preoperative preparation and inadequate vigilance are frequently reported errors that could be avoided. Examples of poor preoperative preparation relevant to the pediatric anesthesiologist include failure to identify patients with unrecognized skeletal myopathy, coronary involvement with Williams syndrome, prolonged QT syndrome, and undiagnosed cardiomyopathy. Another category of preventable causes is inadequate vigilance such as failure to recognize progressive bradycardia and failure to respond to persistent hypotension. In addition to improving preparation and vigilance, the use of test dosing and divided dosing when administering medications (especially to unstable patients) is suggested to minimize medication errors. Other important and preventable etiologies of “anesthesia-related” cardiac arrest include transfusion-related hyperkalemia, local anesthetic toxicity, and inhalational anesthetic overdose ( Morray, 2000) .

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


Successful resuscitation of a victim of cardiopulmonary arrest requires that the anesthesiologist: (1) recognizes inadequate ventilation or circulation and the need for CPR and (2) reestablishes and maintains both ventilation and circulation until the underlying pathology can be corrected. These steps are the same for all resuscitations whether they occur on the street or in the operating room. The more rapidly these steps are achieved, the better is the outcome.


The decision to begin CPR for a child in the operating room requires the recognition that the patient's vital signs are inadequate. An understanding of the cerebral and cardiac perfusion requirements in children is necessary to determine the adequacy of the vital parameters. Exact data are lacking to address these requirements for the wide range of patients who come under the pediatric anesthesiologist's care. Specific pediatric experience and training provide the background necessary to make decisions about these perfusion requirements. A lower rate of CPR in children receiving anesthetics by anesthesiologists trained in pediatric anesthesiology suggests that the knowledge of the appropriate hemodynamic variables in children may reduce the need for resuscitation ( Keenan et al., 1991 ).

CPR should be started immediately when the circulation is inadequate to deliver oxygen, substrates, or resuscitative drugs to the heart or brain. The presence of inadequate ventilation or circulation should be evident in the patient undergoing extensive monitoring in the operating room. In the absence of extensive monitoring, health care workers should use palpation of the umbilical artery for the newly born, of the brachial artery for an infant, and of the carotid artery for a child to detect a heart rate abnormality ( Cavallaro and Melker, 1983 ; Lee and Bullock, 1991 ; American Heart Association, 2000) . The presence of either a heart rate or a blood pressure alone should not result in an assumption that the circulation is sufficient as both an adequate heart rate (>60 beats per minute) and signs of good perfusion are necessary (adequate blood pressure values in Table 33-4 ).

TABLE 33-4   -- Adequate vital signs for children


Heart Rate (bpm)[*]

Blood Pressure (mm Hg)








100 to 170


SBP <60



65 to 140


SBP <70



65 to 140


SBP <70 + (2 . age in years)

>10 Years


60 to 120


SBP <90

Modified from American Heart Association in collaboration with the International Liaison Committee on Resuscitation: Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support. Circulation 102:I291, 2000.

SBP, systolic blood pressure.



Note: Significant bradycardia is typically defined as 2 SDs below the low end of normal.




The finding that exhaled air from the rescuer (16% oxygen) may provide adequate oxygenation of the victim (SaO2 of ≥90%) ( Elam et al., 1954 ) is the basis for bystander CPR when supplemental oxygen is not available. The administration of 100% oxygen via tracheal intubation during CPR helps to maximize oxygen delivery to the vital organs. The need to optimize oxygen delivery outweighs the risk of oxygen toxicity during resuscitation, and 100% oxygen should be used when available. The rare exception is the child with a circulatory condition such as a hypoplastic left heart whose poor systemic perfusion may be the result of pulmonary overcirculation. In such a case, the anesthesiologist needs to decide whether high levels of oxygen administration contribute to the poor systemic circulation.

Initially, researchers believed that closed-chest compression alone provided adequate ventilation to the victim requiring CPR ( Kouwenhoven et al., 1960 ). The possibility that compressions alone may provide ventilation for adult victims has resulted in over-the-telephone instruction for compressions-alone CPR to untrained bystanders or those unwilling to provide mouth-to-mouth ventilation. Unfortunately, soft tissue obstruction may prevent adequate ventilation in humans without intubation and positive-pressure ventilation ( Safar et al., 1961 ). An unprotected airway puts patients at greater risk for aspiration during CPR. Unlike fibrillatory arrests, a model of asphyxial arrest shows the greatest benefit with the combination of compressions and ventilations compared with compression or ventilation alone ( Berg et al., 2000 ). Tracheal intubation is the optimal means to ensure ventilation during CPR for pediatric anesthesiologists because they maintain the training to do so.

The laryngeal mask airway (LMA) is an airway adjunct that the anesthesiologist may be using when a child experiences a cardiac arrest. The LMA compares favorably to mouth-to mouth, mask ventilation, and other airway adjuncts during CPR ( Rumball and MacDonald, 1997 ; Stone et al., 1998 ). There are limited data available for a comparison of LMA to tracheal intubation during CPR ( Samarkandi et al., 1994 ). Tracheal intubation is considered a more stable and protective airway than the LMA during resuscitation. Airway adjuncts remain not recommended as a replacement for tracheal intubation during CPR in children, especially when an anesthesiologist is available ( Grayling et al., 2002 ). The tracheal tube (TT) remains the airway of choice for patients requiring CPR, and appropriate placement can be verified by the presence of end-tidal CO2 (ETCO2).

The incidence of accidentally placing a TT in the esophagus of a child is greater during an arrest than during a nonarrest intubation (19% to 26% esophageal intubation rate during arrest versus 3% for nonarrest situations) ( Bhende et al., 1992 ; Bhende and Thompson, 1995 ). The demonstration of persistent (after six ventilations) ETCO2 after intubation is extremely reliable to confirm correct placement of the TT in children with spontaneous circulation ( Bhende et al., 1992 ). The lack of presence of ETCO2 in the TT implies esophageal intubation but can be misleading during cardiac arrest. After cardiac arrest, the decreased pulmonary blood flow produced during CPR causes ETCO2 to be falsely low or absent in correctly placed TTs (14% to 15% of TTs had no ETCO2 in arrested children receiving CPR) (Bhende et al., 1992 ; Bhende and Thompson, 1995 ). Continually detectable ETCO2 is proof of tracheal intubation even during arrest. In the absence of ETCO2 and unless pulmonary circulation is restored, the TT placement should be visually inspected to discriminate esophageal intubation. The TT also provides access to the circulation for drug administration ( Table 33-5 ).

TABLE 33-5   -- Vascular access during cardiopulmonary resuscitation



Peripheral venous access (IV)

Route of first choice if vascular access not present.

Rapidly and easily placed.

Any drug or fluid may be administered by this route.

Need to flush each drug with 0.25 mL/kg normal saline to ensure central delivery (20 mL in adults)

Intraosseous access (IO)

Easier to obtain in <6 years old; can use for any age.

Any drug or fluid may be administered by this route.

Use flush as with peripheral venous access.

Intratracheal route (TT)

Use only if no IV or IO access.

Only administer lidocaine, atropine, naloxone, and epinephrine (LEAN drugs) by TT.

Note: TT drug delivery requires 2 to 10 times IV dose.

Use 1 to 2 mL of normal saline in TT to increase distribution into distal bronchial tree (10 mL in adults).

Central venous catheter

Central access is first choice if already in place.

Place if no IV or IO access is obtained.

Requires flush if catheter tip is below diaphragm.

Cutdown saphenous

Use when other options have failed.

Requires special skill; high complication rate.

PACU, causes of arrest likely to occur in the postanesthesia care unit.

*Causes of cardiac arrest that have low resuscitation and high mortality rates.




A comparison of different patterns of ventilation during chest compression revealed differences in oxygenation, ventilation, and hemodynamics ( Wilder et al., 1963 ). Both oxygen administration without pressure and continuous positive airway pressure produce adequate oxygenation but not ventilation or hemodynamic effects. Ventilation independent of compression, ventilation interposed between compressions, and ventilation synchronized with compression allow both adequate oxygenation and ventilation, but their effects on hemodynamic pressures vary. Delivery of positive-pressure ventilation has an impact on the hemodynamic variables due to the changes in intrathoracic pressure. Simultaneous compression and ventilation yield improvement in blood flow and survival in this canine model but has not shown the same benefit in humans and remains an experimental mode of CPR. The current recommendations are that in unintubated patients, ventilations should be interposed between compressions to maximize effectiveness of the ventilation, at least until the airway is secured ( American Heart Association et al., 2000) . The recommendations for the ratio of chest compressions to ventilation during CPR vary with the age of the child, whether the airway is secured, and the number of rescuers. The younger the child, the greater the need for an increase in the number of ventilations during CPR. The neonate and newly born require a 3:1 ratio or about 90 compressions and 30 breaths per minute whether one or two rescuers are used and whether or not the child is intubated. The infant and child aged 1 to 8 years require a 5:1 ratio or about 100 compressions and 20 breaths per minute whether one or two rescuers are used and whether or not the child is intubated. The older child and adult should receive a 15:2 ratio or about 90 compressions and 12 breaths per minute for both one and two rescuers if not intubated; the older child or adult should receive a 5:1 ratio for two rescuers once intubated.

The decreased compression-to-ventilation ratios used for the youngest victims allow more ventilation and may be more effective in the age groups at highest risk of respiratory etiology of arrest. The decreased compression ratio in the older victim allows the delivery of more uninterrupted compressions. Data in animals suggest that longer periods of uninterrupted compressions result in increased coronary perfusion (Kern et al., 1992, 1998 [185] [184]). Interposed ventilation interrupts chest compression and, if prolonged, causes a reduction in CPR-generated perfusion pressure and recovery time to regain perfusion pressure when compressions are restarted ( Berg et al., 2001 ).

Other significant interruptions to chest compressions include patient assessment for return or spontaneous circulation, intubation, central line placement, and placement of adhesive pads for defibrillation. These interruptions should be minimized or compressions continued when possible. The accomplishment of 80 chest compressions per minute has been correlated with successful resuscitation ( Yu et al., 2002 ). Once the patient is intubated, the pauses for ventilations can be minimized and the need to interrupt compressions decreased. In the operating room, the airway should be secured as soon as possible and the compressions performed without interruption. The anesthesiologist can use the age-specific ratios of compressions to ventilations or adjust them as appropriate for the etiology of the arrest ( Table 33-6 ).

TABLE 33-6   -- Basic life support procedures


Newly Born (<12 hr old)

Infant (<1 yr old)

Child (1 to 8 yr old)

Adult (>8 yr old)

Ventilation rate, breaths/min





Pulse check

Umbilical cord




Compression area

Below nipples

Lower 1/2 of sternum

Lower 1/2 of sternum

Lower 1/2 of sternum

Compress with

Two fingers/encircle

Two fingers/encircle

One hand

Two hands

Compression depth

1/3 AP diameter

1/3 to 1/2 AP diameter

1/3 to 1/2 AP diameter

1/3 to 1/2 AP diameter

Compression rate,/min


100 minimum



Compression-to-ventilation ratio




15:2 (5:1 if intubated)

Data from American Heart Association in collaboration with the International Liaison Committee on Resuscitation: Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support. Circulation 102:I291, 2000.





Mechanisms of Blood Flow During Cardiopulmonary Resuscitation

Kouwenhoven and others (1960) proposed that external chest compressions squeeze the heart between the sternum and the vertebral column, forcing blood to be ejected. This assumption of direct cardiac compression during external CPR became known as the cardiac pump mechanism of blood flow. The cardiac pump mechanism proposes that the atrioventricular (AV) valves close during ventricular compression and that ventricular volume decreases during ejection of blood. During chest relaxation, ventricular pressures fall below atrial pressures, enabling the AV valves to open and the ventricles to fill. This sequence of events resembles the normal cardiac cycle and occurs with the direct cardiac compression during open-chest CPR.

Several observations of the hemodynamics during external CPR are inconsistent with the cardiac pump mechanism for blood flow ( Table 33-7 ). First, similar elevations in the arterial and venous intrathoracic pressures during closed-chest CPR suggest a generalized increase in intrathoracic pressure ( Weale and Rothwell-Jackson, 1962 ). Second, reconstructing the integrity of the thorax of patients with flail sternums improves the blood pressure during CPR (unexpected, because a flail sternum should allow direct cardiac compression during closed-chest CPR) ( Rudikoff et al., 1980 ). Third, patients who develop ventricular fibrillation (VF) produce sufficient blood flow to maintain consciousness by repetitive coughing or deep breathing ( MacKenzie et al., 1964 ; Criley et al., 1976 ; Niemann et al., 1980 ; Harada et al., 1991 ). These observations suggest a generalized increase in intrathoracic pressure may contribute to the production of blood flow during CPR. The finding that changes in intrathoracic pressure without direct cardiac compression (i.e., a cough) produce blood flow epitomizes the thoracic pump mechanism of blood flow during CPR. Extensive research has addressed the involvement of the cardiac and thoracic pump mechanisms in blood flow during CPR.

TABLE 33-7   -- Comparison of mechanisms of blood flow during closed-chest compressions


Cardiac Pump

Thoracic Pump

Proposed Mechanism


Sternum and spine compress heart

General increase in intrathoracic pressure

Findings During Compression

Atrioventricular valves


Stay open

Aortic diameter



Blood movement

Left ventricle to aorta

Pulmonary veins to aorta

Ventricular volume


Little change

Compression rate


Little effect

Duty cycle

Little effect


Compression force

Increases role

Decreases role

Clinical Situations


Small chest

Large chest


High compliance

Low compliance



Thoracic Pump Mechanism

Chest compressions during CPR generate almost equal pressures in the left ventricle, aorta, right atrium, pulmonary artery, airway, and esophagus ( MacKenzie et al., 1964 ; Criley et al., 1976 ; Chandra et al., 1981a ; Niemann et al., 1981a ; Cohen et al., 1982 ; Raessler et al., 1988 ; Swenson et al., 1988 ; Paradis et al., 1989 ). Because all intrathoracic vascular pressures are equal, the suprathoracic arterial pressures must be greater than the suprathoracic venous pressures for a cerebral perfusion gradient to exist. Venous valves, either functional or anatomic, prevent the direct transmission of the elevation in intrathoracic pressure to the suprathoracic veins. These jugular venous valves are present in animals ( Rudikoff et al., 1980 ; Chandra et al., 1980, 1981a [58] [61]; Niemann et al., 1981b ; Fisher et al., 1982; Guerci et al., 1985 ; Criley et al., 1986 ; Gudipati et al., 1986 ) and humans ( Niemann et al., 1981b ; Swenson et al., 1988 ; Paradis et al., 1989 ; Chandra et al., 1990 ; Goetting and Paradis, 1991a, 1991b [130] [131]). This unequal transmission of the intrathoracic pressure to the suprathoracic vasculature establishes the gradient necessary for cerebral blood flow during closed-chest CPR.

During normal cardiac activity, the lowest pressure occurs on the atrial side of the AV valves, providing a downstream effect that allows venous return to the pump. The extrathoracic shift of this low-pressure area to the cephalic side of the jugular venous valves during the thoracic pump mechanism of blood flow implies that the heart is merely a conduit during this mechanism. Angiographic studies show blood passing from the vena cavae, through the right heart, to the pulmonary artery and from the pulmonary veins, through the left heart, to the aorta during a single chest compression ( Niemann et al., 1981b ; Cohen et al., 1982 ). Unlike normal cardiac activity and open-chest CPR, echocardiographic studies during closed-chest CPR in both dogs ( Niemann et al., 1981b ; Cohen et al., 1982 ) and humans ( Rich et al., 1981 ; Werner et al., 1981 ; Clements et al., 1986 ) have shown that the AV valves are open during blood ejection. In addition, unlike native cardiac activity and open-chest CPR, aortic diameter decreases instead of increasing during blood ejection ( Niemann et al., 1981b ; Werner et al., 1981 ). These findings during closed-chest CPR support the thoracic pump theory and suggest that the heart becomes a passive conduit for blood flow.

Cardiac Pump Mechanism

Despite evidence for the importance of the thoracic pump mechanism of blood flow during external chest compressions, there are specific situations in which the cardiac pump mechanism predominates during closed-chest CPR. First, applying more force during chest compressions (as in High Impulse Cardiopulmonary Resuscitation, see later) increases the likelihood of direct cardiac compression. Increasing the force of chest compressions in animals undergoing CPR increases the closure of the AV valves, implying more direct cardiac compression ( Feneley et al., 1987 ; Hackl et al., 1990 ). Second, a smaller chest size seems to allow more direct cardiac compression. Adult dogs with small chests have better hemodynamics during closed-chest CPR than do dogs with large chests ( Babbs et al., 1982a ). Third, the very compliant infant chest should permit more direct cardiac compression. During closed-chest CPR in an infant swine model, excellent blood flows are produced compared with most adult models ( Schleien et al., 1986 ). Unlike the adult model, the addition of simultaneous ventilation with compression (SCV) does not augment the flow produced during piglet CPR ( Berkowitz et al., 1989 ).SCV-CPR fails to augment the already high flows that occur in small dogs with conventional chest compression ( Babbs et al., 1982a ). The lack of contribution of SCV-CPR in the infant or small adult animal models implies that excellent compression (probably direct cardiac) occurs and that the additional intrathoracic pressure is of no benefit.

Transesophageal echocardiography studies demonstrate the closing of the AV valves during the compression phase of CPR in humans ( Higano et al., 1990 ; Kuhn et al., 1991 ) and during thoracic pump models of CPR in animals ( Beattie et al., 1991 ). These findings support the occurrence of cardiac compression during conventional CPR, suggesting both mechanisms of blood flow may occur during CPR. As seen later, varying the CPR technique may alter the contribution of each mechanism.

Distribution of Blood Flow During Cardiopulmonary Resuscitation

The overall blood flow to the tissues is decreased during CPR compared with the normal physiologic state. A redistribution of the blood flow during CPR optimizes perfusion to the heart and brain. This redistribution toward the vital organs should enhance outcome. Maintenance of myocardial blood flow during CPR is necessary for the ROSC, and maintenance of cerebral blood flow determines the quality of the eventual neurologic outcome. The distribution of blood flow to both the heart and brain during CPR is influenced by the development of regional gradients.

The distribution of blood flow to the brain depends on the development of three regional gradients: intrathoracic-suprathoracic, intracranial-extracranial, and caudal-rostral gradients. The first gradient, intrathoracic-suprathoracic, provides the flow of oxygenated blood from the chest to the upper extremities and head. Either venous collapse, secondary to the elevated intrathoracic pressure, or closure of anatomic valves in the jugular system prevents the transmission of intrathoracic pressure to the suprathoracic venous system ( Rudikoff et al., 1980 ; Niemann et al., 1981b ; Fisher et al., 1982 ). When CPR is effective, arterial collapse does not occur and elevated intrathoracic pressure results in a gradient that promotes suprathoracic blood flow.

The second gradient, intracranial-extracranial, directs blood to the brain away from the extracranial suprathoracic vessels and toward the intracranial vessels. The α-adrenergic agonist vasoconstrictors constrict the extracranial vessels but have little effect on the intracranial vessels, resulting in increased intracranial blood flow. Use of the a-agonist vasoconstrictor epinephrine increases intracranial blood flow while decreasing flow in the cephalic skin, muscle, and tongue ( Schleien et al., 1986 ).

The third gradient, caudal-rostral, occurs within the intracranial vessels. CPR alone seems to increase the distribution of flow to caudal areas of the brain, and ischemia preceding CPR significantly increases the distribution of flow to these areas ( Michael et al., 1984 ; Shaffner et al., 1998, 1999 [333] [332]). This pattern of caudal redistribution of flow also occurs in other models of global ischemia and may provide relative sparing of the brainstem ( Jackson et al., 1981 ).

Myocardial blood flow does not have the advantage of the large extrathoracic pressure gradient that augments cerebral flow. The thoracic pump generates equal increases in all intrathoracic structures. This lack of a gradient causes poor myocardial blood flow during external chest compressions. Several studies have shown much lower blood flow to the myocardium compared with the cerebrum during closed-chest CPR ( Ditchey et al., 1982 ; Michael et al., 1984 ; Schleien et al., 1986 ).

The type of CPR influences the production of myocardial blood flow. Methods that are more likely to cause direct cardiac compression, such as high-impulse CPR, result in increased myocardial blood flow ( Ditchey et al., 1982 ; Maier et al., 1984 ). Myocardial blood flow may be present only during relaxation of chest compression ( Maier et al., 1984 ), correlating with a “diastolic” pressure ( Cohen et al., 1982 ) or, in other methods, during compression, correlating with a “systolic” pressure ( Michael et al., 1984 ; Schleien et al., 1986 ). Regional flow within the heart also changes during CPR, with a shift in the ratio of subendocardial-subepicardial blood flow from the normal 1.5:1 to 0.8:1 ( Schleien et al., 1986 ). This ratio reverts to normal with epinephrine administration.

Blood flow to organs other than the heart or brain falls dramatically during CPR. The lack of valves in the infrathoracic veins causes retrograde transmission of venous pressure and decreases the gradient for blood flow below the diaphragm in animals ( Brown et al., 1987b ). Regional blood flows for infrathoracic organs (small intestine, pancreas, liver, kidney, and spleen) during CPR are usually less than 20% of prearrest rates and often close to zero (Koehler and Michael, 1985a, 1985b [191] [192]; Voorhees et al., 1983 ; Michael et al., 1984 ; Sharff et al., 1984 ). The addition of abdominal compressions does not alter the infrathoracic organ blood flow (Koehler and Michael, 1985a, 1985b [191] [192]; Voorhees et al., 1983 ). Administration of epinephrine during closed-chest CPR almost eliminates flow to the subdiaphragmatic organs, with the exception of the adrenal glands ( Ralston et al., 1984 ).

Few data are available regarding pulmonary blood flow during CPR. Pulmonary blood flow occurs primarily at times of low intrathoracic pressure during closed-chest CPR ( Cohen et al., 1982 ). High extrathoracic venous pressure builds up during compression and results in pulmonary filling during relaxation as intrathoracic pressure falls. Resuscitation methods that lower intrathoracic pressure may augment pulmonary vascular filling.

Rate and Duty Cycle

In 1986, the American Heart Association Guidelines for CPR and Emergency Cardiac Care recommended increasing the rate of chest compressions from 60 to 100 per minute. This change represented a compromise between advocates of the thoracic pump mechanism and those of the cardiac pump mechanism ( Feneley et al., 1988 ). The mechanics of these two theories of blood flow differ, but a faster compression rate could augment both. It is necessary to understand the concepts of compression rateduty cycle, and compression force to understand the mechanics of CPR.

Compression rate is the number of cycles per minute. Duty cycle is the ratio of the duration of the compression phase to the entire compression-relaxation cycle expressed as a percentage. For example, a rate of 60 compressions per minute (total cycle, 1 second), a 0.6-second compression time, produces a 60% duty cycle (0.6 second/1.0 second = 60%). The impact of duty cycle differs between the two mechanisms of blood flow (see Table 33-7 ). Compression force is the pressure and the acceleration applied to the chest.

If direct cardiac compression generates blood flow (cardiac pump mechanism), then the force of the compression determines the stroke volume. Prolonging the compression (increasing the duty cycle) beyond the time necessary for full ventricular ejection fails to produce any additional increase in stroke volume in this model. In contrast, increasing the rate of compressions increases cardiac output since a fixed ventricular blood volume ejects with each cardiac compression. Therefore, in the cardiac pump mechanism, blood flow is rate sensitive but duty cycle insensitive.

If the thoracic pump mechanism is producing blood flow, the reservoir of blood to be ejected is the large capacitance of the intrathoracic vasculature. With the thoracic pump mechanism, increasing either the force of compression or the duty cycle enhances flow by emptying more of the large intrathoracic capacity. Changes in the compression rate have less effect on flow over a wide range of rates ( Halperin et al., 1986a ). Blood flow in the thoracic pump mechanism is duty cycle sensitive but rate insensitive.

Mathematical models of the cardiovascular system confirm that both the applied force and the duration of compression determine blood flow with the thoracic pump mechanism ( Beyar et al., 1984 ;Halperin et al., 1987 ). Animal data suggest that either the thoracic pump or the cardiac pump mechanism can effectively generate blood flow during closed-chest CPR. Discrepancies among the results of various studies can be attributed to differences in CPR models and in compression techniques. These differences may involve issues of chest compliance and geometry, maturity of different animal species, or chest compression techniques. For example, either mechanism may come into play in an infant with a very compliant chest wall. Differences in techniques may include the magnitude of sternal displacement, compression force, compression rate, and duty cycle.

Several studies in dogs show a benefit of a fast compression rate (120 per minute) over slower rates during conventional CPR ( Maier et al., 1984 ; Feneley et al., 1988 ; Sanders et al., 1990 ). Studies in piglets ( Dean et al., 1990 ), puppies ( Fleisher et al., 1987 ), and humans ( Taylor et al., 1977 ; Ornato et al., 1988 ; Chandra et al., 1990 ) find no difference in the effectiveness of conventional CPR at various rates. A piglet CPR study found that the duty cycle was the major determinant of cerebral perfusion pressure ( Chandra et al., 1981a ). The duty cycle at which venous return becomes limited varies with age. Increasing the duty cycle is more effective in younger piglets and more likely to limit venous return in the adult models.

The discrepancy between the importance of rate and duty cycle in various models by different investigators generates confusion. 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. This is true because it is physically easier for a rescuer to produce a 50% duty cycle at a rate of 100 than at 60 compressions per minute (holding compression is physically difficult at slow rates). This is the reason behind the rate change in the 1986 American Heart Association guidelines for CPR. This increased rate continues to be recommended ( American Heart Association et al., 2000) .

Chest Geometry

Chest geometry plays an important role in the ability of chest compressions to generate intrathoracic pressures. Shapecompliance, and deformability are chest characteristics with a significant impact during CPR. The age of the patient affects each of these characteristics, which may explain some differences in CPR between the pediatric and adult models.

Chest Shape

During anterior-to-posterior-delivered compressions, the change in cross-sectional area of the chest relates to its shape ( Fleisher et al., 1987 ). The thoracic index refers to the ratio of the anteroposterior diameter to the lateral diameter. A keel-shaped chest, as 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 of less than 1. A circular chest would have a thoracic index of 1. A circular chest with the same perimeter would also have a larger cross-sectional area than either elliptical chest shape. As an anteroposterior compression flattens a circle, it decreases the cross-sectional area and compresses the contents. In contrast, as a keel-shaped chest approaches a circular shape, the cross-sectional area increases during the application of anteroposterior compression. The cross-sectional area of the keel-shaped chest does not decrease until the compression continues past the circular shape to flatten the chest. This implies a threshold distance past which the compression must proceed before the intrathoracic contents are compressed ( Dean et al., 1987 ). The rounded, flatter chests of small dogs and pigs may require less displacement than the keel-shaped chests of adult dogs to generate thoracic ejection of blood. The rounded chests of small dogs improve the efficacy of external thoracic compression compared with the keel-shaped chests of adult dogs ( Babbs et al., 1982a ).

Chest Compliance

With increasing age, the cartilage in the chest becomes calcified and the compliance changes. The stiffer or less compliant, older chest may require greater compression force to generate the same anteroposterior displacement. Three-month-old swine require a much greater pressure for anteroposterior displacement than do their 1-month-old counterparts ( Dean et al., 1987 ). The compliance of the chest affects not only the amount of displacement but also the structures compressed. Direct cardiac compression is more likely to occur in the more compliant chests of younger animals. Cerebral blood flow production in a piglet model of external CPR was much greater than expected compared with the usual findings in adult animals ( Schleien et al., 1986 ). The more compliant infant chest may allow more direct cardiac compression, accounting for the high flows that resemble those produced by open-chest cardiac massage.

Chest Deformation

Chest deformation occurs as CPR becomes prolonged. The chest assumes a flatter shape as compressions continue, producing larger decreases in cross-sectional area at the same 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 release of compression. Decreased chest recoil with progressing deformation limits the displacement and produces less effective compression. A pediatric model of conventional CPR shows a progressive decrease over time in the effectiveness of chest compressions to produce blood flow ( Schleien et al., 1986 ; Dean et al., 1991 ). The permanent deformation of the chest in this model approaches 30% of the original anteroposterior diameter. Attempting to limit the deformation by increasing intrathoracic pressure from within during compression with SCV-CPR was ineffective ( Berkowitz et al., 1989 ). Neither the amount of deformation nor the time to deterioration of flow was different. In an attempt to limit the production of deformation, investigators used a third mode of infant animal CPR with a vest to deliver compressions. The vest distributes compression force diffusely around the thorax and greatly decreases permanent deformation (3% versus 30%) ( Fisher et al., 1982 ; Shaffner et al., 1990 ). Unfortunately, the deterioration of blood flows with time still occurs and appears to be unrelated to the amount of deformation in this model.

The relevance to humans of chest geometry characteristics found in animal studies is unclear. Body weight, surface area, chest circumference, and chest diameter did not correlate with the aortic pressure produced during CPR in a study of nine adults already declared dead ( Swenson et al., 1988 ). There has not been a direct comparison of adult and pediatric human CPR. The increased compliance and deformability of the infant chest make it likely that CPR would be more effective in children than in adults (as seen in animal models).


Efficacy of Blood Flow During Cardiopulmonary Resuscitation

The level of blood flow to the vital organs produced by conventional closed-chest CPR without pharmacologic support is disappointingly low. The range of cerebral blood flow in dogs is 3% to 14% of prearrest levels ( Bircher and Safar, 1981 ; Koehler and Michael, 1985a ; Koehler et al., 1983 ; Jackson et al., 1984 ; Luce et al., 1984 ). Cerebral perfusion pressures are also low, 4% to 24% of prearrest levels in animals ( Bircher and Safar, 1981 ; Koehler and Michael, 1983a, 1983b ; Luce et al., 1984 ) and only 21 mm Hg in humans ( Goetting and Paradis, 1991b) . Myocardial blood flows in this basic CPR mode are also discouragingly low at 1% to 15% of prearrest levels in dogs ( Chandra et al., 1981a ; Voorhees et al., 1983 ; Koehler and Michael, 1985a ; Halperin et al., 1986a ; Shaffner et al., 1990 ). Myocardial perfusion pressures (MPPs) correlate with myocardial blood flow. Plotting myocardial blood flow in milliliters per minute per gram versus MPP in mm Hg gives a slope of 0.01 to 0.015 (Voorhees et al., 1983 ; Ralston et al., 1984 ). These data imply a one-to-one relationship between myocardial blood flow (when measured in mL/min per 100 g) and MPP (mm Hg).

In addition to pharmacologic support, several other factors affect cerebral and myocardial blood flow during CPR. These factors include the victim's age, intracranial pressure, duration of CPR, and duration of preresuscitation ischemia.

Age affects cerebral blood flow during closed-chest CPR. Two-week-old piglets have substantially higher cerebral blood flows (50% of prearrest) and slightly higher myocardial flows (17% of prearrest) than those reported for adult models ( Schleien et al., 1986 ). Two studies on slightly older pigs yielded opposing results. Cerebral blood flows in these two studies were 26% to 95% and 1% to 4% of prearrest values, with corresponding myocardial values of 2% to 8% and 1% to 6% ( Sharff et al., 1984 ; Brown et al., 1987b ). The cerebral blood flow in the first of these two studies was markedly higher than in adult models during closed-chest CPR, and neither of the myocardial flows was different from that in adult models. No human data exist with blood flows at different ages during CPR.

Intracranial pressure can represent the downstream pressure for cerebral blood flow and, if elevated, can inhibit cerebral perfusion. Increasing intrathoracic pressure with closed-chest CPR causes intracranial pressure to increase ( Rogers et al., 1979 ). This relationship is linear; one third of the increase in intrathoracic pressure is transmitted to the intracranial pressure ( Guerci et al., 1985 ). The carotid arteries and jugular veins do not appear to be involved in the transmission of intrathoracic pressure to the intracranial contents. The transmission can be partially blocked by occluding the cerebrospinal fluid or vertebral vein flow ( Guerci et al., 1985 ). The rise in intracranial pressure with chest compressions becomes more significant in the setting of baseline increased intracranial pressure (two thirds of the intrathoracic pressure is transmitted to the intracranial pressure). Clinicians need to be aware that the efficacy of CPR to perfuse the brain deteriorates markedly in the face of elevated intracranial pressure. When possible, the intracranial pressure should be lowered early in the resuscitation (i.e., shunt tapped, hematoma drained) to increase the effectiveness of resuscitative efforts.

Increased duration of CPR has a negative effect on cerebral blood flow and seems to be most detrimental in the infant preparation ( Sharff et al., 1984 ; Schleien et al., 1986 ). The length of the ischemic period before CPR begins also has a negative effect on cerebral blood flow ( Lee et al., 1984 ; Shaffner et al., 1999 ). Forebrain blood flow during subsequent CPR is reduced more than brainstem as the preceding ischemic interval is increased (Shaffner et al., 1998, 1999 [333] [332]). Hypothermia has some protective effect and prevents this reduction in the ischemic intervals tested in dogs ( Shaffner et al., 1998 ). The cause of these detrimental effects on cerebral blood flow is unclear. It remains obvious that a short ischemic period and quick resuscitation improve eventual outcome.

There appear to be thresholds for minimal vital organ blood flow during CPR. The inability to maintain blood flow above these thresholds during CPR results in organ malfunction. A myocardial blood flow of 20 mL/min per 100 g or greater is necessary for successful defibrillation in dogs ( Guerci et al., 1985 ; Sanders et al., 1985a ). A cerebral blood flow of greater than 15 to 20 mL/min per 100 g is necessary to maintain normal electrical activity during CPR ( Michael et al., 1984 ).

Monitoring the Effectiveness of Cardiopulmonary Resuscitation

Monitoring the effectiveness of CPR can be difficult. An adequate MPP is necessary to allow the ROSC during CPR. The above data suggest that a myocardial blood flow of 20 mL/min per 100 g is necessary for the resumption of myocardial activity. This flow would correlate with an MPP of 20 mm Hg (aortic relaxation pressure minus right atrial relaxation pressure). Data from CPR in humans show that an MPP of 15 mm Hg was necessary for, but did not guarantee, ROSC ( Paradis et al., 1990 ). Often right atrial relaxation pressure is low and the aortic “diastolic” pressure represents the MPP. In clinical practice, detection of this “diastolic” pressure is difficult without an arterial catheter. With the use of an arterial catheter during CPR, the anesthesiologist can observe the effects of interventions on the “diastolic” pressure as a guide to optimizing the impact of the resuscitative efforts on myocardial perfusion.

Measurement of venous oxygen saturation has been described as a method to monitor the effectiveness of CPR. In humans, the level of venous oxygen saturation correlated with the ROSC ( Snyder et al., 1991 ; Rivers et al., 1992 ). This observation may be of use during CPR in victims with central venous access. Patients with a mixed venous oxygen saturation of less than 30% are receiving inadequate CPR and are unlikely to have an ROSC ( Rivers et al., 1992 ). If the patient has a central catheter in place during CPR, this method can be useful in determining the effectiveness of resuscitative efforts.

Another method to monitor the effectiveness of myocardial perfusion during CPR is to follow the production of ETCO2. ETCO2 monitoring is particularly useful to determine the effectiveness of CPR as the patient is more likely to have a TT than an arterial catheter or a central catheter in place. The amount of CO2 exhaled and measured in the TT depends on the amount of pulmonary blood flow. In general, ETCO2 levels decrease as pulmonary blood flow falls during CPR. As blood flow to the heart and lungs improves during CPR, ETCO2 should return toward arterial CO2 levels.

An ETCO2 measured during CPR of less than 10 mm Hg predicts an inability to restore spontaneous circulation ( Callaham and Barton, 1990 ; Wayne et al., 1995 ; Levine et al., 1997 ). Alternatively, an ETCO2 greater than 15 mm Hg during CPR predicts resuscitation ( Sanders et al., 1989 ; Barton and Callaham, 1991 ; Bhende and Thompson, 1995 ). These studies imply that ETCO2 monitoring is useful in determining the effectiveness of CPR and that the production of ETCO2 of less than 10 to 15 mm Hg suggests a need to modify CPR to improve blood flow. ETCO2 may also detect cardiac output during electromechanical dissociation ( Barton and Callaham, 1991 ), the ROSC during CPR ( Garnett et al., 1987 ), and the presence of spontaneous circulation during cardiopulmonary bypass ( Gazmuri et al., 1991 ).

The measurement of ETCO2 for patients receiving CPR without tracheal intubation appears useful to determine the effectiveness of cardiac compressions at producing blood flow ( Nakatani et al., 1999 ). In nonintubated patients receiving CPR with a facemask or an LMA, the level of ETCO2 correlated with the rate of ROSC. The correlation was not as close as reports from studies of intubated patients. More patients with a low ETCO2 had ROSC and fewer with higher ETCO2 had ROSC than in studies with intubated patients.

There are pitfalls in the monitoring of ETCO2 during CPR. Administration of bicarbonate increases CO2 production and may elevate ETCO2 without a corresponding increase in pulmonary blood flow. Administration of epinephrine may decrease ETCO2 despite an increase in the myocardial perfusion, causing a misinterpretation that CPR has become less effective ( Martin et al., 1990 ). Contamination of disposable ETCO2 detectors by resuscitation medications (epinephrine, atropine, or lidocaine) or gastric acid may decrease their accuracy in the assessment of CPR effectiveness or the detection of esophageal intubation ( Muir et al., 1990 ). The cause of the arrest may also have an impact on the interpretation of quantitative ETCO2 analysis during CPR. The initial ETCO2 measurements are higher following asphyxial arrest than fibrillatory arrest, and the ability to predict ROSC from initial ETCO2 levels was better with fibrillatory arrest ( Grmec et al., 2003 ).

An additional monitoring tool during CPR is the electrocardiogram (ECG) when the arrest is ventricular fibrillation. Fibrillation with large amplitude and long latency (coarse fibrillation) represents less cellular ischemia and a more resuscitatable situation. As myocardial perfusion continues to deteriorate, the fibrillation becomes finer and the myocardium less responsive to resuscitation. Effective CPR can reverse a fine fibrillation to a coarser pattern. Analyzing fibrillation waveforms may allow the rescuer to deliver defibrillation attempts when optimal ( Hayes et al., 2003 ).

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier



Conventional CPR includes closed-chest compressions delivered manually with ventilations interposed after every 5th or 15th compression (see Table 33-6 for basic life support procedures). This method of CPR can be delivered in any setting without additional equipment and with a minimum of training. No large randomized study exists to demonstrate the superiority of any alternative method of CPR over conventional CPR.

Rescuer fatigue is a major problem with manual CPR in the field. Individual variation among rescuers performing manual CPR is another problem both in the field and in the laboratory. Mechanical devices are available to deliver chest compressions to prevent fatigue and to standardize compression delivery. Mechanical devices are presently limited to adult CPR and are not recommended for children (American Heart Association, 2000) . The overall low efficacy of conventional CPR has led to investigations of multiple CPR modalities. These methods usually reflect attempts to enhance the contribution of the thoracic pump or cardiac pump to blood flow during CPR ( Table 33-8 ). For example, the use of both hands to encircle the chest of an infant while using the thumbs to apply sternal compression attempts to both raise intrathoracic pressure and compress the heart ( Todres and Rogers, 1975 ; David, 1988 ). This two-thumb encircling technique of CPR generates higher blood pressures and is recommended over the two-finger technique for infants ( Dorfsman et al., 2000 ).

TABLE 33-8   -- Contribution of cardiac or thoracic pump to various methods of cardiopulmonary resuscitation

Cardiopulmonary Resuscitation Method

Cardiac Pump

Thoracic Pump

Open chest



High impulse






Abdominal binding



Interposed abdominal compression[*]



Simultaneous compression













Also moves blood by abdominal compression alone.



Conventional CPR is usually performed with sternal compressions applied to a supine patient. Patients who are prone and undergoing posterior cranial or spine surgery when they sustain a cardiac arrest may not be able to be quickly repositioned to supine. An unstable spine, protruding stabilizers, and ongoing blood loss are factors that may delay turning the patient and necessitate starting CPR in the prone position. Two methods for posterior chest compression in the prone position have been suggested. The first uses two hands, one spread over each scapula ( Tobias et al., 1994 ), and the second uses the heel of one hand on the spine (second hand on top of the first, similar to sternal compression) ( Sun et al., 1992 ; Dequin et al., 1996 ). Counterpressure under the sternum with another rescuer's hand or fist or a sandbag has also been suggested ( Sun et al., 1992 ; Dequin et al., 1996 ; Mazer et al., 2003 ). When possible, patients should be turned supine for CPR.

In addition to chest compressions, defibrillation may be necessary in a prone patient. Prior placement of self-adhesive defibrillation pads is advised as intraoperative placement of paddles may be difficult and requires time. The placement of gel pads on the back on either side of the surgical incision in left mid-axillary lines and inferior to right scapula has been successful for defibrillating a prone patient (Miranda and Newton, 2001 ). Prone defibrillation may be attempted if the defibrillator is available and the patient cannot be turned immediately. Risk factors for cardiac arrest in the prone position are listed in Table 33-9 .

TABLE 33-9   -- Risk factors for cardiac arrest in the prone position



Cardiac anomalies in patients undergoing major spinal surgery






Air embolism



Wound irrigation with H2O2



Poor positioning



Occluded venous return

From Brown J, Rogers J, Soar J: Cardiac arrest during surgery and ventilation in the prone position: A case report and systematic review. Resuscitation 50:233, 2001.





Simultaneous compression-ventilation CPR (SCV-CPR) represents an attempt to augment conventional CPR by increasing the contribution of the thoracic pump mechanism to blood flow. Delivering ventilation simultaneously with every compression (instead of interposed after every fifth compression) adds to the intrathoracic pressure and should augment blood flow produced by conventional closed-chest CPR. Several studies suggest that SCV-CPR increases the carotid blood flow compared with conventional CPR ( Harris et al., 1967 ; Chandra et al., 1980, 1981a [58] [61]; Bircher and Safar, 1981 ;Babbs et al., 1982a ). Studies show an advantage of SCV-CPR in large canine models (Koehler and Michael, 1985a, 1985b [191] [192]; Luce et al., 1983 ) but no advantage over conventional CPR in infant pigs ( Berkowitz et al., 1989 ) and small dogs (Babbs et al., 1982a, 1982b [10] [9]; Sanders et al., 1982 ). In small animals, the compliance of the chest may allow more direct cardiac compression and higher intravascular pressure than with conventional CPR ( Schleien et al., 1986 ; Dean et al., 1987, 1990 [90] [91]). Human studies comparing SCV to conventional CPR show minimal improvement ( Harris et al., 1967 ) or detrimental effect ( Martin et al., 1986 ) on the coronary perfusion pressure. Survival is worse in both animals ( Sanders et al., 1982 ) and humans ( Krischer et al., 1989 ) when SCV-CPR is compared with conventional CPR. No study has shown an increased survival with this CPR technique.


Researchers have used abdominal binders and military anti-shock trousers (MAST) to augment closed-chest CPR. Both methods apply continuous compression circumferentially below the diaphragm. Abdominal binding augments CPR by (1) decreasing the compliance of the diaphragm, which results in increased intrathoracic pressure; (2) forcing blood out of the subthoracic structures, which increases the circulating blood volume (an autotransfusion effect); and (3) increasing the resistance in the subdiaphragmatic vasculature, which increases suprathoracic blood flow. The increases in intrathoracic pressure and blood volume lead to increased aortic pressure and carotid blood flow in both animals ( Lee et al., 1981 ; Koehler and Michael, 1985a, 1985b [191] [192]; Niemann et al., 1984 ) and humans (Chandra et al., 1981b ; Lilja et al., 1981 ).

Unfortunately, as the aortic pressure increases, the right atrial “diastolic” pressure increases to a greater extent, resulting in a decrease in the coronary perfusion pressure ( Sanders et al., 1982 ; Niemann et al., 1984 ). This deterioration of coronary perfusion pressure is coincidental with a decreased myocardial blood flow ( Niemann et al., 1984 ). This technique also decreases the cerebral perfusion pressure via transmission of the intrathoracic pressure to the intracranial vault, raising the intracranial pressure ( Guerci et al., 1985 ). The use of abdominal binders or MAST to augment CPR does not increase survival in clinical studies ( Sanders et al., 1982 ; Mahoney and Mirick, 1983 ; Niemann et al., 1990 ). Liver laceration from abdominal binder CPR has been reported ( Harris et al., 1967 ) but is no more frequent than with conventional CPR ( Redding, 1971 ; Rudikoff et al., 1980 ; Mahoney and Mirick, 1983 ; Niemann et al., 1984 ). There are no data to support the use of these techniques clinically during CPR, and the potential for complications should discourage their application.


Interposed abdominal compression CPR (IAC-CPR) is the delivery of an abdominal compression during the relaxation phase of chest compression. IAC-CPR may augment conventional CPR (1) by increasing venous return to the chest during the abdominal compression/chest relaxation phase and “priming the pump” ( Ralston et al., 1982 ; Voorhees et al., 1983 ), (2) through abdominal compression during IAC-CPR, increasing intrathoracic pressure and adding to the duty cycle of the chest compression ( Einagle et al., 1988 ), and (3) via the effect of abdominal compression on the aorta, which may send blood retrograde to the carotids or coronaries ( Ralston et al., 1982 ). Several studies have shown hemodynamic improvements secondary to IAC-CPR. In animals, cardiac output and cerebral and coronary blood flow improved when IAC-CPR was compared with conventional CPR in adult models ( Ralston et al., 1982 ; Voorhees et al., 1983 ; Walker et al., 1984 ; Einagle et al., 1988 ) but not in an infant swine model ( Eberle et al., 1990 ). Human studies have also shown an increase in aortic pressure and coronary perfusion pressure during IAC-CPR compared with conventional CPR ( Berryman and Phillips, 1984 ; Howard et al., 1984, 1987 [169] [168]; Ward et al., 1989 ; Barranco et al., 1990 ; Chandra et al., 1990 ).

Although one study reports a 10% aspiration rate ( Walker et al., 1984 ), most report no aspiration or liver lacerations ( Voorhees et al., 1983 ; Berryman and Phillips, 1984 ; Mateer et al., 1985 ; Einagle et al., 1988 ; Ward et al., 1989 ; Barranco et al., 1990 ; Sack et al., 1992 ). Clinically, IAC-CPR requires extra manpower or equipment and remains experimental. Outcome studies have mixed results, showing no increase in survival with prehospital arrests but increased survival with in-hospital arrests ( Mateer et al., 1985 ; Sack et al., 1992 ). Although an alternative technique for in-hospital CPR in adults, a lack of data prevents a recommendation for the use of IAC-CPR in children.


Vest CPR uses an inflatable bladder that is wrapped circumferentially around the chest and is cyclically inflated. This method of delivering chest compressions by diffuse application of pressure has two unique characteristics. First, the increase in intrathoracic pressure occurs with only minimal change in chest dimensions, making direct cardiac compression unlikely (an almost purely thoracic pump technique). Second, the diffuse distribution of pressure decreases the likelihood of trauma. Vest CPR in dogs improves cerebral and myocardial blood flows and survival compared with conventional CPR (Luce et al., 1983 ; Criley et al., 1986 ; Halperin et al., 1986a, 1986b [151] [148]). In a pediatric model of vest CPR, only a 3% permanent chest deformation occurred after 50 minutes of vest CPR ( Shaffner et al., 1990 ) compared with almost 30% deformation produced by an equivalent period of conventional CPR ( Schleien et al., 1986 ). In humans, vest CPR increases aortic systolic pressure but does not significantly increase diastolic pressure compared with conventional CPR ( Swenson et al., 1988 ). In a preliminary study of vest CPR in victims of out-of-hospital arrest, increased aortic and coronary perfusion pressures were demonstrated, and there was a trend toward a greater ROSC compared with standard CPR ( Halperin et al., 1993 ). A large clinical trial is under way to determine if these benefits improve outcome.

The lack of metallic parts has allowed vest CPR to be used experimentally during nuclear magnetic resonance spectroscopy to study brain intracellular pH ( Eleff et al., 1992 ). In addition, the vest has been used as an external cardiac assist device in nonarrested dogs with heart failure ( Beyar et al., 1989 ; Chandra et al., 1991 ). Clinically, the use of vest CPR depends on sophisticated equipment and the technique remains experimental at this time.


High-impulse CPR involves the application of greater than usual force during chest compression. This increase in force can be in the form of greater mass, greater velocity, or both. It is hypothesized that the larger impulses result in greater chest deflection, causing more contact with the heart ( Kernstine et al., 1982 ). Direct cardiac compression is more likely with this form of closed-chest CPR. High-impulse CPR can generate myocardial blood flows as high as 60% to 75% of prearrest values ( Maier et al., 1984 ). In humans, high-impulse CPR generates increased aortic pressures ( Swenson et al., 1988). An outcome study in dogs compared high-impulse CPR with conventional closed-chest CPR and found no significant improvement in resuscitation, survival, or neurologic outcome ( Kern et al., 1986 ).


Active compression-decompression (ACD) CPR requires a device that attaches to the chest and allows the rescuer to pull up on the sternum and decompress the thorax between compressions. The theoretical advantages of decompressing the chest between compressions include restoring chest wall shape and creating a negative intrathoracic pressure that pulls gas into the lungs and pulls blood into the intrathoracic vessels. These characteristics allow for more effect from the subsequent compression as more intrathoracic pressure can be generated and more blood ejected. Preliminary studies in humans have shown that after advanced cardiac life support failed, ACD-CPR was more effective than standard CPR at improving hemodynamic variables ( Cohen et al., 1992 ). Following a witnessed in-hospital arrest, more patients had ROSC and survival at 24 hours and had a better Glasgow Coma Scale score when they received ACD-CPR than when standard CPR was applied ( Cohen et al., 1993 ).

A larger study of in-hospital cardiac arrest victims failed to show any difference in the resuscitation or outcomes between patients receiving ACD-CPR or standard CPR ( Stiell et al., 1996 ). Several large studies of patients who sustained an out-of-hospital cardiac arrest did not find a difference in the effectiveness of ACD-CPR or standard CPR for improving the incidence of ROSC, hospital admission, hospital discharge, or short-term neurologic outcome ( Lurie et al., 1994 ; Schwab et al., 1995 ; Mauer et al., 1996 ; Stiell et al., 1996 ; Nolan et al., 1998 ).

Complication rates following CPR were not different following ACD-CPR or standard CPR in most studies ( Lurie et al., 1994 ; Schwab et al., 1995 ; Mauer et al., 1996 ). It is interesting that the same study that showed that ACD had more complications than standard CPR (hemoptysis and sternal dislodgment) was also one of the few large studies that found ACD-CPR more effective than standard CPR for out-of-hospital arrests ( Plaisance et al., 1997 ). ACD-CPR is considered an optional technique for adults, but the absence of clinical data prevents a recommendation for children.

Impedance threshold valve (ITV) describes the use of a device on the TT or facemask that impedes the inflow of inspiratory gas during reexpansion of the chest between CPR compressions when not actively ventilating the patient. The impedence of gas inflow promotes negative intrathoracic pressure development during chest reexpansion. This increase in the negative intrathoracic pressure facilitates blood return to the thorax with chest recoil prior to the next chest compression ( Lurie et al., 2002 ). The use of an ITV has been shown to improve coronary perfusion pressure and vital organ blood flow with both standard and ACD-CPR in adult and pediatric animal models ( Langhelle et al., 2002 ; Voelckel et al., 2002 ). Improved levels of ETCO2, diastolic and coronary perfusion pressure occurred in a prospective, randomized controlled trial in adults with ACD-CPR with ITV compared with ACD-CPR without ITV. A decrease in the time to achieve an ROSC was also seen with ACD-CPR with ITV (Plaisance et al., 2000 ). A prospective controlled trial comparing standard CPR and ACD-CPR with ITV found significantly improved short term survival (24 hours) in adult patients ( Wolcke et al., 2003 ). Further studies are needed to determine the effectiveness of the use of an ITV for pediatric resuscitation.


Another manual method of CPR known as phased chest and abdominal compression-decompression (PCACD) cardiopulmonary resuscitation is being investigated ( Tang et al., 1997 ). PCACD CPR resembles a combination of active compression-decompression CPR and interposed abdominal compression CPR. It requires a device that attaches to both the abdomen and chest and is used to alternately compress and reexpand both structures. It offers the theoretical advantages of both methods because chest shape is restored and blood and gas are pulled into the thorax during active decompression of the chest and blood flow is augmented due to the compression, and active decompression, of the abdomen. Coronary perfusion pressure, ROSC, short-term survival, and neurologic outcome were improved in a porcine model of fibrillatory cardiac arrest resuscitated using PCACD CPR ( Tang et al., 1997 ).


Open-chest CPR involves a thoracotomy and the application of direct compression of the heart to generate blood flow. The use of this technique requires a high level of preparation and training as well as special equipment and facilities. These requirements limit open-chest CPR to certain hospital settings.

Open-chest CPR represents a model of the cardiac pump mechanism for generation of blood flow. In theory, this model eliminates the production of intrathoracic pressure, which, if transmitted, could reduce the gradients for blood flow. This enhanced gradient combined with directly applied compression can result in near-normal blood flows. In experimental models, open-chest CPR produces cardiac outputs of 25% to 61% of prearrest values ( Weiser et al., 1962 ; Bircher and Safar, 1981 ; Bartlett et al., 1984 ). These studies and others demonstrate cardiac outputs two to three times larger than those with conventional closed-chest CPR ( Weiser et al., 1962 ; Del Guercio et al., 1965 ; Bircher et al., 1980 ; Bircher and Safar, 1981 ; Bartlett et al., 1984 ). Increases in cerebral perfusion pressure have been significant in some studies ( Bircher et al., 1980 ) but not in others ( Del Guercio et al., 1965 ; Bircher and Safar, 1981 ). MPPs are significantly increased compared with closed-chest CPR ( Bircher et al., 1980 ; Sanders et al., 1984b ). Cerebral blood flow in dogs of 150% of prearrest values can be produced with open-chest CPR ( Jackson et al., 1984 ). Cross-clamping the descending aorta during open-chest CPR further increases carotid blood flow.

Survival in dogs can be improved by use of open-chest CPR after inadequate closed-chest CPR ( Sanders et al., 1984a ). Dogs with MPP of less than 30 mm Hg after 15 minutes of closed-chest external CPR received 2 to 4 minutes of either open-chest or closed-chest external CPR before defibrillation was attempted. The dogs that received open-chest CPR had significantly greater MPPs and survival rates.

The length of time of closed-chest CPR affects the success of subsequent open-chest CPR ( Sanders et al., 1985b ). After 20 and 25 minutes of closed-chest CPR, the success rate of open-chest CPR decreased to 38% and 0%, respectively. This implies that the benefits from open-chest CPR are time limited and that early application is crucial. There are no data to recommend the routine use of open-chest CPR in the pediatric patient. Postoperative cardiac patients with a recent sternotomy may benefit from open-chest CPR. They have easier access and can be inspected for tamponade, and suture lines can be inspected and avoided.


Cardiopulmonary bypass (CPB) is a very effective way to restore circulation after cardiac arrest. Animal studies show that CPB increases 72-hour survival and recovery of consciousness and preserves myocardium better than conventional CPR ( Levine et al., 1987 ; Pretto et al., 1987 ). In dogs, CPB results in better neurologic outcome than continued conventional CPR after a 4-minute ischemic period (neurologic outcome was dismal in both groups when the ischemic period lasted for 12 minutes) ( Levine et al., 1987 ; Pretto et al., 1987 ). Twenty-four-hour survival is possible for at least 90% of dogs after 15 or 20 minutes of cardiac arrest but for only 10% of dogs after 30 minutes of arrest with CPB stabilization during defibrillation ( Reich et al., 1990 ). CPB decreases myocardial infarct size in a model involving coronary artery occlusion, compared with conventional CPR ( Angelos et al., 1990 ). In most animal models, CPB facilitates resuscitation and improves success compared with conventional CPR.

There is increasing experience with CPB for cardiac arrest in humans outside the cardiac operating room. Timely application of percutaneous femoral artery and vein bypass has been successful in resuscitating patients with “refractory” cardiac arrest. Unfortunately, many patients who are stabilized on CPB after standard CPR fails cannot be weaned off CPB or have a low likelihood of long-term survival or of good neurologic outcome ( Mattox and Beall, 1976 ; Phillips et al., 1983 ; Hartz et al., 1990 ; Reichman et al., 1990 ; Martin et al., 1998 ). There are reports of patients with cardiac arrest in the operating room or catheterization suite who have cardiac arrest under anesthesia and fail to respond to conventional CPR but benefit from the institution of CPB. These patients are reported to have good neurologic outcomes despite over 30 minutes (even over 2 hours and transfer to another facility) of failed conventional resuscitation efforts ( Lee et al., 1994 ; Cochran et al., 1999 ).

CPB rescue for cardiac arrest requires considerable technical support and experience. It is remarkable that the procedure can be fully operational in less than 10 minutes after it is requested ( Mattox and Beall, 1976 ; Phillips et al., 1983 ). Despite rapid availability and restoration of circulation, the lack of effective resuscitation before institution of CPB reduces the ability to preserve neurologic or cardiac function. Because of these limitations, CPB may have limited value for patients who sustain out-of-hospital cardiac arrest or require more than 30 minutes of conventional CPR ( Hartz et al., 1990 ;Tisherman et al., 1991 ; Martin et al., 1998 ). The patient who arrests in the operating room may be the ideal candidate for CPB rescue because intervention and application of CPR are usually immediate. For many patients, extracorporeal membrane oxygenation (ECMO) can extend the CPR duration to 50 to 60 minutes with an acceptable survival and neurologic outcome rates ( del Nido et al., 1992 ; Dalton et al., 1993 ; Duncan et al., 1998 ; Chen et al., 2003 ).

The anesthesiologist can make a decision about attempting CPB rescue of cardiac arrest that is based on the reversibility of the cause of arrest and the effectiveness of the ongoing CPR. A patient with a cardiac arrest from a very reversible cause, with immediate and effective CPR, may be helped by CPB rescue even if the CPR has been longer than 60 minutes.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier



Emergency situations require prioritization of events and escalation to invasive measures ( Fig. 33-1 ). Vascular access is crucial to the effective administration of drugs and fluids for resuscitation but may be difficult to achieve in pediatric victims. During cardiac arrest, attempts to obtain peripheral venous access in infants and children should be limited to 90 seconds. After 90 seconds, an intraosseous catheter should be placed, percutaneous central venous access attempted, or the administration of drugs in the TT started.


FIGURE 33-1  Algorithm for peripheral vascular access during emergencies in pediatric patients.  (From Chameides L, Hazinski MF, editors: Vascular access. In Pediatric advanced life support. Dallas, TX, 1997, American Heart Association.)




During CPR, the optimal location of intravascular catheterization provides ready access to the anesthesiologist and minimizes the interruption of resuscitation efforts. Peripheral venous access, femoral central access, and intraosseous access can usually be accomplished without interruption of airway management or chest compressions. The use of a saline flush for medications administered in peripheral access, intraosseous access, and central lines with the catheter tip below the diaphragm improves drug delivery to the heart in the low-flow state of CPR. A flush with 5 to 20 mL of normal saline administered into the catheter should drive the medication up the vein into the central circulation (0.25 mL/kg was effective in an animal model). For most CPR events, peripheral vascular access should be adequate for resuscitation medication if obtainable (see Table 33-5 ).


Intraosseous cannulation provides a rapid, safe, and reliable route to vascular access via the marrow venous plexus. Intraosseous access can be accomplished within 30 to 60 seconds ( American Heart Association, 2000) . All drugs, crystalloids, colloids, and blood can be administered via this route. The preferred site of placement is the anterior tibia, but other sites include the distal femur, medial malleolus, iliac crest, and, in older children and adults, the distal radius, distal ulna, proximal tibia, and sternum (risk of cardiac laceration) ( Glaeser et al., 1993 ; Guy et al., 1993 ; Waisman and Waisman, 1997 ; Calkins et al., 2000 ) ( Fig. 33-2 ). Specially designed intraosseous needles should be readily available in the operating room for such emergencies. Complications from intraosseous placement are rare (<1% of patients) but include bone fracture, compartment syndrome, fluid extravasation, osteomyelitis, and fat embolism ( Orlowski et al., 1989 ).


FIGURE 33-2  Intraosseous needle insertion into the proximal tibia.




Intratracheal administration of drugs may be used for lipid-soluble resuscitation drugs. As most anesthetized children have this route available, it must be considered early, especially if vascular access is a problem. Drugs that can be administered via the trachea are lidocaine, epinephrine, naloxone (remember the pneumonic “LEAN”). Animal studies suggest that standard intravenous doses given via the trachea achieve serum concentrations approximately 10% of that of intravenously administered drugs ( Kleinman et al., 1999 ); the recommended intratracheal dose of epinephrine is 10 times the intravascular dose. The technique for administration is to flush the drug with 2 to 5 mL of normal saline into the endotracheal tube and provide five manual ventilation breaths to deliver the medication into the distal airways and alveoli. This technique is favored over catheter or feeding tube delivery because of ease and practicality ( Jasani et al., 1994 ).


Based on the initial findings of the POCA Registry ( Morray, 2000) , hemorrhage and inadequate fluid therapy accounted for 41% of arrests and contributed to 18% of reported pediatric deaths from cardiac arrest. Postarrest fluid status can often be difficult to evaluate, necessitating the need for more invasive monitoring (such as an arterial and central venous catheters) and secure large-bore vascular access for continued resuscitation. In the setting of hemorrhage or surgical bleeding, for every 1.5 mL/kg lost, 5 mL/kg of isotonic fluid or 1:1 mL/kg of red blood cells should be administered ( Barcelona and Coté, 2001 ). Such bleeding, if anticipated, often necessitates a large-bore catheter to permit such rapid fluid and blood product administration. Complications of any fluid resuscitation, whether colloid or crystalloid, are electrolyte abnormalities, hypothermia, metabolic abnormalities, coagulopathies, and hemoglobin abnormalities (both high and low).

Isotonic crystalloid solutions are preferred for volume expansion. Normal saline or Ringer's lactate should be used. Early volume expansion has been shown to prevent progression of patients to shock and cardiac arrest ( Carcillo et al., 1991 ). Dextrose-containing fluids should not be used for the initial fluid resuscitation because of the complications of hyperglycemia and secondary osmotic diuresis. Further, hyperglycemia prior to cerebral ischemia may worsen neurologic outcome ( Cherian et al., 1997 ). Dextrose-containing fluids should be used to treat documented or suspected hypoglycemia (see section on glucose). The use of colloid fluids (e.g., albumin, dextran, hetastarch, etc.) has the theoretic benefit of improving oncotic pressure as well as volume expansion, although studies have failed to prove a clear benefit over crystalloid ( Schierhout and Roberts, 1998 ). Caution should be taken in aggressively fluid-resuscitating a postarrest patient with poor urine output as acute tubular necrosis or other forms of renal insufficiency often occur. If low blood pressure and poor perfusion persist after adequate pharmacologic and fluid resuscitation, other causes should be sought and consideration given to evaluation of cardiac function (pulmonary artery catheter placement, echocardiogram, or both).

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier



Physiology of Bradycardia

Bradyarrhythmias are frequent complications in children in the perioperative period. Bradycardia is the rhythm seen most often before cardiac arrest in pediatric patients. Most bradycardic events are precipitated by hypoxemia, vagal stimuli, or the side effects of medications. In the operating room, hypoxia and the depressant effects of anesthetic agents were identified as the probable causes of intraoperative arrest ( Morray, 2000) . Sinus bradycardia results from slowed or suppressed sinoatrial node depolarization. However, other pathologic causes can occur and should be considered in the settings of trauma, cardiac surgery, and structural heart disease. A sufficiently slow heart rate in an infant or a child that results in hemodynamic instability and fails to respond to ventilation, oxygenation, epinephrine, atropine, and chest compressions is an indication for electrical pacing. The probability of successfully pacing after cardiac arrest decreases with increasing duration of CPR ( Quan et al., 1992 ;Beland et al., 1988 ). If a bradyarrhythmia persists despite resuscitative and electrical support, other forms of cardiovascular support should be sought, such as cardiopulmonary bypass or extracorporeal membrane oxygenation therapies. These therapies, although not curative, can provide the time needed to diagnose the cause and allow time for recovery ( Fig. 33-3 ).


FIGURE 33-3  PALS bradycardia algorithm.



Sinus Bradycardia

Great variability exists in infant and pediatric heart rates (see Table 33-4 ). Lower limits of heart rates serve as guidelines but may not be indicative of pathology. Diagnosis of sinus bradycardia implies a normal p wave is present on 12-lead ECG evaluation. Hemodynamic instability may or may not be present and is not essential for the diagnosis. Underlying etiologies should be sought, such as increased vagal tone (from suctioning), elevated intracranial pressure, hypoxia, metabolic abnormalities (hypokalemia or hypercalcemia), and drug effects.

Sinus and Atrioventricular Block

Sinus block or sinus node arrest is characterized by absence of p waves from the ECG and are usually related to a toxic ingestion or underlying cardiac disease. This can also present as asystole. Often suppression of the sinus node results in activation of extranodal pacing with variable electrical activity. Recognition and prompt initiation of transcutaneous pacing are crucial and can be lifesaving (Beland et al., 1998).

Various degrees of heart block will appear as a bradyarrhythmia ( Fig. 33-4A ). Often, 12-lead ECG is needed to diagnose properly the changes in the PR interval and potential conduction delays. First-degree heart block is the prolongation of the PR interval beyond normal. Normal PR interval can vary with age and is usually less than 0.16 second. Such prolongation can be normal or a sign of underlying disease or medication effects. Most patients with first-degree heart block are hemodynamically stable and require monitoring for progression to other forms of heart block. No emergent intervention is required.


FIGURE 33-4  Arrhythmias from delayed atrioventricular node conduction.



Second-degree AV block is characterized by the frequent occurrence of a p wave without a QRS complex. Progressive lengthening of the PR interval until this occurs is called a Wenkebach (Mobitz type 1) rhythm ( Fig. 33-4B ). If the QRS complex is repeatedly absent without a change in the PR interval, it is a Mobitz type 2 rhythm ( Fig. 33-4C ). Both are forms of second-degree heart block caused by delayed or absent conduction through the AV node. Some medications, hypoxia, myocarditis, and other conditions can create this problem. As long as the ventricular rate maintains adequate perfusion, no immediate treatment is necessary and diagnosis with elimination of the causative factor can be curative. However, if the patient is not stable hemodynamically, attempts to increase conduction by increasing heart rate with atropine or epinephrine may be useful and a pediatric cardiologist should be consulted.

Third-degree AV block, or complete heart block, is characterized by electrical and mechanical dissociation between the atrial and ventricular rates (see Fig. 33-4 D). ECG analysis demonstrates p waves at one rate and QRS complexes at a separate rate. As with other forms of heart block, finding and eliminating the causative agent is curative. If the patient is hemodynamically stable, observation is often suitable. However, if the rhythm becomes unstable, some form of supportive pacing is required. If the third-degree block is congenital in origin or is persistent from surgical correction of a congenital heart lesion, a permanent pacemaker is often required.

Treatment of Bradycardia

The treatment of all causes of bradycardia should focus on the underlying causes. If the patient is hemodynamically unstable, immediate measures should be taken to improve the heart rate. The first line of therapy is administration of a vagolytic agent and stopping all sedatives/medications that may be aggravating the bradycardia. The POCA Registry demonstrated epinephrine with or without atropine was most successful in returning stable heart rate and circulation ( Morray, 2002 ). If the patient is hypoxic, improved oxygenation and oxygen delivery often result in improvement in heart rate. Bradycardia secondary to vagal stimulus can be treated by the removal of the stimulus and the administration of a vagolytic agent. Further treatment with adrenergic agents, such as epinephrine, is reasonable and effective. Epinephrine may be the first-choice treatment for bradycardia if severe hypotension is present. Severe hypotension refers to inadequate cerebral or myocardial perfusion. Prior recommendations had included use of isoproterenol, a potent β-selective adrenergic agent, to provide “chemical” pacing. However, because of its tendency to vasodilate resulting in inadequate coronary perfusion, it is no longer recommended and the next line of therapy is electrical pacing (see Fig. 33-3 ).

Transcutaneous/Transesophageal Pacing

Most modern defibrillator/cardioversion units have the capability for cardiac pacing. For transcutaneous pacing, the external pacing electrodes should be positioned anteriorly on the left sternal border and posteriorly between the scapulae. For most resuscitative efforts, the demand mode (not sensing) is selected for overdrive pacing. Initial settings should be set at the desired age appropriate rate, and the output set at zero. The output should then be increased rapidly until capture or pacing occurs. Note that the required output is usually “relative” and not in a defined milliamperage (mA) like epicardial or transvenous pacer units. Once electrical capture is seen on the monitor, mechanical association should be assessed by palpation of a pulse ( Beland et al., 1988 ).

Transesophageal pacing is another minimally invasive technique. Lead positioning is technically more challenging and can be uncomfortable depending on the level of consciousness of the infant or child. Positioning can be confirmed by ECG monitoring with an arm lead attached to the esophageal electrode, by chest radiograph, by clinical response of electrical capture, or by a combination. Because of the higher electrical output required, ventricular capture rarely occurs with a conventional epicardial pacer unit. A defibrillator/cardioversion unit should be used for transesophageal pacing as described for transcutaneous pacing to obtain electrical capture. Although both transcutaneous and transesophageal methods can be used to overdrive pace complex supraventricular tachycardia (SVT), the transesophageal electrode also has diagnostic use. Diagnostic and therapeutic consultation with a cardiologist is usually warranted.

Transvenous Pacing

When transcutaneous pacing fails to provide results, transvenous pacing is often attempted in adults. Experience with transvenous pacing techniques in children has primarily occurred in the catheterization laboratory. Limited data are available using this technique in children during CPR. One study retrospectively reviewed experience with five children with various etiologies for cardiac arrest. Four of the patients had restoration of cardiac output and stability for 10 to 60 minutes with subsequent arrest and death ( Greissman et al., 1995) . The pacing wire can be placed via a vascular introducer into the right atrium or right ventricle. The techniques for setting the output and sensitivity are then the same as for the epicardial pacing leads. Although there is limited published experience, many anesthesiologists and intensivists attempt such interventions as a bridge to more definitive or supportive therapy.

Epicardial Pacing

Many infants and children undergoing cardiac surgery have atrial and ventricular epicardial pacing wires placed during the procedure for postoperative management. During resuscitation, if both atrial and ventricular leads are available, the ventricular leads should be used to maintain cardiac output. Note that most pacemaker generators can sense as well as pace depending on the mode. It may be necessary to adjust the sensitivity and output to produce ventricular depolarization with every paced beat. Pacing in asynchronous ventricular fixed-rate or ventricle-inhibited pacing ensures the most consistent cardiac output. Again, electrical activity should be monitored for electrical association and pulses or arterial waveform monitored to evaluate the mechanical results.

For routine use in the postoperative period, the pacer is often attached with the output at 0 mA and sensitivity at 0 mA. For initial testing of the pacemaker, the rate of the pacer should be set 20 beats per minute higher than the monitored rate. The sensitivity can then be increased until the pacing unit indicates adequate sensing with every stimulus. Then the pacer output can be adjusted up until electrical capture is seen on the monitor. The pacer rate can be adjusted down to an age appropriate backup rate, and the output turned up (by several mA) to provide a margin of safety if emergent pacing is required.


Physiology of Tachycardia

The analysis of a tachyarrhythmia must discriminate between a supraventricular, junctional, or ventricular etiology. The source of the reentrant or excitable focus creates variations of wide or narrow QRS complexes. Those tachycardias with narrow complexes (<0.08 second) are often atrial or high junctional (between the atria). Sinus tachycardia is the most common tachycardia seen in the pediatric patient during anesthesia. This may be brought on by pain, hyperthermia, light anesthesia, hypercarbia, hypoxia, and hypoglycemia. Most often, these events are limited. It is not uncommon to have a child with a heart rate 20% to 30% above baseline in the postoperative period because of these various factors. However, many common pathologic tachycardias present in the perioperative period that must be recognized and distinguished from sinus tachycardia. Often, a 12-lead ECG is required along with evaluation and elimination of the possible causes. SVT, atrial fibrillation, atrial flutter, and junctional ectopic tachycardia (JET) present with narrow QRS complex. Ventricular tachycardia, often surprisingly well tolerated by pediatric patients, usually presents with widened QRS complex.

Supraventricular Tachycardia

SVT is the most common tachyarrhythmia and nonarrest arrhythmia in children and is the most common arrest arrhythmia in infancy ( Fig. 33-5 ). The cause is often a reentrant mechanism producing a rapid and narrow QRS complex (<0.08 second) on ECG. Rates are greater than 220 beats per minute in infancy and greater than 180 beats per minute in childhood. Ability of a pediatric patient to tolerate the arrhythmia depends on prior diseases, possible initiating factors, and duration of the SVT event. In infants, the presenting signs are irritability and poor feeding. Older children often complain of lightheadedness, fatigue, and sometimes chest discomfort. In anesthetized patients, the difficulty often is in differentiating the rhythm from sinus tachycardia. In this situation, it is important to rule out possible causes of sinus tachycardia such as fever, pain, hypovolemia, hypoxia, hypercarbia, and myocardial failure. If the patient becomes hemodynamically unstable, synchronized cardioversion is the treatment of choice. The initial dose is 0.5 to 1 J/kg. If SVT persists or recurs after countershock, additional synchronized cardioversion can occur with the dose increased to 2 J/kg. If vascular access is available, adenosine (see Adenosine) can be attempted but no delays for vascular access should occur.


FIGURE 33-5  Supraventricular tachycardia.



In a stable pediatric patient with SVT, vagal maneuvers can be the first line of therapy. Gag and diving reflex can be attempted. Ocular rubs are no longer recommended because of the risk for ocular trauma. Intraoperatively, some of these maneuvers are unavailable. Ice applied to the face, Valsalva maneuvers, and carotid sinus massage have potential benefit and may be attempted until adenosine dosing is prepared (see Adenosine). An electrical intervention can be overdrive pacing with an esophageal or epicardial lead. Failure of these methods may indicate a more complex arrhythmia and necessitate consultation with a pediatric cardiologist.

Atrial Fibrillation/Atrial Flutter

Atrial flutter and fibrillation are extremely rare in children (Figs. 33–6 and 33-7 [6] [7]). Atrial flutter is often a rapid and regular rhythm, and in children, it is often associated with structural heart disease. The underlying physiology is a macroreentry phenomenon creating a rapid, regular tachycardia. Atrial rates of 200 to 500 beats per minute are usually present, and “saw-toothed” waves or f waves are present on the ECG.


FIGURE 33-6  Atrial flutter.




FIGURE 33-7  Atrial fibrillation.



Atrial fibrillation is a rapid, irregular tachycardia. Atrial rates can be as high as 600 beats per minute. The rhythm occurs because of multiple microreentrant circuits in the atria creating the variable ventricular response and irregular atrial rate. Treatment options are variable and depend on the underlying structural etiology. If the rhythm is sustained and stable, consultation with a pediatric cardiologist should be sought. If unstable, synchronized cardioversion with 0.5 to 2 J/kg is treatment of choice.

Anesthesiologists are often consulted to provide sedation or anesthesia during elective cardioversion. Of importance are the duration and level of anticoagulation that should be present in these patients to help decrease the risk of atrial thrombus formation. Heparin or coumadin or both are often used and should be considered if the atrial fibrillation/flutter has been present for longer than 2 days. Presence of clot can often be excluded by transthoracic or transesophageal echocardiography but is not absolute. Chemical conversion with amiodarone and/or ibutilide may be an alternative consideration in those patients with clot or stable atrial fibrillation or flutter ( Bernard et al., 2003 ).

Junctional Ectopic Tachycardia

JET is seen in the days following cardiac surgery for congenital heart disease. It is narrow and rapid with complete AV dissociation resulting in a ventricular rate that is more rapid than the atrial rate. Many describe the arrhythmia as a narrow-complex SVT brought on by increased automaticity of the AV node or in the bundle of His. Occasional atrial beats are conducted through to the ventricles depending on the timing of the electrical pulse. Hemodynamic instability depends on the heart rate and degree of AV dissociation. If there is no instability, there is no need for treatment. However, as cardiac output becomes diminished, more aggressive treatment is warranted. Current treatment involves elimination of painful stimuli, metabolic acidosis, electrolyte abnormalities, and fever. Further interventions involve cooling the child, intravenous amiodarone (see Amiodarone) and pacing the atria at an increased rate to provide improved atrial synchrony and forward blood flow. These efforts are supportive until the JET rhythm resolves over 2 to 4 days ( Hoffman et al., 2002 ).

Ventricular Tachycardia

Ventricular tachycardia (VT) is defined as a rapid, wide-complex QRS tachycardia ( Fig. 33-8 ). The etiology of the tachycardia is below the bifurcation of the bundle of His. This can result from either a reentrant phenomena or increased automaticity. At a minimum, there are three wide QRS complexes in series with no evident p wave. When there are more wide-complex beats that persist in runs of 30 seconds or greater, sustained VT is present. If there are unstable vital signs, VT should be treated emergently with synchronized cardioversion. Countershock should start with 0.5 to 1 J/kg and increase to 2 J/kg with subsequent countershocks ( Fig. 33-9 , Pediatric Advanced Life Support [PALS] algorithm for VT without pulse). If the patient is stable, then medical intervention with intravenous amiodarone can be attempted ( Fig. 33-10 , PALS algorithm for VT with pulse). After initial stabilization, electrolytes should be measured, medications reviewed for possible toxic/adverse effects, and potential structural or conductive abnormalities explored. Myocarditis and myocardial ischemia, although rare in pediatrics, should always be in the differential diagnosis.


FIGURE 33-8  Ventricular tachycardia.




FIGURE 33-9  PALS tachycardia algorithm for infants and children with rapid rhythm and evidence of poor perfusion.




FIGURE 33-10  PALS tachycardia algorithm for infants and children with rapid rhythm and adequate perfusion.



Technique for Synchronized Cardioversion

Correct paddle size and position are important to the success of defibrillation. Three paddle sizes are available for external defibrillation: 13 cm in diameter for adults, 8 cm for older children, and 4.5 cm for infants. The largest paddle size that can be used without causing the paddles to touch should be used. The large surface area reduces the density of current flow, which reduces myocardial damage. However, if the entire paddle does not rest firmly on the chest wall, a high-density current is delivered to a small contact point on the skin. The paddles should be positioned on the chest wall with most of the myocardium included between them. If for some reason two paddles cannot be placed on the anterior chest, an alternate approach is to place one paddle anteriorly over the left precordium and the other paddle posteriorly between the scapulae.

The interface between the paddle and chest wall can be electrode cream, saline, paste, soap, or moist gauze pads. The cream produces lower impedance than does the paste. Care should be taken not to allow the substance from one paddle to touch that from the other paddle, as electrical current follows the path of least resistance. This is especially important in infants, in whom the distance between electrodes is very small.

The defibrillator/cardioversion unit should be set with “sensing” on. This mode permits coordination of the countershock with the electrical activity of the heart. Once the patient is adequately sedated (if necessary) and the paddles are positioned for optimal contact, the monitor should be indicating adequate sensing and charge set to deliver 0.5 to 1 J/kg. Once cardioversion occurs, a follow-up ECG should be obtained and the rhythm strip reviewed to identify the resulting rhythm. Repeat dosing with up to 2 J/kg is sometimes required with the same technique. Often other antiarrhythmic therapy can be initiated if the rhythm is transiently responsive to the cardioversion. For resistant tachycardias, hypoxemia, acidosis, hypoglycemia, and electrolyte abnormalities should be sought and corrected if present as they can interfere with the success of cardioversion.


Physiology of Ventricular Fibrillation

Ventricular fibrillation (VF) is a sustained burst of multiple, uncoordinated regional ventricular depolarizations and contractions, resulting in an ineffective cardiac output and cessation of myocardial blood flow. Reentrant impulses, generated within the ventricles with multiple, shifting circuits, maintain VF. Several physiologic conditions lower the threshold for fibrillation, including hypoxia, hypercapnia, myocardial ischemia, hypothermia, metabolic acidosis, and electrolyte disturbances, including those of potassium, calcium, sodium, and magnesium. VF is a relatively uncommon rhythm during cardiac arrest in children. Typically, an initial ECG finding of a bradyarrhythmia, which progresses to asystole, is found. One study showed VF to be the initial rhythm in 19% of children presenting with cardiac arrest ( Mogayzel et al., 1995 ). The etiologies of VF in that study were medical illness, toxic overdose, drowning, trauma, and congenital heart defect.

When either VT or VF with significant hypotension or absent pulses is present, electrical countershock is the treatment of choice ( Fig. 33-11 ). Drug treatment by itself cannot be relied on to terminate VF. High-voltage electrical countershock, when correctly applied, sends more than 2 amps through the heart and can terminate VF by simultaneously depolarizing and causing a sustained contraction of the myocardium. This allows spontaneous cardiac contractions to commence if the myocardium is in a well-oxygenated environment with a normal acid-base status. The amount of myocardial damage produced by the countershock relates proportionally to the amount of energy delivered ( Dahl et al., 1974 ; DiCola et al., 1976 ). In addition, the incidence of postdefibrillation arrhythmias increases as the energy dose increases ( Peleska, 1963 ). Frequent, concentrated, high-density electrical current can damage the myocardium, decrease the likelihood of successful defibrillation, and lead to postdefibrillation arrhythmias ( Weaver et al., 1982 ). These arrhythmias are thought to be associated with prolonged depolarization of the myocardial cell membrane, which increases with the intensity of the stimulus ( Jones et al., 1978 ; Anderson et al., 1980 ) and provides an ideal setting for reentrant arrhythmias ( Anderson et al., 1980 ).


FIGURE 33-11  PALS pulseless arrest algorithm.



Technique for Defibrillation

When the onset of VF is observed, defibrillation should be attempted as soon as possible. If three rapidly administered shocks of 2, 4, and 4 J/kg are unsuccessful, basic life support should be continued, epinephrine administered, and sodium bicarbonate given (if metabolic acidosis is documented or if the duration of cardiac arrest warrants its administration). If subsequent defibrillation attempts are necessary, 360 J of delivered energy to adults or 4 J/kg in children should be used. If VF is recurrent, an antiarrhythmic can be used (see later). It is not necessary to increase the energy dose for successive defibrillation attempts. On the contrary, the potential for successful defibrillation often increases after CPR and the administration of resuscitation medications improves the metabolic environment in the myocardium.

Open-chest internal defibrillation should be performed when the chest is already open during surgery or is reopened following surgery. Paddles made specifically for this purpose are applied directly to the heart. Internal paddles have a diameter of 6 cm for adults, 4 cm for children, and 2 cm for infants. The handles should be insulated. The paddles are applied with saline-soaked pads. One electrode is placed behind the left ventricle and the other over the right ventricle on the anterior surface of the heart. To perform open-chest defibrillation, a dose of 5 to 20 J of delivered energy should be used, beginning with the lowest energy level.

The optimal dose of electrical energy required to defibrillate the heart of an infant or a child is not conclusively established. The available data suggest an initial dose of 2 J/kg for external defibrillation (no data exist for internal defibrillation doses in children) ( Gutgesell et al., 1976 ). If the initial dose is unsuccessful, then a dose of 4 J/kg should be used on the second and subsequent attempts at defibrillation. The initial three defibrillation doses should be given in succession, followed by medication administration. This pattern can be maintained as either cycles of “CPR-drug-shock, CPR-drug-shock” or as “CPR-drug-shock-shock-shock.” The technique for lead/paddle placement/paddle size is similar to synchronized cardioversion (see Technique for Synchronized Cardioversion). The major difference is that no sensing of the intrinsic rhythm is performed, and countershock is administered immediately on discharge of the device.

Defibrillators are available that use a biphasic waveform. This waveform variant appears to be more effective at producing defibrillation/cardioversion at lower dosages ( Tang et al., 2002 ). The advantage of lower dose and potentially less risk of injury have made biphasic defibrillators appealing. Data are not available for lower dosages of shocks for defibrillation in pediatric patients.

Several clinical factors affect the success rate of ventricular defibrillation in humans. Success decreases with an increased duration of VF. Short fibrillation time is the most accurate predictor for successful defibrillation ( Kerber and Sarnat, 1979 ; Pionkowski et al., 1983 ). Defibrillation attempts were successful in patients shocked within 8 minutes of fibrillation, whereas attempts were unsuccessful in patients shocked at an average of 17 minutes after the onset of VF ( Kerber and Sarnat, 1979 ). A brief period of myocardial perfusion in dogs before electrical countershock improves cardiac resuscitation outcome after prolonged VF ( Niemann et al., 1992 ). Acidosis and hypoxia also decrease the success of defibrillation ( Kerber and Sarnat, 1979 ). The temperature of the patient does not alter the energy dose required for successful defibrillation ( Tacker et al., 1981 ). Patients with terminal illness are more resistant to successful defibrillation ( Gascho et al., 1979 ), as are those who fibrillate later in the course of a myocardial infarction.


Automated external defibrillators (AEDs) are becoming the standard therapy for VF in the out-of-hospital environment because early defibrillation is the key to successful resuscitation for adult patients (Weaver et al., 1988 ). Current recommendations support the use of AEDs in early rhythm identification in children as young as 1 year. A retrospective review of 18 adolescents and children aged 5 to 15 years receiving AED treatment by emergency medical services crews showed accurate rhythm detection and shock delivery ( Atkins et al., 1998 ). The use of AEDs has generally been limited to adults because of the lower incidence of VF in children. However, with the increasing recognition of VF as a cause of cardiac arrest in children, the use of AED for children may be more beneficial, resulting in the alteration of AED use from older than 8 years to older than 1 year ( Atkins et al., 1998 ). The dosage delivered by most AED devices (150 to 200 J) exceeds recommended dosages for children younger than 8 years of age or weighing less than 25 kg. The risk of delivering a shock that may cause myocardial injury needs to be considered against the possibility of diagnosis and early treatment of VF. Low-energy (150-J) impedance-adjusted shocks for adults appear to be clinically safe and effective ( Cummins et al., 1987 ). Studies have demonstrated biphasic waveform transthoracic defibrillation to be effective with lower energy levels in both adults and children ( Samson et al., 2003 ). Triphasic waveform defibrillation is under investigation ( Tang et al., 2002 ). These AED units continue to improve with the implementation of lower energy delivery and pediatric rhythm recognition software ( Niskanen, 1997 ).

Pulseless Electrical Activity

Pulseless electrical activity (PEA) is a clinical state characterized by monitored electrical activity in the absence of detectable cardiac output. This is a preterminal condition that often leads to asystole. Typically, it is characterized by a slow wide-complex rhythm in a child without pulses who has experienced a prolonged period of hypoxia, ischemia, and/or hypercarbia. If the condition occurs rapidly, it may result from a reversible cause. This subcategory of cardiac arrest was previously described as electromechanical dissociation (EMD). Recognition and treatment of one of the four H's (hypoxemia,hypothermia, and hyperkalemia) or the four T's (tension pneumothorax, pericardial tamponade, thromboembolism to the lungs, and toxins) often result in rapid relief of PEA.


PEA should be managed according to the pulseless arrest algorithm (see Fig. 33-11 ). In the setting of the operating room, many episodes of PEA in children are fortunately related to treatable causes that, if promptly recognized and treated, result in survival.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier




Clinical Effects

Atropine is a parasympatholytic agent that acts by reducing vagal tone to the heart. This results in the increased discharge rate of the sinus node, enhanced AV conduction, and activated latent ectopic pacemakers ( Gillette and Garson, 1981 ). Atropine has minimal effects on systemic vascular resistance, myocardial perfusion, and myocardial contractility ( Gilman et al., 1990 ).


Atropine is indicated for treatment of asystole, pulseless electrical activity, bradycardia associated with hypotension ( Goldberg, 1974 ), ventricular ectopy, second- and third-degree heart block, and slow idioventricular rhythms ( Scheinman et al., 1975 ). Atropine is a useful drug in clinical states associated with excessive parasympathetic tone. Acute myocardial infarction may augment parasympathetic tone and lead to arrhythmias (including asystole) that are responsive to atropine. In pediatric patients who present in cardiac arrest, bradycardia or asystole is commonly the initial rhythm and atropine is therefore a first-line drug for such patients. In infants, during the perioperative period, any type of stress (e.g., laryngoscopy) may result in severe bradycardia or even asystole secondary to enhanced parasympathetic tone. These conditions should be treated with atropine.


The recommended adult dose of atropine is 0.5 mg IV given every 5 minutes until a desired heart rate is obtained or to a maximal dose of 2 mg. Full vagal blockade occurs in adults who receive a dose of 2 mg. For asystole, 1 mg IV should be given and repeated every 5 minutes if asystole persists. The pediatric dose for atropine is 0.02 mg/kg, with a minimal dose of 0.15 mg and a maximal dose of 2 mg. A minimal dose is necessary because of the possible occurrence of paradoxical bradycardia resulting from a central stimulating effect on the medullary vagal nuclei ( Kottmeier and Gravenstein, 1968 ). Atropine may be given via any route: intravenous, endotracheal, intraosseous, intramuscular, or subcutaneous (intramuscular and subcutaneous routes may not have adequate perfusion and absorption during cardiac arrest or CPR). Onset of action occurs within 30 seconds and the peak effect occurs 1 to 2 minutes after an IV dose.

Adverse Effects

Atropine should not be used in patients in whom tachycardia is undesirable. Following myocardial infarction or ischemia with persistent bradycardia, atropine should be administered in the lowest dose possible that increases heart rate. Tachycardia, which increases myocardial oxygen consumption and can lead to VF, is common after large doses of atropine in patients with myocardial ischemia. Caution should also be used when administering atropine to patients with pulmonary or systemic outflow tract obstruction or idiopathic hypertrophic subaortic stenosis as tachycardia can decrease ventricular filling and lower cardiac output ( Table 33-10 ). Electrical pacing may be a safer means of maintaining a desired heart rate in these patients.

TABLE 33-10   -- First-line antiarrhythmic administration during cardiopulmonary resuscitation



Symptomatic bradycardia with AV node block

Vagal bradycardia during intubation attempts

After epinephrine for bradycardia with poor perfusion


0.02 mg/kg intravenous or intraosseous after ensuring oxygenation (2.5 times dose if given via TT)

Repeat every 3 to 5 min at the same dose

Maximum single dose 0.5 mg in a child and 1 mg in an adolescent

Maximum total dose 1 mg in a child and 2 mg in an adolescent



First line after vagal maneuvers fail for supraventricular tachycardia (SVT)


First dose 0.1 mg/kg rapid IV bolus, second dose increase to 0.2 mg/kg rapid IV bolus (maximum single dose, 12 mg)

Note: Must be followed with 0.5 to 1 mL/kg normal saline flush over 1 to 2 seconds to have effect



Supraventricular and ventricular tachyarrhythmia


5 mg/kg IV over 20 to 60 minutes

Data from American Heart Association in collaboration with the International Liaison Committee on Resuscitation: Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support. Circulation 102:I291, 2000.





Clinical Effects

Adenosine is a purine nucleoside that is first-line treatment for SVT for children and adults. It is an endogenous molecule involved in the various precursors of adenosine triphosphate (ATP). Adenosine acts by binding directly to adenosine receptors in the myocardium and peripheral vasculature. Binding to the receptor initiates intracellular signaling via G proteins and results in a prolonged refractory period of the AV node and slowed conduction. This breaks the reentrant circuit responsible for most SVT ( Crosson et al., 1994 ).

Clinical Indications

Treatment of narrow-complex QRS tachyarrhythmia (<0.08 second) with adenosine results in conversion to sinus rhythm in 75% of patients with few side effects. Adenosine can be used diagnostically to differentiate between VT and SVT. The half-life is <10 seconds because of rapid uptake by red blood cells and endothelial cells and metabolism by adenosine deaminase on the red cell surface. Adenosine is then completely cleared from the plasma in less than 30 seconds, giving it the rapid onset and short duration of action ( Losek et al., 1999 ).


A rapid primary bolus of 100 mcg/kg (maximum, 6 mg) is given. If SVT resumes or no electrical response is seen, another higher bolus of 200 mcg/kg (maximum, 12 mg) is given and may be repeated if no response is seen or there is a limited duration of effect. The technique of administration is critical for the success of the therapy. Central administration is preferable when available. A stopcock should be used with 10 mL of saline flush and the adenosine dosages drawn up. Thus, three saline doses and the progressively larger dosages of adenosine should be prepared and ready for administration. Each dose should be followed by a rapid injection of the 10-mL flush to drive the medicine centrally.

Adverse Effects

Reported complications are rare considering adenosine's frequency of use. Hypotension, bradycardia, and brief AV block have been reported. Less likely complications are bronchospasm, facial flushing, headache, dyspnea, chest pain, nausea, lightheadedness, complete AV block, and ventricular standstill. If complications do occur, they are often brief secondary to the short duration of action of adenosine. Note that doses may need to be increased in patients receiving methylxanthines (e.g., theophylline) because these agents are adenosine antagonists (see Table 33-10 ).


Clinical and Pharmacologic Effects

Amiodarone hydrochloride is a diiodinated benzofuran derivative containing a diethylated tertiary amine chain. It is strongly lipophilic and has extensive tissue distribution. The drug is metabolized in the liver with mainly bile elimination. There is little renal elimination. Amiodarone has a long elimination half-life ranging from 20 to 47 days ( Chow, 1996 ).

Amiodarone has a broad range of pharmacologic effects, including all four antiarrhythmic classes ( Singh et al., 1989 ). It has potassium channel-blocking action, blocks the inward sodium current, is a noncompetitive b-blocker, and has calcium channel-blocking properties. Interestingly, its major electrophysiologic effect is dependent on the route (and duration) of administration ( Bauman, 1997 ). With long-term treatment, its predominant activity is in its ability to increase the action potential duration in most cardiac tissue, a class III effect. When used acutely via the intravenous route, the drug increases AV node refractoriness and intranodal conduction interval time, class II antiadrenergic effect, or calcium channel blocker effect ( Nattel, 1993 ). Additionally, amiodarone causes both coronary and systemic vasodilation ( Coté et al., 1979 ). It does have phosphodiesterase inhibition ( Harris et al., 1993 ) and is a selective inhibitor of thyroid hormone metabolism ( Singh et al., 1989 ).

Clinical Indications

Amiodarone has been studied as both a prophylactic long-term medication for patients with high arrhythmogenic potential due to organic heart disease and for the use of acute life-threatening arrhythmias. It has been shown to be effective when lidocaine or bretylium was not for VT or VF in over 15 studies ( Bauman et al., 1987 ; Helmy et al., 1988 ; Roberts et al., 1994 ; Podrid, 1995 ; Chow, 1996 ). When intravenous amiodarone was compared with placebo in a randomized trial (ARREST trial), there was significant improvement in the number of patients surviving to the emergency department following an out-of-hospital arrest ( Gonzales et al., 1998) . Amiodarone was shown to improve survival to admission when given to adults with out-of-hospital arrest and shock-resistant VF ( Kudenchuk et al., 1999 ). A study comparing the efficacy of lidocaine with that of amiodarone for shock-resistant VF in out-of-hospital arrest demonstrated 15% versus 27% survival of adult patients to admission ( Dorian et al., 2002 ). These adult studies further support the superior performance of amiodarone for ventricular arrhythmias.

Amiodarone has been studied in children with generally favorable outcome. Perry and others (1993) showed arrhythmia resolution in 6 of 10 children (mean age, 6.8 years) for whom multiple other antiarrhythmic agents had failed. Figa and others (1994) studied 30 infants and children with life-threatening arrhythmias, including SVT and VT. Arrhythmias were eliminated in 71% of patients, and an additional 23% experienced a significant improvement in clinical status and rhythm ( Figa et al., 1994 ). Burri and others (2003) demonstrated safety and efficacy of treatment in infants. They treated 23 infants with hemodynamically unstable tachycardias. Of the infants treated, only one was unresponsive. Dosages ranged from 5 to 26 mcg/kg per min (mean, 15 mcg/kg per min). Adverse effects were seen in only four infants.


There are limited data of amiodarone pharmacokinetics in children. Intravenous administration for active arrhythmias is common practice, followed by a continuous infusion and/or transition to oral medication if ongoing treatment is indicated. An initial intravenous dose of 5 mg/kg can be followed by additional doses or a continuous infusion of 5 mcg/kg per min. Increases in the infusion can occur up to a maximum of 10 mcg/kg per min or 20 mg/kg per 24 hr ( Perry et al., 1996 ).

Adverse Effects

All of the adverse effects of amiodarone appear to be less frequent at lower dosages ( Singh, 1996 ). Cardiovascular effects appear to be the most common and include hypotension due to the acute vasodilation and negative inotropic effects of the drug. Bradyarrhythmias, congestive heart failure, cardiac arrest, and VT have all been reported. Proarrhythmias, although possible, are seen less frequently than with other class III antiarrhythmic agents with an incidence of approximately 2%. Torsades de pointes occurs in one third of these cases ( Perry et al., 1993 ). The most common noncardiovascular toxicities are pulmonary complications. Interstitial pneumonitis is the most frequent, usually associated with oral long-term treatment. A hypersensitivity pneumonitis can occur early in the course of treatment. Symptoms include cough, low-grade fever, dyspnea, weight loss, respiratory associated chest pain, and bilateral interstitial infiltrates. These symptoms are usually reversible upon cessation of the drug ( Jessurun et al., 1998 ). Hepatotoxicity can occur and is more common with oral use. Thyroid dysfunction may occur in as many as 10% of patients resulting in either hypo- or hyperthyroidism. Optic neuritis or neuropathy resulting in decreased acuity or blurred vision can progress to permanent blindness. Neurologic symptoms include ataxia, tremor, peripheral neuropathy, malaise or fatigue, sleep disturbance, dizziness, and headache. Dermatologic reactions include allergic rash, photosensitivity, and blue-gray skin discoloration ( Hilleman et al., 1998 ) (see Table 33-10 ).



Lidocaine, a class 1B antiarrhythmic, depresses the fast inward sodium channel, which results in an increased refractory period and shortening of the total action potential. The drug is metabolized primarily in the liver by the microsomal enzyme system ( Collinsworth et al., 1974) . Up to 10% of the drug is excreted unchanged in the urine. The amount excreted unchanged increases in acidic urine. There is no biliary excretion or intestinal absorption in humans.


Lidocaine causes a decrease in automaticity and in spontaneous phase 4 depolarization of pacemaker tissue. The drug increases the VF threshold while having essentially no effect on the ventricular diastolic threshold for depolarization. It decreases the duration of the action potential of Purkinje fibers and ventricular muscle while increasing the effective refractory period of these fibers. Lidocaine does not affect conduction time through the AV node or intraventricular conduction time. By decreasing automaticity, lidocaine prevents or terminates ventricular arrhythmias caused by accelerated ectopic foci. Lidocaine abolishes reentrant ventricular arrhythmias by decreasing action potential duration and conduction time of Purkinje fibers, thus reducing the nonuniformity of action. The effect on ischemic tissues in which lidocaine delivery may be limited is unknown ( Collinsworth et al., 1974) .

Hemodynamic Effects

In animal models, rapid intravenous delivery of lidocaine causes a decrease in stroke work, blood pressure, systemic vascular resistance ( Constantino et al., 1967 ), and left ventricular contractility ( Austen and Moran, 1965 ) and a slight increase in heart rate. In healthy adults, the drug does not appear to cause any change in heart rate or blood pressure ( Jewitt et al., 1968 ; Schumacher et al., 1968 ) but patients with cardiac disease have a slight decrease in ventricular function (Schumacher et al., 1969). In most patients, even in those who have sustained a recent myocardial infarction, a 1- to 2-mg/kg bolus of lidocaine does not alter cardiac output, heart rate, or blood pressure ( Jewitt et al., 1968 ). Excessive doses of lidocaine given via rapid infusion may decrease cardiac function in patients with cardiac disease, especially in those with an acute myocardial infarction. Slow intravenous administration at no greater than 50 to 100 mg/min in adults is recommended ( Collinsworth et al., 1974) . When given to a patient with a normal heart, lidocaine usually causes few, if any, hemodynamic changes.

Antiarrhythmic Effects

Lidocaine is effective in terminating ventricular premature beats (VPBs) and VT in humans during the perioperative period of general or cardiac surgery, after an acute myocardial infarction, and in patients with digitalis intoxication. Treatment of VPBs after myocardial infarction is indicated if they are of unifocal origin occurring at a rate of more than five per minute, they occur on a normal T wave, they are multifocal, or if runs of VPBs occur (VT). Lidocaine is also effective in preventing and treating ventricular arrhythmias during cardiac catheterization. The drug is second-line therapy for VF, especially when VF or tachycardia recurs. Lidocaine is not effective in the treatment of atrial or AV junctional arrhythmias.


To achieve and maintain therapeutic levels of lidocaine, a bolus dose should be given at the initiation of a constant infusion. If an infusion is begun without an initial bolus, approximately five half-lives are required to approach a plateau serum concentration (half-life of 108 minutes) ( Collinsworth et al., 1974) . When a bolus administration is used alone, ventricular arrhythmias often return within 15 to 20 minutes because of its rapid clearance ( Bartlett et al., 1984 ). Lidocaine toxicity with serum concentration greater than 7 to 8 mcg/mL occurs most commonly in patients with severe hepatic disease or severe congestive heart failure. Decreased cardiac output results in decreased hepatic blood flow, which leads to decreased lidocaine clearance. During CPR, lidocaine clearance is decreased because of the inherent decrease in cardiac output and very low hepatic blood flow. In dogs, with the use of conventional CPR to obtain a blood pressure of 20% of control values, an intravenous lidocaine bolus of 2 mg/kg resulted in very elevated blood and tissue concentrations. During CPR, distribution of the drug, which is usually complete in 20 minutes, was still not complete after 1 hour. In addition, lidocaine clearance and distribution may be altered owing to changes in protein binding and metabolism during CPR ( Chow et al., 1983 ). In humans, high peak blood and tissue concentrations of lidocaine occur during CPR, with a delay in the time to peak concentration. Comparison of the peripheral, central, and intraosseous routes of administration of lidocaine during open-chest CPR in dogs revealed no difference in time to peak serum concentration ( Chow et al., 1981 ).


In patients with normal cardiac and hepatic function, an initial intravenous bolus of lidocaine 1 mg/kg is given, followed by a constant intravenous infusion at a rate of 20 to 50 mcg/kg per min ( Table 33-11). If the arrhythmia recurs, a second intravenous bolus of the same dose can be given ( Greenblatt et al., 1976 ). In patients with severe diminution of cardiac output, a bolus no greater than 0.75 mg/kg should be administered, followed by an infusion at a rate of 10 to 20 mcg/kg per min. In patients with hepatic disease, dosages should be decreased by 50% of normal. Patients with chronic renal disease on hemodialysis have normal lidocaine pharmacokinetics. Drug interactions with lidocaine are common. Phenobarbital increases lidocaine metabolism, requiring increased doses. Isoniazid and chloramphenicol decrease lidocaine metabolism, so a decreased dosage should be used. Any drug that decreases cardiac output (e.g., β-blockers) increases the serum concentration of lidocaine, whereas drugs (e.g., isoproterenol) that increase cardiac output and hepatic blood flow cause the serum concentration to be lower than predicted.

TABLE 33-11   -- Second-line antiarrhythmic administration during cardiopulmonary resuscitation



Ventricular arrhythmias (not ventricular escape rhythm)

Suppress ventricular ectopy

Raise threshold for fibrillation


1 mg/kg intravenous or intraosseous bolus (2.5 times dose if given via TT)

20 to 50 mcg/kg per min intravenous or intraosseous infusion

Reduce infusion rate if low cardiac output or liver failure



Refractory ventricular arrhythmias

Possibly junctional ectopic tachycardia


5 to 15 mg/kg load intravenously or intraosseously over 30 to 60 minutes

20 to 80 mcg/kg per min intravenous or intraosseous infusion



Torsades de pointes



25 to 50 mg/kg intravenous or intraosseous (maximum, 2 g/dose)

Data from American Heart Association in collaboration with the International Liaison Committee on Resuscitation: Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support. Circulation 102:I291, 2000.

TT, intratracheal route





Adverse Effects

The toxic effects of lidocaine generally involve the central nervous system and include seizures, psychosis, drowsiness paresthesias, disorientation, muscle twitching, agitation, and respiratory arrest. Treatment for seizures and psychosis consists of a benzodiazepine or a barbiturate. True allergic reactions to lidocaine are extremely rare. Cardiovascular side effects (discussed earlier) are usually observed in patients whose myocardial function is already decreased. Conversion of second-degree to complete heart block has been described ( Lichstein et al., 1973 ). Further slowing of sinus bradycardia has also been observed. These effects are infrequent and occur with large-dose administration. These potential side effects do not prohibit the use of lidocaine in these patients (see Table 33-11 ).


Clinical Effects

Procainamide is a class IA antiarrhythmic agent and is metabolized to N-acetyl procainamide (NAPA), which has class III antiarrhythmic properties. Procainamide is a sodium channel-blocking agent that prolongs the refractory period of both atria and ventricles and slows the conduction velocity of the conducting system. This results in prolonged QT and PR intervals. Approximately 50% of the dose is excreted as unchanged drug and the other half is metabolized by N-acetylation in the liver ( Elson et al., 1975 ). The efficacy and toxicity of procainamide has been linked to the activity of the N-acetyl transferase enzyme. Rapid acetylators have a significantly higher NAPA/procainamide ratio than do slow acetylators ( Reidenberg et al., 1975 ). Reports in children have demonstrated potential benefit in the treatment of atrial fibrillation, flutter, and SVT, as well as the treatment of postoperative JET ( Luedtke et al., 1997 ; Walsh et al., 1997 ). Despite its long history of use in children, there are little data to compare its effectiveness with that of other antiarrhythmic agents. In PALS recommendations, procainamide serves as a second- or third-line treatment for perfusing VT.


An initial intravenous load of 5 to 15 mg/kg over 30 to 60 minutes is followed by an infusion of 20 to 80 mcg/kg per min. The infusion should be stopped if the QRS complex increases to greater than 50% of baseline, hypotension occurs, or both occur. Because the use of procainamide increases the likelihood of torsades de pointes, it should not be used in combination with other agents that prolong the QT interval, such as amiodarone. Both Walsh and others (1997) and Mandapati and others (2000) reported infusion rates of greater than 40 mcg/kg per min were necessary to cause decreases in heart rates greater than180 beats per minute. Therapeutic serum levels of procainamide are 4 to 10 mcg/mL, and procainamide plus NAPA is 10 to 30 mcg/mL. The medication can also be given intramuscularly and orally.

Adverse Effects

Reported complications of procainamide use in children are hypotension, arrhythmia (both bradyarrhythmias and tachyarrhythmias), and hepatotoxicity. Prolongation of the QT interval is a real and demonstrated complication of coadministration with other agents such as amiodarone. However, careful monitoring during administration and halting of infusion represents the treatment for most of the adverse effects. Patients on therapy should have blood levels monitored, with rising levels thought to be associated with a greater number of side effects (see Table 33-11 ).


Clinical Effects and Indications

Only two clinical scenarios are indications for emergent magnesium therapy in children: hypomagnesemia and polymorphic VT (torsades de pointes). Magnesium is an intracellular cation with less than 1% of the body's store available in the serum. The ionized fraction is physiologically active, much like calcium, and serves as a cofactor in enzymatic reactions. Low serum magnesium levels often develop in critically ill patients and patients post cardiopulmonary bypass. Whether the decline is caused by stress or creates any physiologic problems in many clinical scenarios is unknown.

Other situations have been studied, but the benefits of magnesium administration are controversial, such as myocardial ischemia, premature ventricular contractions and atrial arrhythmias, and asthma.Dorman and others (2000) and Dittrich and others (2003) , in a randomized, double-blind prospective study, demonstrated that treatment with magnesium (whether via bolus or infusion) prevented falls in magnesium levels and resulted in a lower incidence of hemodynamically unstable arrhythmias than in the placebo groups. However, another study showed other interventions (such as amiodarone treatment) to be more efficacious in dealing with these same postoperative arrhythmias ( Hoffman et al., 2002 ). The exact mechanism of magnesium on the conduction pathways of the heart is not known. Studies have demonstrated antagonism of calcium channels. Such “antagonism of calcium” has been shown to block the rise of intracellular calcium during periods of hypoxia.


Treatment with magnesium sulfate starts with 25 to 50 mg/kg per dose (maximum, 2 g/dose) given intravenously over 20 to 30 minutes. Serum levels should be monitored, with some controversy regarding the usefulness of monitoring ionized levels. Note that 1 g of magnesium sulfate is equivalent to 4 mmol, 8 mEq, or 98 mg of elemental magnesium. Magnesium sulfate can be administered via the intravenous, intraosseous, and oral route.

Adverse Effects

A rapid rate of administration can cause a fall in systemic vascular resistance by as much as 30%. This hypotension can be treated by either slowing or stopping the dose. Apnea and weakness are possible complications but are not routinely seen until toxic levels of greater than 4 mmol/L. Potentiation of neuromuscular blockade and neuromuscular weakness has been reported at lower serum levels (see Table 33-11 ).


Bretylium is no longer part of the American Heart Association protocols for treatment of ventricular arrhythmias. Its removal is based primarily on the lack of supporting evidence for its continued use in the face of more effective treatments. Bretylium is included here in the discussion of CPR medications because of its historical and controversial significance.


Bretylium, a class III antiarrhythmic (prolongs phase 3 repolarization, prolonging the refractory period), is a bromobenzyl quaternary ammonia compound not structurally related to lidocaine. Its half-life gradually increases over time, with a mean elimination half-life of 9.8 hours ( Romhilt et al., 1972 ). The drug is 80% excreted unchanged in the urine over the first 24 hours. An additional 10% of the drug is excreted in the urine during the next 72 hours ( Kuntzman et al., 1970 ).

Mechanism of Action

The mechanism of action of bretylium is still controversial, but it appears to act by adrenergic stimulation. There is an initial release of norepinephrine from adrenergic nerve endings, with subsequent inhibition of norepinephrine release ( Markis and Koch-Weser, 1971 ). There is also blockade of the reuptake of norepinephrine and epinephrine by adrenergic nerve endings, thereby potentiating the action of these agonists on adrenal receptors. Bretylium increases the action potential duration of cardiac muscle and increases the effective refractory period of Purkinje and ventricular muscle fibers. In dogs, bretylium decreases the disparity in action potential duration between normal and infarcted areas of the heart, probably the major physiologic explanation for its antiarrhythmic actions ( Chatterjee et al., 1973 ). Bretylium also increases the VF threshold in normal and infarcted myocardium. It has been known to defibrillate a heart without electrical countershock ( Bacaner, 1966 ).

Clinical Effects

Bretylium has been shown to be effective in suppressing ventricular arrhythmias when other antiarrhythmics are not, including VF resistant to electrical countershock ( Zipes et al., 1975 ). There was no difference in out-of-hospital resuscitation of VF when the number of patients achieving a stable rhythm, the time needed to achieve that rhythm, the number of defibrillation shocks required, and the numbers of patients discharged from the hospital were compared. In that study, none of the patients were defibrillated with electrocardioversion ( Bacaner, 1966 ).


The dose of bretylium used to treat VF or VT is 5 to 10 mg/kg given via rapid intravenous bolus. If the drug can be given less urgently, 500 mg should be diluted in not less than 50 mL of fluid given over 10 minutes. The slower regimen decreases the incidence of nausea in the awake patient. Close monitoring, including an ECG and blood pressure, is critical during bretylium administration. Its onset of action in suppressing VF and facilitating electrocardioversion is within minutes, although it can be delayed by up to 10 to 15 minutes. After an intramuscular injection, the drug is effective in 20 to 60 minutes. Its duration of action is 6 to 12 hours ( Haynes et al., 1981 ). After an intravenous dose of bretylium is given, an electrical countershock should be administered. If the arrhythmia then persists, the drug can be repeated every 15 to 30 minutes, up to a total dose of 30 mg/kg. If the arrhythmia is abolished, a maintenance dose, the same as the initial dose, can be given every 6 to 8 hours. In treating VT, the second dose should be repeated in 1 to 2 hours and then for maintenance every 6 to 8 hours. The drug can also be given by constant infusion at 1 to 2 mg/min in adults.

Adverse Effects

Hypertension due to norepinephrine release is the most commonly seen side effect due to bretylium ( Bernstein and Koch-Weser, 1972 ). In addition, bretylium appears to have an inotropic effect ( Markis and Kock-Weser, 1971) . After the initial hypertensive response, more than half of the patients subsequently show a mild decrease in blood pressure including orthostatic changes ( Koch-Weser, 1979 ) due to the adrenergic-blocking effects of bretylium ( Markis and Koch-Weser, 1971 ). After receiving bretylium, a patient may have an exaggerated response to dopamine, norepinephrine, and epinephrine because of the impaired uptake of those drugs. With rapid infusions of the drug, nausea and vomiting are common. Parotid swelling and pain are complications of oral use of bretylium ( Koch-Weser, 1979).


Adrenergic Agonists

Redding and Pearson (1963) first described the use of adrenergic agonists during CPR and demonstrated that early administration of epinephrine during cardiac arrest improved the success rate for resuscitation. They also demonstrated that the increase in diastolic pressure produced by administration of adrenergic agonist drugs was responsible for the success of resuscitation ( Pearson and Redding, 1965 ). They theorized that vasopressors such as epinephrine were of value because they increase systemic vascular resistance. Since that time, epinephrine has been the drug of choice during CPR, without compelling evidence for a change in that role.

Yakaitis and others (1979) investigated the relative importance of α- and β-adrenergic agonist actions during resuscitation. Also using a canine model of arrest, they found only one in four animals receiving both the pure β-adrenergic agonist isoproterenol and an α-adrenergic antagonist was successfully resuscitated. In contrast, all the dogs treated with both an α-adrenergic agonist drug and a β-adrenergic antagonist were successfully resuscitated. These data suggest that the α-adrenergic agonist action of epinephrine is responsible for successful resuscitation after cardiac arrest. Further studies have confirmed this notion. Michael and others (1984) demonstrated that the effects of epinephrine during CPR are mediated by selective vasoconstriction of peripheral vessels, excluding those supplying the brain and heart. Epinephrine infusions maintain a higher aortic pressure and result in a higher perfusion pressure to both the heart and the brain ( Michael et al., 1984 ). Even with the increase in both mean and diastolic aortic pressure, the flow to other “nonvital” organs, such as the kidneys and small intestine, becomes compromised because of intense vasoconstriction of their blood supply ( Michael et al., 1984 ;Koehler and Michael, 1985b ; Schleien et al., 1986 ).

Effects on Coronary Blood Flow

The increase and maintenance of aortic diastolic pressure associated with the administration of α-adrenergic agonists during CPR are critical for coronary blood flow and ultimately the rate of successful resuscitation. In the beating heart, the contractile state of the myocardium is increased by β-adrenergic receptor agonist action. During CPR, these drugs may stimulate spontaneous myocardial contractions and increase the intensity of VF. In the fibrillating heart, the inotropic effect of β-adrenergic agonists might be deleterious by increasing intramyocardial wall pressure ( Livesay et al., 1978 ). This increased wall pressure contributes to decreased coronary perfusion pressure and diminished myocardial blood flow. In addition, β-adrenergic stimulation increases myocardial oxygen demand by increasing cellular metabolism and oxygen consumption. The superimposition of an increased oxygen demand on the low myocardial blood flow available during CPR probably contributes to ischemia.

Other α-adrenergic agonist drugs (such as methoxamine and phenylephrine) have been used successfully during CPR. As with epinephrine, the increase in aortic diastolic pressure results in an increased coronary blood flow. However, the absence of direct β-adrenergic stimulation avoids an increase in oxygen uptake by the myocardium, resulting in a more favorable oxygen demand-to-supply ratio in the ischemic heart. These nonepinephrine, α-adrenergic agonists have contributed to successful resuscitation ( Redding and Pearson, 1963 ; Pearson and Redding, 1965 ; Yakaitis et al., 1979 ; Schleien et al., 1989 ) and maintain myocardial blood flow during CPR as effectively as epinephrine ( Yakaitis et al., 1979 ). Schleien and others (1989) found that high aortic pressures can be sustained in a canine model of CPR with phenylephrine, a pure α-adrenergic agonist. The debate continues about the merits of pure α-adrenergic agonist drugs for resuscitation because of the confusion regarding the benefit versus detriment of the β-adrenergic effects of epinephrine (Brown et al., 1987a, 1987c [46] [41]; Holmes et al., 1980 ).

Effects on Cerebral Blood Flow

During CPR, the generation of cerebral blood flow, like coronary blood flow, depends on the vasoconstriction of peripheral vessels. This vasoconstriction is enhanced by administration of α-adrenergic agonists. Epinephrine and other α-agonist drugs produce selective vasoconstriction of noncerebral peripheral vessels to areas of the head and scalp (e.g., tongue, facial muscle, and skin) without causing cerebral vasoconstriction in adult (Koehler and Michael, 1985a, 1985b [191] [192]; Beattie et al., 1991 ) and infant models of CPR ( Schleien et al., 1986 ). Infusion of either epinephrine or phenylephrine maintained cerebral blood flow and oxygen uptake at prearrest levels for 20 minutes in a canine model of CPR. This implies that blood flow was higher than that needed to maintain adequate cerebral metabolism ( Schleien et al., 1989 ). There were no differences in neurologic outcome 24 hours after resuscitation when either epinephrine or phenylephrine was administered 9 minutes after VF ( Brillman et al., 1985 ). Other investigators found epinephrine to be more beneficial in generating vital organ blood flow (Brown et al., 1986b, 1987, 1987c [40] [41]). This may have been due to the use of drug dosages that were not equipotent in generating vascular pressure and subsequent blood flow.

Cerebral oxygen uptake may be increased by a central β-adrenergic receptor effect if sufficient amounts of epinephrine cross the blood-brain barrier during or after resuscitation ( MacKenzie et al., 1976 ;Carlsson et al., 1977 ). In addition, epinephrine may have either a vasoconstriction or vasodilation effect on cerebral vessels, depending on the balance between α- and β-adrenergic actions ( Winquist et al., 1982 ). When cerebral ischemia is brief and the blood-brain barrier remains intact, epinephrine and phenylephrine have similar effects on cerebral blood flow and metabolism ( Schleien et al., 1989 ). Catecholamines may cross the blood-brain barrier when mechanical disruption occurs or when enzymatic barriers to vasopressors (e.g., monoamine oxidase inhibitors) are overwhelmed during tissue hypoxia ( Edvinsson et al., 1978 ; Lasbennes et al., 1983 ). During CPR, the blood-brain barrier may be disrupted owing to the generation of large fluctuations in cerebral venous and arterial pressures during chest compressions. In addition, the permeability of the barrier may increase because of the arterial pressure surge that occurs in a maximally dilated vascular bed after resuscitation ( Arai et al., 1981). An increase in cerebral oxygen demand when cerebral blood flow is limited could affect cerebral recovery adversely. In an infant model of CPR producing 8 minutes of cardiac arrest, disruption of the blood-brain barrier was present 4 hours after defibrillation ( Schleien et al., 1991 ). In similar protocols involving 8 minutes of cardiac arrest, endothelial vacuolization has been shown, with extravasation of protein through the blood-brain barrier ( Schleien et al., 1992a ). These theoretical effects of catecholamines on the cerebral circulation need to be further clarified and do not represent a contraindication to the administration of epinephrine during cardiac arrest.


The administration of high-dose epinephrine is no longer recommended. Studies have examined the physiologic response of animals and humans to higher doses of epinephrine. Cerebral blood flow increases further in response to administration of larger doses of epinephrine ( Brillman et al., 1985 ; Brown et al., 1986a ; Berkowitz et al., 1991 ). Although myocardial and endocardial blood flow increases, animal models suggest that there is a disproportionate rise in myocardial oxygen consumption with high-dose epinephrine ( Jackson et al., 1984 ; Maier et al., 1984 ; Brown et al., 1988a, 1988b [44] [43]; Ditchey and Lindenfeld, 1988 ). In a swine model, high-dose epinephrine failed to increase myocardial blood flow to levels achieved with a lower dose ( Berkowitz et al., 1991 ). Studies in humans have been contradictory regarding survival of patients who were given high-dose epinephrine after cardiac arrest. In earlier studies, investigators were optimistic that higher doses of epinephrine would increase aortic diastolic pressure and therefore improve the return to spontaneous circulation compared with standard epinephrine doses. Gonzalez and others (1988, 1989) [136] [135] demonstrated a dose-dependent increase in aortic blood pressure for patients who failed to respond to prolonged resuscitation efforts.

Paradis and others (1990) showed increased aortic diastolic pressure and successful resuscitation for patients for whom advanced cardiac life support (ACLS) protocols failed. They also reported on seven pediatric patients treated successfully with 0.2 mg/kg of epinephrine ( Goetting and Paradis, 1989 ). Other investigators have also reported higher aortic diastolic pressures and an improvement in ROSC (Martin et al., 1990 ; Paradis et al., 1990 ; Cipolotti et al., 1991 ). In these nonrandomized, unblinded studies, there were few survivors, although three patients survived in the pediatric study. Subsequently, three large multicenter studies were published that dampened the enthusiasm for the use of high-dose epinephrine.

Stiell and others (1992) reported on 650 adult patients who sustained cardiac arrest. These patients were randomly assigned to either a standard or a high-dose (7 mg) epinephrine protocol. High-dose epinephrine did not improve survival (18% versus 23% 1-hour survival; 3% versus 5% hospital discharge) or alter neurologic outcome. In a multicenter prospective study, Brown and others (1992) reported on 1280 adult patients who received either standard (0.02 mg/kg) or high-dose (0.2 mg/kg) epinephrine after cardiac arrest. Again, no differences were seen between groups in ROSC, short-term survival, survival to hospital discharge, or neurologic outcome between patients treated with a standard dose of epinephrine and those treated with a high dose. Callaham and others (1992) , in a study of 816 adults, reported a higher rate of ROSC in the high-dose epinephrine group. However, there were no differences in the rate of hospital discharge or survival of these patients.

There is concern that high-dose epinephrine may account for some of the adverse effects that occur after resuscitation. High doses may worsen myocardial ischemia and result in arrhythmias, hypertensive crisis, pulmonary edema, digitalis toxicity, hypoxemia, and cardiac arrest ( Brown et al., 1992 ; Schleien et al., 1992b ). Tang and others (1991) showed that epinephrine induced a decrease in PaO2 and an increase in alveolar dead space ventilation, thought to be due to a redistribution of pulmonary blood flow, compared with an α-agonist.

In children, the dosing scheme is explicit. Higher doses are preferred when epinephrine is given through the endotracheal tube because of its decreased bioavailability. All endotracheal tube doses are 0.1 mg/kg (1:1000). To treat a pulseless arrest in children, the first and all intravenous or intraosseous doses are 0.01 mg/kg (1:10,000) repeated every 3 to 5 minutes ( Table 33-12 ).

TABLE 33-12   -- Epinephrine administration during cardiopulmonary resuscitation


Decreases perfusion to nonvital organs (α-adrenergic effect)

Improves coronary perfusion (aortic diastolic pressure) (α-adrenergic effect)

Increases intensity of ventricular fibrillation (β-adrenergic effect)

Stimulates cardiac contractions (β-adrenergic effect)

Intensifies cardiac contractions (β-adrenergic effect)


Bradyarrhythmia with hemodynamic compromise

Asystole or pulseless arrest


Bradycardia 0.01 mg/kg intravenous or intraosseous or 0.1 mg/kg TT

Repeat every 3 to 5 min at the same dose


First dose 0.01 mg/kg intravenous or intraosseous or 0.1 mg/kg TT

Repeat every 3 to 5 min.

Data from American Heart Association in collaboration with the International Liaison Committee on Resuscitation: Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support. Circulation 102:I291, 2000.

TT, intratracheal route





Other Adrenergic Agents

Dopamine and dobutamine are additional agents used for vasopressor support postarrest in infants and children. Guidelines for postresuscitation support advocate their use as they cause less tachycardia, myocardial ectopy, and hypertension in the postarrest patient. Dopamine hydrochloride is often used as an infusion of 2 to 20 mcg/kg per min. At higher doses (>10 mcg/kg per min rates), α-adrenergic activity is dominant and increases in blood pressure are observed. Dobutamine hydrochloride is another adrenergic agent used for inotropic support. Dobutamine infusions are maintained between 2 and 20 mcg/kg per min. Decreases in systemic vascular resistance can be observed with its predominant β-adrenergic effects ( Table 33-13 ).

TABLE 33-13   -- Vasopressor infusions in the postarrest period





0.05 to 1.0 mcg/kg per min



2 to 20 mcg/kg per min

Inotrope, chronotrope; dilates the splanchnic vasculature at lower doses, pressor effect at higher doses


2 to 20 mcg/kg per min

Inotrope, decreased SVR


Load: 50 to 75 mcg/kg

Inotrope, improve diastolic


Infusion: 0.5 to 1 mcg/kg per min

relaxation, decreased SVR



Phosphodiesterase Inhibitors

Amrinone and milrinone are increasingly used to support inotropic myocardial function during the perioperative period in children undergoing congenital heart surgery. The benefits of these agents are multiple: increased inotropy (force of left ventricular contraction), increased dromotropy (speed and efficiency of myocardial conductive pathways), and increased lusitropy (left ventricular diastolic relaxation). There is no effect on chronotropy (rate of contraction). All of these effects are accomplished through nonadrenergic mechanisms; the impact on myocardial oxygen consumption is minimal. In addition, the risk of arrhythmias is lower. The side effects of these agents are predominantly on platelet function and decreases in systemic vascular resistance.

Because of the increased incidence of thrombocytopenia seen with amrinone, milrinone has become the drug of choice. Milrinone is usually administered with a loading dose of 50 mcg/kg over 30 minutes followed by an infusion of 0.5 to 1 mcg/kg per min. The benefits in the postoperative cardiac period at preventing low cardiac output states has been supported by a single double-blind randomized prospective study ( Hoffman et al., 2003 ). In an animal model of CPR with fibrillatory arrest, the treatment with a loading dose and maintenance infusion of milrinone improved stroke volume and sustained rhythm after arrest ( Niemann et al., 2003 ). There are no studies of this category of drug in adults or children in the postarrest period.


Clinical Effects.

Renewed interest for the vasoconstrictor vasopressin has brought it into the CPR literature. Vasopressin is a pituitary hormone that binds to specific receptors located throughout the vasculature (V1receptors, vasoconstriction) and in the renal tubules (V2 receptors, facilitate water reabsorption). L-Arginine vasopressin is the exogenously administered compound used to treat diabetes insipidus, gastric hemorrhage, and adult cardiac arrest. Both endogenous and administered vasopressin are cleared and inactivated from plasma during passage through the liver and kidneys. This results in an elimination half-life of about 10 to 20 minutes. Studies in both animals and humans show its efficacy in restoring a life-sustaining rhythm for cardiac arrest with VF (Lindner et al., 1993, 1995 [210] [213]; Prengel et al., 1996b ; Strohmenger et al., 1996 ) but not with pulseless electrical activity ( Morris et al., 1997 ).

Mechanism of Action.

For the treatment of VF, vasopressin has a theoretical advantage compared with epinephrine because it is a vasoconstrictor without adrenergic activity, so it does not increase oxygen demand at a time when oxygen delivery is limited. In addition, vasopressin should result in less ventricular ectopy and tachycardia in the postresuscitation period. These advantages may be offset by intense vasoconstriction following ROSC, potentially worsening myocardial ischemia (Prengel et al., 1996a, 1998 [283] [285]; Wenzel and Lindner, 2002 ).


No controlled study has shown improved survival in humans with vasopressin compared with epinephrine. One study showed improved ROSC with vasopressin. In an uncontrolled trial, vasopressin (40 units IV) was administered to eight adults with in-hospital VF in whom conventional resuscitation, including epinephrine, had failed. All eight patients recovered spontaneous circulation, and three were eventually discharged with intact neurologic function ( Lindner et al., 1996 ). In an out-of-hospital randomized controlled trial comparing epinephrine with vasopressin (40 units IV) after an initial unsuccessful defibrillation attempt, there was a trend (14 of 20 versus 7 of 20, P = 0.06) toward more successful resuscitation with vasopressin. However, survival to hospital discharge was not different (P= 0.16) between the groups ( Lindner et al., 1997 ).

In a prospective double-blind randomized study, the efficacy of 1 mg of epinephrine was compared against two injections of 40 U of vasopressin in adults with out-of-hospital arrest. Vasopressin and epinephrine were comparable in ROSC and hospital admission in those patients with VF and PEA. However, vasopressin was superior to epinephrine for patients with asystole (29.0% versus 20.3% in the epinephrine group, P = 0.02 for hospital admission; 4.7% versus 1.5%, P = 0.04 for hospital discharge). Further, a subgroup of 732 adults who failed to have return of circulation after vasopressin dosing were treated with epinephrine with significant improvement in rates of admission. This may suggest that refractory arrest patients are more responsive to epinephrine after treatment with vasopressin (Wenzel et al., 2004 ).

Adult dosing recommendations are the administration of vasopressin 40 U intravenously times two dosages for refractory VF. Currently, there are no recommendations for the use of vasopressin in pediatric patients. However, there has been some limited experience to suggest that it may be of similar benefit in children. Two studies have evaluated the effects of vasopressin therapy in children with cardiac compromise. A retrospective review showed that four of six children experiencing cardiac arrest had ROSC after dosage with vasopressin (0.4 U/kg) ( Mann et al., 2002 ). In two of the patients, a second dose of vasopressin of 0.4 U/kg was given. In a non-cardiac arrest study of 11 children undergoing cardiac surgery, vasopressin infusion improved blood pressure and myocardial function for 9 of 11 patients ( Cipolotti et al., 1991 ). Vasopressin was administered by continuous infusion of 0.0003 to 0.002 U/kg per min. During the first hour of treatment, systolic blood pressure rose from 65 to 87 mm Hg (P < 0.0001; n = 11), and epinephrine administration was decreased in a majority of children. Ultimately, the role of vasopressin in the treatment of children is to be determined.

Adverse Effects.

The most common side effects seen with vasopressin treatment in children receiving treatment for diabetes insipidus and gastric hemorrhage are nausea, vomiting, and abdominal pain. This may be related to vasoconstriction of the splanchnic vasculature. Rare adult reports of bowel ischemia, skin necrosis, and myocardial ischemia have been made that may be related to the vasoconstrictive effects of vasopressin. Anaphylactic and other allergic reactions have also been reported but are rare ( Wenzel et al., 2002) .


Sodium Bicarbonate

Clinical Effects.

Sodium bicarbonate use during CPR is one of the most controversial issues in the cardiac arrest literature. This stems from its potential side effects and the lack of evidence in animals and humans of any benefit from receiving bicarbonate during CPR. Administration of sodium bicarbonate results in an acid-base reaction in which bicarbonate combines with hydrogen ions to form water and carbon dioxide, resulting in an elevated blood pH:

Because bicarbonate generates carbon dioxide, adequate alveolar ventilation must be present before its administration. Sodium bicarbonate administration transiently elevates CO2 levels in the blood so that administration during cardiac arrest may worsen preexisting respiratory acidosis if ventilation is not adequate to remove the elevated CO2. This may be more of an issue for children because a major cause of cardiac arrest is respiratory failure.


Sodium bicarbonate is indicated for correction of significant metabolic acidosis. Acidosis depresses myocardial function by decreasing spontaneous cardiac activity, the electrical threshold for VF, the inotropic state of the myocardium, and the cardiac responsiveness to catecholamines and by prolonging diastolic depolarization ( Pannier and Leusen, 1968 ; Cingolani et al., 1970 ; Orlowski, 1980 ;Steinhart et al., 1983 ). Acidosis also decreases systemic vascular resistance and blunts the vasoconstrictive response of peripheral vessels to catecholamines ( Wood et al., 1963 ). In addition, pulmonary vascular resistance increases with acidosis in patients with a reactive pulmonary vascular bed ( Rudolph and Yuan, 1966 ). Therefore, correction of acidosis may be of help in resuscitating patients who have the potential for right-to-left shunting. Sodium bicarbonate is also indicated in hyperkalemic arrest because the increase in pH drives potassium intracellularly, resulting in a lowered serum potassium concentration. Hypermagnesemia, tricyclic antidepressant overdose, and overdose from sodium channel-blocking medications including cocaine, β-blockers, and diphenhydramine are other indications for bicarbonate ( Kilecki and Curry, 1997 ; Donovan et al., 1999 ; Mullins et al., 1999 ; American Heart Association, 2000) .


When the PaCO2 and pH are known, the dose of bicarbonate needed to correct the pH to 7.40 can be calculated from the formula (0.3 × weight [kg] × base deficit) = mEq bicarbonate. Because of the possible side effects of bicarbonate and the large arterial-to-venous carbon dioxide gradient that develops during CPR, giving half the dose based on a volume of distribution of 0.6 is recommended. If blood gases are not available, the initial dose is 1 mEq/kg, followed by 0.5 mEq/kg every 10 minutes of ongoing arrest ( Martinez et al., 1979 ). The importance of alveolar ventilation cannot be overemphasized, as well as the need for repeated arterial blood gas analyses.

Adverse Effects.

The multiple side effects that are seen with bicarbonate administration include metabolic alkalosis, hypernatremia ( Worthley, 1976 ), hypercapnia, and hyperosmolarity ( Mattar et al., 1974 ), all of which are associated with an increased mortality rate. Metabolic alkalosis causes a leftward shift of the oxyhemoglobin dissociation curve that impairs the release of oxygen from hemoglobin to tissues at a time of low cardiac output and low oxygen delivery ( Bishop and Weisfeldt, 1976 ). Hypernatremia and hyperosmolarity may decrease organ perfusion by increasing interstitial edema in microvascular beds.

There are various theoretical adverse effects created by bicarbonate administration. A marked hvpercapnic acidosis in both systemic venous and coronary sinus blood develops during cardiac arrest and may be worsened by administration of bicarbonate ( Grundler et al., 1986 ; Weil et al., 1986 ). Hypercapnic acidosis in the coronary sinus may cause decreased myocardial contractility ( Pannier and Leusen, 1968 ; Cingolani et al., 1970 ; Deshmukh et al., 1986 ). Falk and others (1988) measured the mean venoarterial difference of PaCO2 as 23.8 ± 15.1 mm Hg in five patients during CPR. In one patient, the difference increased from 16 mm Hg to 69 mm Hg after administration of sodium bicarbonate. In another study of 16 patients during CPR, the venoarterial gradient for carbon dioxide was 42 mm Hg ( Weil et al., 1986 ).

In the central nervous system, intracellular acidosis probably does not occur unless overcorrection of the pH occurs. After administration of two doses of bicarbonate of 5 mEq/kg to neonatal rabbits recovering from hypoxic acidosis, the arterial pH increased to 7.41 and the intracellular brain pH increased to prehypoxic levels ( Sessler et al., 1987 ). A paradoxical intracellular acidosis did not develop. In a study in rats, the intracellular brain ATP concentration did not change during 70 minutes of extreme hypercarbia, despite a decrease in the intracellular brain pH to 6.5 ( Cohen et al., 1990 ). After hypercarbia, these animals could not be distinguished from normal controls and their brains were not morphologically different from those of control animals. Eleff and others (1992) , using nuclear magnetic resonance spectroscopy to measure brain pH in dogs during CPR, showed that brain pH decreased to 6.29 after 6 minutes of VF, with total depletion of brain ATP. After 6 minutes of effective CPR, the ATP level returned to 86% of prearrest levels, and after 35 minutes of CPR, brain pH had returned to normal despite ongoing peripheral arterial acidosis ( Eleff et al., 1992 ) ( Table 33-14 ).

TABLE 33-14   -- Sodium bicarbonate administration during cardiopulmonary resuscitation



Preexisting metabolic acidosis

Long cardiopulmonary resuscitation time without blood gas availability

Pulmonary hypertensive crisis


1 mEq/kg intravenous or intraosseous empirically, or calculated from base deficit

Ensure adequate ventilation when administering bicarbonate


Metabolic alkalosis

Impairs O2 delivery by shift of oxyhemoglobin dissociation

Decreases cardiac contractility

Increases possibility for fibrillation

Decreases plasma K+ and Ca2+ by intracellular shift




Paradoxical intracellular acidosis



Several other alkalinizing agents have been used experimentally in animals and humans to avoid the real and theoretical side effects of sodium bicarbonate. Unfortunately, none have demonstrated real advantages over sodium bicarbonate. Carbicarb (International Medication Systems, Ltd.), a solution of equimolar amounts of sodium bicarbonate and sodium carbonate, works by consuming carbon dioxide and water to generate bicarbonate ion and sodium:

In animal models, Carbicarb administration resulted in a higher elevation of pH and a lesser increase in PaCO2, lactate, and serum osmolarity compared with sodium bicarbonate ( Sun et al., 1987 ; Bersin and Arieff, 1988 ; Gazmuri et al., 1990 ).

Dichloroacetate (DCA), another alkalinizing agent, works by stimulating the activity of pyruvate dehydrogenase, which facilitates the conversion of lactate to pyruvate ( Stacpoole, 1989 ). Initial studies have shown that DCA decreased lactate concentration by half and increased bicarbonate concentration and pH when administered to humans ( Stacpoole et al., 1988 ). It was also shown to improve cardiac output, possibly by enhancing myocardial metabolism of lactate and carbohydrate ( Wargovich et al., 1988 ; Stacpoole et al., 1987 ). Unfortunately, in a multicenter trial that studied patients with lactic acidosis, DCA did not improve outcome or survival compared with standard alkalinizing agents ( Stacpoole et al., 1992 ).

Tromethamine (THAM), or tris-[hydroxymethyl] aminomethane, is an organic amine that attracts and combines with hydrogen ions. It is available as a 0.3 mol/L solution adjusted to pH 8.6. A dose of 3 mL/kg should raise the bicarbonate concentration by 3 mEq/L. Side effects of this drug include hyperkalemia, hypoglycemia, acute hypocarbia, and apnea. Most important, it also acts as a peripheral vasodilator when administered during CPR, which may worsen myocardial perfusion. THAM is contraindicated for patients with renal failure.


Clinical Effects

Indications for the administration of calcium during CPR are now limited to a few specific problems. This is primarily due to the possibility that in the setting of ischemia-reperfusion injury, calcium administration may worsen postischemic hypoperfusion and hasten the development of intracellular cytotoxic events that lead to cell death. Intracellular calcium overload occurs in many pathologic conditions, including ischemia, and may be a part of the common pathway of cell death ( Katz and Reuter, 1979 ; White et al., 1983 ). Nevertheless, no study has shown that transient elevation of plasma calcium concentration worsens the outcome after cardiac arrest.

The calcium ion is essential in myocardial excitation-contraction coupling and myocardial contractility, and it enhances ventricular automaticity during asystole ( Greenblatt et al., 1976 ). Therefore, calcium should be useful in the setting of asystole or EMD. Ionized hypocalcemia leads to decreased ventricular performance, peripheral vasodilatation, and blunting of the hemodynamic response to catecholamines ( Bristow et al., 1977 ; Scheidegger et al., 1977 ; Drop and Scheidegger, 1980 ; Marquez et al., 1986 ; Urban et al., 1986 ). Severe ionized hypocalcemia (mean, 0.67 mmol/L) was present in adult patients who experienced out-of-hospital cardiac arrest ( Urban et al., 1988 ). Evidence for beneficial clinical effects of calcium during these clinical situations is lacking ( Dembo, 1981 ; Stueven et al., 1985a, 1985b [353] [354]).

Calcium channel blockers improve blood flow and function after ischemia to the heart ( Clark et al., 1979 ), kidney ( Burke et al., 1984 ), and brain ( Holthoff et al., 1990 ). Calcium channel blockers also raise the threshold of the ischemic heart to VF ( Resnekov, 1981 ). The use of calcium in these settings seems contraindicated.


The few firm indications for calcium use during CPR include cardiac arrest secondary to total or ionized hypocalcemia, hyperkalemia, hypermagnesemia, or an overdose of a calcium channel blocker. Hypocalcemia occurs with a vast array of conditions that predispose to low total body calcium stores, including the long-term use of loop diuretics. Ionized hypocalcemia may coexist with a normal total plasma calcium concentration. This occurs in the presence of severe alkalosis, which may be seen in the operating room secondary to iatrogenic hyperventilation. Ionized hypocalcemia also follows massive or rapid transfusion of citrated blood products into patients during surgery. The degree of hypocalcemia caused by citrated products depends on the rate of administration, the total dose, and the hepatic and renal function of the patient. Administration of 2 mL/kg per min of citrated whole blood causes a significant but transient decrease in the ionized calcium in anesthetized patients ( Denlinger et al., 1976 ). Because calcium administration is not a first-line treatment during CPR, hypocalcemia must be considered as a cause of cardiac arrest, particularly in the operating room, and, if present, must be treated aggressively.


The pediatric dose is 20 mg/kg or 0.2 mL/kg of the 10% calcium chloride solution. Calcium gluconate is as effective as calcium chloride in raising ionized calcium concentration during CPR ( Heining et al., 1984 ). However, calcium chloride was more effective than calcium gluconate in supporting blood pressure in the hypotensive child ( Broner et al., 1990 ). Calcium gluconate can be given as a dose of 30 to 100 mg/kg, with a maximal dose of 2 g in pediatric patients. Note that calcium preparations can be administered via the intraosseous route.

Adverse Effects

Calcium should be given slowly through a large-bore, free flowing intravenous line, preferably a central venous line. Severe tissue necrosis can occur when calcium infiltrates subcutaneous tissue. When administered too rapidly, calcium may cause severe bradycardia, heart block, or ventricular standstill ( Table 33-15 ).

TABLE 33-15   -- Calcium chloride administration during cardiopulmonary resuscitation





Calcium channel blocker overdose


20 mg/kg IV or IO




Glucose administration during CPR should be restricted to documented hypoglycemia because of the possible detrimental effects of hyperglycemia during brain ischemia. Myers and others (1979) first hypothesized that hyperglycemia worsens the neurologic outcome after cardiac arrest. Siemkowicz and Hansen (1978) confirmed this finding when they found that after 10 minutes of global brain ischemia the neurologic recovery of hyperglycemic rats was worse than in normoglycemic control animals. Hyperglycemia exaggerates ischemic neurologic injury by increasing the production of lactic acid in the brain by anaerobic metabolism. During ischemia under normoglycemic conditions, brain lactate concentration reaches a plateau. However, when hyperglycemia is present, the brain lactate concentration continues to rise for the duration of the ischemic period ( Siesjo, 1984 ). The severity of intracellular acidosis during brain ischemia is directly proportional to the preischemic plasma glucose concentration. These negative effects of hyperglycemia during brain ischemia are based on the existence of at least a small amount of blood flow to brain tissue. In one study, collaterally perfused but not end-arterial brain tissue had greater neuronal damage during hyperglycemic focal ischemia ( Prado et al., 1988 ).

Clinical studies have shown a direct correlation between the initial glucose concentration and a poor neurologic outcome ( Pulsinelli et al., 1983 ; Longstreth and Inui, 1984 ; Woo et al., 1988 ; Ashwal et al., 1990 ). Longstreth and others (1986) suggested that a higher admission plasma glucose concentration may be an endogenous response to severe stress and not the cause of more severe brain injury. Given the likelihood of additional ischemic events during the postresuscitation period, it may be wise to maintain serum glucose in the normal range. Voll and Auer (1988) showed that administration of insulin to hyperglycemic rats after global brain ischemia improved the neurologic outcome. It is not known if active treatment of hyperglycemia enhances the clinical outcome after an ischemic episode. This effect of insulin may be independent of its glucose-lowering properties, because normoglycemic treated rats had a better outcome than did placebo-treated controls ( Voll and Auer, 1991 ). Before any surgical procedure, when the possibility of brain ischemia exists, tight preoperative and intraoperative control of the serum glucose level is desirable ( Sieber and Traystman, 1992 ).

Infants, patients with hepatic disease, and debilitated patients with low endogenous glycogen stores are prone to hypoglycemia when stressed, as may occur during surgery. In these patients, bedside monitoring of the serum glucose level is critical during the perioperative period. In cardiac arrest, glucose is administered to the hypoglycemic patient to maintain normal substrate delivery to vital organs. To treat hypoglycemia, an intravenous dose of l mL/kg of 50% dextrose for adults, 2 mL/kg of 25% dextrose in children, or 3 to 5 mL/kg of 10% dextrose for infants can be administered.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


Vital organ perfusion remains a priority in the postresuscitation period. Maintaining adequate metabolic supply requires attention to blood pressure, ventilation, and glucose levels. Avoiding increased metabolic demand involves prevention of hyperthermia and seizures. In the postresuscitation period, blood pressure should be maintained adequate for vital organ perfusion and may require the administration of fluids, vasoactive medication, or pacing. Ventilation should be normocapnic, avoiding hyperventilation and hypoventilation. Blood glucose levels should also be normalized, avoiding hyperglycemia and hypoglycemia. Postresuscitation hyperthermia is common as patients are often excessively rewarmed. This overwarming can increase metabolic demands, thus worsening outcome, and should be avoided. In fact, hypothermia may be neuroprotective and was recommended for adult victims of cardiac arrest.

The International Liaison Committee on Resuscitation recommends that unconscious adult patients with spontaneous circulation after out-of-hospital cardiac arrest be cooled to 32° to 34°C for 12 to 24 hours when the initial rhythm was VF ( Nolan et al., 2003 ). Consideration should also be given to cooling for other rhythms or in-hospital cardiac arrest. These recommendations are based on the results from two prospective randomized clinical studies that compared hypothermia to normothermia in comatose survivors of cardiac arrest ( Bernard et al., 2002 ; Hypothermia after Cardiac Arrest Study Group, 2002 ). These European and Australian studies excluded children and cardiac arrests of noncardiac etiology. Animal models of asphyxial arrest are reporting protective effects of hypothermia in rats ( Xiao et al., 1998 ; Hickey et al., 2000 ; Hicks et al., 2000 ; Hachim-Idrissi et al., 2001 ) and piglets ( Agnew et al., 2003 ). There is insufficient evidence to make a recommendation on the use of therapeutic cooling for children resuscitated from cardiac arrest.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


CPR is not a definitive treatment for cardiac arrest but rather a means of maintaining respiration and circulation until the underlying pathology can be corrected. In children, the inciting event is often related to the airway, and restoring the airway by initiating CPR can be therapeutic. Because of the limited efficacy of CPR, it is important to identify the underlying pathology as early as possible and correct it so that spontaneous circulation can be restored. During anesthesia, this often means restoring the airway, reducing the anesthetic level, or correcting vascular volume deficits. If EMD is present, hypovolemia, tension pneumothorax, and pericardial tamponade must be considered if no improvement occurs with restoration of the airway. If VF is present, then electrolyte disorders, hypoglycemia, hypothermia, and digitalis or tricyclic antidepressant overdose must be considered.

Preparation for CPR during anesthetic management begins with knowledge of the most current recommendations for treating children during cardiopulmonary arrest. However, the anesthesiologist must have at his or her disposal the appropriate medications and devices necessary to treat these emergencies.

Copyright © 2008 Elsevier Inc. All rights reserved. -

Motoyama & Davis: Smith's Anesthesia for Infants and Children, 7th ed.

Copyright © 2005 Mosby, An Imprint of Elsevier


Agnew et al., 2003. Agnew DM, Koehler RC, Guerguerian AM, et al: Hypothermia for 24 hours after asphyxic cardiac arrest in piglets provides striatal neuroprotection that is sustained 10 days after rewarming.  Pediatr Res  2003; 54:253.

American Heart Association in collaboration with the International Liaison Committee on Resuscitation, 2000. American Heart Association in collaboration with the International Liaison Committee on Resuscitation : Guidelines 2000 for cardiopulmonary resuscitation and emergency cardiovascular care. Part 10: Pediatric advanced life support.  Circulation  2000; 102:I291.

Anderson et al., 1980. Anderson GJ, Reiser J, McAllister H, et al: Electro-physiological characterization of myocardial injury induced by defibrillation shocks.  Med Instrum  1980; 14:54.

Angelos et al., 1990. Angelos MG, Gaddis ML, Gaddis GM, et al: Improved survival and reduced myocardial necrosis with cardiopulmonary bypass reperfusion in a canine model of coronary occlusion and cardiac arrest.  Ann Emerg Med  1990; 19:1122.

Arai et al., 1981. Arai T, Watanabe T, Nagaro T, et al: Blood-brain barrier impairment after cardiac resuscitation.  Crit Care Med  1981; 9:444.

Ashwal et al., 1990. Ashwal S, Schneider S, Tomasi L, et al: Prognostic implications of hyperglycemia and reduced cerebral blood flow in childhood near-drowning.  Neurology  1990; 40:820.

Atkins et al., 1998. Atkins DL, Hartley LL, York DK: Accurate recognition and effective treatment of ventricular fibrillation by automated external defibrillators in adolescents.  Pediatrics  1998; 101:393.

Austen and Moran, 1965. Austen WG, Moran JM: Cardiac and peripheral vascular effects of lidocaine and procainalol.  Am J Cardiol  1965; 16:701.

Babbs et al., 1982b. Babbs CF, Fitzgerald KR, Voorhees WD, et al: High-pressure ventilation during CPR with 95% O2:5% CO2.  Crit Care Med  1982; 10:505.

Babbs et al., 1982a. Babbs CF, Tacker WA, Paris RL, et al: CPR with simultaneous compression and ventilation at high airway pressure in 4 animal models.  Crit Care Med  1982; 10:501.

Bacaner, 1966. Bacaner MB: Bretylium tosylate for suppression of induced ventricular fibrillation.  Am J Cardiol  1966; 17:528.

Barcelona and Coté, 2001. Barcelona SL, Coté CJ: Pediatric resuscitation in the operating room.  Anesthesiol Clin North Am  2001; 19:339.

Barranco et al., 1990. Barranco F, Lesmes A, Irles JA, et al: Cardiopulmonary resuscitation with simultaneous chest and abdominal compression: Comparative study in humans.  Resuscitation  1990; 20:67.

Bartlett et al., 1984. Bartlett RL, Stewart NJ, Taymond J, et al: Comparative study of three methods of resuscitation: Closed-chest, open chest manual, and direct mechanical ventricular assistance.  Ann Emerg Med  1984; 13:773.

Barton and Callaham, 1991. Barton C, Callaham M: Lack of correlation between end-tidal carbon dioxide concentrations and PaCO2 in cardiac arrest.  Crit Care Med  1991; 19:108.

Bauman et al., 1987. Bauman JL, Berk SI, Hariman RJ, et al: Amiodarone for sustained ventricular tachycardia: Efficacy, safety, and factors influencing long-term outcome.  Am Heart J  1987; 114:1436.

Bauman, 1997. Bauman JL: Class III antiarrhythmic agents: The next wave.  Pharmacotherapy  1997; 17:76S.

Beattie et al., 1991. Beattie C, Guerci AD, Hall T, et al: Mechanisms of blood flow during pneumatic vest cardiopulmonary resuscitation.  J Appl Physiol  1991; 70:454.

Beland et al., 1988. Beland MJ, Hesslein PS, Rowe RD: Ventricular tachycardia related to transcutaneous pacing.  Ann Emerg Med  1988; 17:279.

Berg et al., 2000. Berg RA, Hilwig RW, Kern KB, et al: “Bystander” chest compressions and assisted ventilation independently improve outcome from piglet asphyxial pulseless “cardiac arrest.”.  Circulation  2000; 101:1743.

Berg et al., 2001. Berg RA, Sanders AB, Kern KB, et al: Adverse hemodynamic effects of interrupting chest compressions for rescue breathing during cardiopulmonary resuscitation for ventricular fibrillation cardiac arrest.  Circulation  2001; 104:2465.

Berkowitz et al., 1989. Berkowitz ID, Chantarojanasiri T, Koehler RC, et al: Blood flow during cardiopulmonary resuscitation with simultaneous compression and ventilation in infant pigs.  Pediatr Res  1989; 26:558.

Berkowitz et al., 1991. Berkowitz ID, Gervais H, Schleien CL, et al: Epinephrine dosage effects on cerebral and myocardial blood flow in an infant swine model of cardiopulmonary resuscitation.  Anesthesiology  1991; 75:1041.

Bernard et al., 2003. Bernard EO, Schmid ER, Schmidlin D, et al: Ibutilide versus amiodarone in atrial fibrillation: A double-blinded, randomized study.  Crit Care Med  2003; 31:1031.

Bernard et al., 2002. Bernard SA, Gray TW, Buist MD, et al: Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia.  N Engl J Med  2002; 346:557.

Bernstein and Koch-Weser, 1972. Bernstein JG, Koch-Weser J: Effectiveness of bretylium tosylate against refractive ventricular arrhythmias.  Circulation  1972; 45:1024.

Berryman and Phillips, 1984. Berryman CR, Phillips GM: Interposed abdominal compression-CPR in human subjects.  Ann Emerg Med  1984; 13:226.

Bersin and Arieff, 1988. Bersin RM, Arieff AI: Improved hemodynamic function during hypoxia with Carbicarb, a new agent for the management of acidosis.  Circulation  1988; 77:227.

Beyar et al., 1989. Beyar R, Halperin HR, Tsitlik JE, et al: Circulatory assistance by intrathoracic pressure variations: optimization and mechanisms studied by a mathematical model in relation to experimental data.  Circ Res  1989; 64:703.

Beyar et al., 1984. Beyar R, Kishon Y, Sideman S, et al: Computer studies of systemic and regional blood flow mechanisms during cardiopulmonary resuscitation.  Med Biol Eng Comput  1984; 22:499.

Bhende et al., 1992. Bhende MS, Thompson AE, Cook DR, et al: Validity of a disposable end-tidal CO2 detector in verifying endotracheal tube placement in infants and children.  Ann Emerg Med  1992; 21:142.

Bhende and Thompson, 1995. Bhende MS, Thompson AE: Evaluation of an end-tidal CO2 detector during pediatric cardiopulmonary resuscitation.  Pediatrics  1995; 95:395.

Biboulet et al., 2001. Biboulet P, Aubas P, Dubourdieu J, et al: Fatal and nonfatal cardiac arrests related to anesthesia.  Can J Anaesth  2001; 48:326.

Bircher et al., 1980. Bircher N, Safar P, Stewart R: A comparison of standard, “MAST”-augmented, and open-chest CPR in dogs: A preliminary investigation.  Crit Care Med  1980; 8:147.

Bircher and Safar, 1981. Bircher N, Safar P: Comparison of standard and “new” closed-chest CPR and open-chest CPR in dogs.  Crit Care Med  1981; 9:384.

Bishop and Weisfeldt, 1976. Bishop RL, Weisfeldt ML: Sodium bicarbonate administration during cardiac arrest: Effect on arterial pH, PCO2, and osmolality.  JAMA  1976; 235:506.

Brillman et al., 1985. Brillman JA, Sanders AB, Otto CW, et al: Outcome of resuscitation from fibrillatory arrest using epinephrine and phenylephrine in dogs.  Crit Care Med  1985; 13:912.

Bristow et al., 1977. Bristow MR, Schwartz HD, Binetti G, et al: Ionized calcium and the heart. Elucidation of in vivo concentration-response relationships in the open chest dog.  Circ Res  1977; 41:565.

Broner et al., 1990. Broner CW, Stidham GL, Westenkirchner DF, et al: A prospective, randomized, double-blind comparison of calcium chloride and calcium gluconate therapies for hypocalcemia in critically ill children.  J Pediatr  1990; 115:988.

Brown et al., 1986b. Brown CG, Birinyi F, Werman HA, et al: The comparative effects of epinephrine versus phenylephrine on regional cerebral blood flow during cardiopulmonary resuscitation.  Resuscitation  1986; 14:171.

Brown et al., 1987c. Brown CG, Davis EA, Werman RL: Methoxamine versus epinephrine on regional cerebral blood flow during cardiopulmonary resuscitation.  Crit Care Med  1987; 15:682.

Brown et al., 1992. Brown CG, Martin DR, Pepe PE, et al: A comparison of standard-dose and high-dose epinephrine in cardiac arrest outside the hospital.  N Engl J Med  1992; 327:1051.

Brown et al., 1988b. Brown CG, Taylor RB, Werman HA, et al: Effect of standard doses of epinephrine on myocardial oxygen delivery and utilization during cardiopulmonary resuscitation.  Crit Care Med  1988; 16:536.

Brown et al., 1988a. Brown CG, Taylor RB, Werman HA, et al: Myocardial oxygen delivery/consumption during cardiopulmonary resuscitation: A comparison of epinephrine and phenylephrine.  Ann Emerg Med  1988; 17:302.

Brown et al., 1986a. Brown CG, Werman HA, Davis EA, et al: Comparative effects of graded doses of epinephrine on regional brain blood flow during CPR in a swine model.  Ann Emerg Med  1986; 15:1138.

Brown et al., 1987a. Brown CG, Werman HA, Davis EA, et al: The effect of high-dose phenylephrine versus epinephrine on regional cerebral flow during CPR.  Ann Emerg Med  1987; 16:743.

Brown et al., 1987b. Brown CG, Werman HA, Davis EA, et al: The effects of graded doses of epinephrine on regional myocardial blood flow during cardiopulmonary resuscitation in swine.  Circulation  1987; 75:491.

Brown et al., 2001. Brown J, Rogers J, Soar J: Cardiac arrest during surgery and ventilation in the prone position: A case report and systematic review.  Resuscitation  2001; 50:233.

Burke et al., 1984. Burke TJ, Arnold PE, Gordon JA, et al: Protective effect of intrarenal calcium membrane blockers before or after renal ischemia: Functional, morphological, and mitochondrial studies.  J Clin Invest  1984; 74:1830.

Burri et al., 2003. Burri S, Hug MI, Bauersfeld U: Efficacy and safety of intravenous amiodarone for incessant tachycardias in infants.  Eur J Pediatr  2003; 162:880.

Calkins et al., 2000. Calkins MD, Fitzgerald G, Bentley TB, et al: Intraosseous infusion devices: A comparison for potential use in special operations.  J Trauma  2000; 48:1068.

Callaham and Barton, 1990. Callaham M, Barton C: Prediction of outcome of cardiopulmonary resuscitation from end-tidal carbon dioxide concentration.  Crit Care Med  1990; 19:358.

Callaham et al., 1992. Callaham M, Madsen CD, Barton CW, et al: A randomized clinical trial of high-dose epinephrine and norepinephrine versus standard-dose epinephrine in prehospital cardiac arrest.  JAMA  1992; 268:2667.

Carcillo et al., 1991. Carcillo JA, Davis AL, Zaritsky A: Role of early fluid resuscitation in pediatric septic shock.  JAMA  1991; 266:1242.

Carlsson et al., 1977. Carlsson C, Hagerdal M, Kaasid AE, et al: A catecholamine-mediated increase in cerebral oxygen uptake during immobilization stress in rats.  Brain Res  1977; 119:223.

Carpenter and Stenmark, 1997. Carpenter TC, Stenmark KR: High-dose epinephrine is not superior to standard-dose epinephrine in pediatric in-hospital cardiopulmonary arrest.  Pediatrics  1997; 99:403.

Cavallaro and Melker, 1983. Cavallaro DL, Melker RJ: Comparison of two techniques for detecting cardiac activity in infants.  Crit Care Med  1983; 11:189.

Chandra et al., 1980. Chandra N, Rudikoff M, Weisfeldt ML: Simultaneous chest compression and ventilation at high airway pressure during cardiopulmonary resuscitation.  Lancet  1980; 1:175.

Chandra et al., 1981b. Chandra N, Snyder LD, Weisfeldt ML: Abdominal binding during cardiopulmonary resuscitation in man.  JAMA  1981; 246:351.

Chandra et al., 1990. Chandra N, Tsitlik J, Halperin HR, et al: Observations of hemodynamics during human cardiopulmonary resuscitation.  Crit Care Med  1990; 18:929.

Chandra et al., 1981a. Chandra N, Weisfeldt ML, Tsitlik J, et al: Augmentation of carotid flow during cardiopulmonary resuscitation by ventilation at high airway pressure simultaneous with chest compression.  Am J Cardiol  1981; 48:1053.

Chandra et al., 1991. Chandra NC, Beyar R, Halperin HR, et al: Vital organ perfusion during assisted circulation by manipulation of intrathoracic pressure.  Circulation  1991; 84:279.

Chatterjee et al., 1973. Chatterjee K, Mandel WJ, Vyden JK, et al: Cardiovascular effects of bretylium tosylate in acute myocardial infarction.  JAMA  1973; 223:757.

Chen et al., 2003. Chen YS, Chao A, Yu HY, et al: Analysis and results of prolonged resuscitation in cardiac arrest patients rescued by extracorporeal membrane oxygenation.  J Am Coll Cardiol  2003; 41:197.

Cherian et al., 1997. Cherian L, Goodman JC, Robertson CS: Hyperglycemia increases brain injury caused by secondary ischemia after cortical impact injury in rats.  Crit Care Med  1997; 25:1378.

Chow, 1996. Chow MS: Intravenous amiodarone: Pharmacology, pharmacokinetics, and clinical use.  Ann Pharmacother  1996; 30:637.

Chow et al., 1983. Chow MSS, Ronfeld RA, Hamilton RA, et al: Effect of external cardiopulmonary resuscitation on lidocaine pharmacokinetics in dogs.  J Pharmacol Exp Ther  1983; 224:531.

Chow et al., 1981. Chow MSS, Ronfeld RA, Ruffett D, et al: Lidocaine pharmacokinetics during cardiac arrest and external cardiopulmonary resuscitation.  Am Heart J  1981; 102:799.

Cingolani et al., 1970. Cingolani HE, Mattiazi AR, Blesa ES: Contractility in isolated mammalian heart muscle after acid-base changes.  Circ Res  1970; 26:269.

Cipolotti et al., 1991. Cipolotti G, Paccagnella A, Simini G: Successful cardiopulmonary resuscitation using high doses of epinephrine.  Int J Cardiol  1991; 33:430.

Clark et al., 1979. Clark RE, Kristlieb IY, Henry PD: Nifedipine: A myocardial protective agent.  Am J Cardiol  1979; 44:825.

Clements et al., 1986. Clements FM, De Bruijn NP, Kisslo JA: Transesophageal echocardiographic observations in a patient undergoing closed-chest massage.  Anesthesiology  1986; 64:826.

Cleveland, 1971. Cleveland JC: Complete recovery after cardiac arrest for three hours.  N Engl J Med  1971; 284:334.

Cochran et al., 1999. Cochran JB, Tecklenburg FW, Lau YR, et al: Emergency cardiopulmonary bypass for cardiac arrest refractory to pediatric advanced life support.  Pediatr Emerg Care  1999; 15:30.

Cohen et al., 1982. Cohen JM, Chandra M, Alderson PO, et al: Timing of pulmonary and systemic blood flow during intermittent high intrathoracic pressure cardiopulmonary resuscitation in the dog.  Am J Cardiol  1982; 49:1883.

Cohen et al., 1993. Cohen TJ, Goldner BG, Maccaro PC, et al: A comparison of active compression-decompression cardiopulmonary resuscitation with standard cardiopulmonary resuscitation for cardiac arrests occurring in the hospital.  N Engl J Med  1993; 329:1918.

Cohen et al., 1992. Cohen TJ, Tucker KJ, Lurie KG, et al: Active compression-decompression. A new method of cardiopulmonary resuscitation. Cardiopulmonary Resuscitation Working Group.  JAMA  1992; 267:2916.

Cohen et al., 1990. Cohen Y, Chang LH, Litt L, et al: Stability of brain intracellular pH during prolonged hypercapnia in rats.  J Cereb Blood Flow Metab  1990; 10:277.

Collinsworth et al., 1974. Collingsworth KA, Kalman SM, Harrison DC: The clinical pharmacology of lidocaine as an antiarrhythmic drug.  Circulation  1974; 50:1217.

Constantino et al., 1967. Constantino RT, Crockett SE, Vasko JS: Cardiovascular effects and dose response relationships of lidocaine.  Circulation  1967; 36:89.

Coté et al., 1979. Coté P, Bourassa MG, Delaye J, et al: Effects of amiodarone on cardiac and coronary hemodynamics and on myocardial metabolism in patients with coronary artery disease.  Circulation  1979; 59:1165.

Criley et al., 1976. Criley JM, Blaufuss AH, Kissel GL: Cough-induced cardiac compression: self-administered form of cardiopulmonary resuscitation.  JAMA  1976; 236:1246.

Criley et al., 1986. Criley JM, Niemann JT, Rosborough JP, et al: Modification of cardiopulmonary resuscitation based on the cough.  Circulation  1986; 74(suppl IV):IV-42.

Crosson et al., 1994. Crosson JE, Etheridge SP, Milstein S, et al: Therapeutic and diagnostic utility of adenosine during tachycardia evaluation in children.  Am J Cardiol  1994; 74:155.

Cummins et al., 1987. Cummins RO, Eisenberg MS, Litwin PE, et al: Automatic external defibrillators used by emergency medical technicians. A controlled clinical trial.  JAMA  1987; 257:1605.

Dahl et al., 1974. Dahl CF, Ewy GA, Warner ED, et al: Myocardial necrosis from direct current countershock: effect of paddle electrode size and time interval between discharges.  Circulation  1974; 50:956.

Dalton et al., 1993. Dalton HJ, Siewers RD, Fuhrman BP, et al: Extracorporeal membrane oxygenation for cardiac rescue in children with severe myocardial dysfunction.  Crit Care Med  1993; 21:1020.

David, 1988. David R: Closed chest cardiac massage in the newborn infant.  Pediatrics  1988; 81:552.

Davies et al., 1987. Davies CR, Carrigan T, Wright JA, et al: Neurologic outcome following pediatric resuscitation.  J Neurosci Nurs  1987; 19:205.

Dean et al., 1987. Dean JM, Koehler RC, Schleien CL, et al: Age-related changes in chest geometry during cardiopulmonary resuscitation.  J Appl Physiol  1987; 62:2212.

Dean et al., 1990. Dean JM, Koehler RC, Schleien CL, et al: Age-related effects of compression rate and duration in cardiopulmonary resuscitation.  J Appl Physiol  1990; 68:554.

Dean et al., 1991. Dean JM, Koehler RC, Schleien CL, et al: Improved blood flow during prolonged cardiopulmonary resuscitation with 30% duty cycle in infant pigs.  Circulation  1991; 84:896.

Del Guercio et al., 1965. Del Guercio LR, Feins NR, Cohn JD, et al: Comparison of blood flow during external and internal cardiac massage in man.  Circulation  1965; 32(suppl I):I-171.

del Nido et al., 1992. del Nido PJ, Dalton HJ, Thompson AE, et al: Extracorporeal membrane oxygenator rescue in children during cardiac arrest after cardiac surgery.  Circulation  1992; 86(suppl II):II-300-II-304.

Dembo, 1981. Dembo DH: Calcium in advanced life support.  Crit Care Med  1981; 9:358.

Denlinger et al., 1976. Denlinger JK, Nahrwold MLL, Gibbs PS, et al: Hypocalcemia during rapid blood transfusion in anaesthetized man.  Br J Anaesth  1976; 48:995.

Dequin et al., 1996. Dequin PF, Hazouard E, Legras A, et al: Cardiopulmonary resuscitation in the prone position: Kouwenhoven revisited.  Intensive Care Med  1996; 22:1272.

Deshmukh et al., 1986. Deshmukh HG, Gudipati CV, Weil MH, et al: Myocardial respiratory acidosis during CPR.  Crit Care Med  1986; 14:433.

DiCola et al., 1976. DiCola VC, Freedman GS, Downing SE, et al: Myocardial uptake of technetium-99m stannous pyrophosphate following direct current transthoracic countershock.  Circulation  1976; 54:980.

Ditchey and Lindenfeld, 1988. Ditchey RV, Lindenfeld J: Failure of epinephrine to improve the balance between myocardial oxygen supply and demand during closed chest resuscitation in dogs.  Circulation  1988; 78:382.

Ditchey et al., 1982. Ditchey RV, Winkler JV, Rhodes CA: Relative lack of blood flow during closed-chest resuscitation in dogs.  Circulation  1982; 66:297.

Dittrich et al., 2003. Dittrich S, Gemanakis J, Dahnert I, et al: Randomized trial on the influence of continuous magnesium infusion on arrhythmias following cardiopulmonary bypass surgery for congenital heart disease.  Intensive Care Med  2003; 29:1141.

Donovan et al., 1999. Donovan KD, Gerace RV, Dreyer RF: Acebutolol-induced ventricular tachycardia reversed with sodium bicarbonate.  J Toxicol Clin Toxicol  1999; 37:481-484.

Dorfsman et al., 2000. Dorfsman ML, Menegazzi JJ, Wadas RJ, et al: Two-thumb versus two-finger chest compression in an infant model of prolonged cardiopulmonary resuscitation.  Acad Emerg Med  2000; 7:1077-1082.

Dorian et al., 2002. Dorian P, Cass D, Schwartz B, et al: Amiodarone as compared with lidocaine for shock-resistant ventricular fibrillation.  N Engl J Med  2002; 346:884.

Dorman et al., 2000. Dorman BH, Sade RM, Burnette JS, et al: Magnesium supplementation in the prevention of arrhythmias in pediatric patients undergoing surgery for congenital heart defects.  Am Heart J  2000; 139:522.

Drop and Scheidegger, 1980. Drop LJ, Scheidegger D: Plasma ionized calcium concentration: Important determinant of the hemodynamic response to calcium infusion.  J Thorac Cardiovasc Surg  1980; 79:425.

Duncan et al., 1998. Duncan BW, Ibrahim AE, Hraska V, et al: Use of rapid-deployment extracorporeal membrane oxygenation for the resuscitation of pediatric patients with heart disease after cardiac arrest.  J Thorac Cardiovasc Surg  1998; 116:305.

Eberle et al., 1990. Eberle B, Schleien CL, Shaffner DH, et al: Effects of three modes of abdominal compression on vital organ blood flow in a piglet CPR model.  Anesthesiology  1990; 73:A300.

Edvinsson et al., 1978. Edvinsson L, Hardebo JE, MacKenzie ET, et al: Effect of endogenous noradrenaline on local cerebral blood flow after osmotic opening of the blood-brain barrier in the rat.  J Physiol (Lond)  1978; 274:149.

Einagle et al., 1988. Einagle V, Bertrand F, Wise RA, et al: Interposed abdominal compressions and carotid blood flow during cardiopulmonary resuscitation.  Chest  1988; 93:1206.

Elam et al., 1954. Elam JO, Brown ES, Elder JD: Artificial respiration by mouth-to-mask method.  N Engl J Med  1954; 250:749.

Eleff et al., 1992. Eleff SM, Schleien CL, Koehler RC, et al: Brain bioenergetics during cardiopulmonary resuscitation in dogs.  Anesthesiology  1992; 76:77.

Elson et al., 1975. Elson J, Strong JM, Lee WK, et al: Antiarrhythmic potency of N-acetylprocainamide.  Clin Pharmacol Ther  1975; 17:134.

Falk et al., 1988. Falk JL, Rackow EC, Weil MH: End-tidal carbon dioxide concentration during cardiopulmonary resuscitation.  N Engl J Med  1988; 318:607.

Feneley et al., 1987. Feneley MP, Maier GW, Gaynor JW, et al: Sequence of mitral valve motion and transmitral blood flow during manual cardiopulmonary resuscitation in dogs.  Circulation  1987; 76:363.

Feneley et al., 1988. Feneley MP, Maier GW, Kern KB, et al: Influence of compression rate on initial success of resuscitation and 24 hour survival after prolonged manual cardiopulmonary resuscitation in dogs.  Circulation  1988; 77:240.

Figa et al., 1994. Figa FH, Gow RM, Hamilton RM, et al: Clinical efficacy and safety of intravenous Amiodarone in infants and children.  Am J Cardiol  1994; 74:573.

Fisher et al., 1982. Fisher J, Vaghaiwall F, Tsitlik J, et al: Determinants and clinical significance of jugular venous valve competence.  Circulation  1982; 65:188.

Fleisher et al., 1987. Fleisher G, Delgado-Paredes C, Heyman S: Slow versus rapid closed-chest cardiac compression during cardiopulmonary resuscitation in puppies.  Crit Care Med  1987; 15:939.

Garnett et al., 1987. Garnett AR, Ornato JP, Gonzales ER, et al: End-tidal carbon dioxide monitoring during cardiopulmonary resuscitation.  JAMA  1987; 257:512.

Gascho et al., 1979. Gascho JA, Crampton RS, Cherwek ML, et al: Determinants of ventricular defibrillation in adults.  Circulation  1979; 60:231.

Gazmuri et al., 1990. Gazmuri RJ, Planta M, Weil MH, et al: Cardiac effects of carbon dioxide-consuming and carbon dioxide-generating buffers during cardiopulmonary resuscitation.  JAMA  1990; 15:482.

Gazmuri et al., 1991. Gazmuri RJ, Weil MH, Bisera J, et al: End-tidal carbon dioxide tension as a monitor of native blood flow during resuscitation by extracorporeal circulation.  J Thorac Cardiovasc Surg  1991; 101:984.

Gillette and Garson, 1981. Gillette PC, Garson A: Pediatric cardiac dysrhythmias,  New York, Grune & Stratton, 1981.

Gillis et al., 1986. Gillis J, Dickson D, Rieder M, et al: Results of inpatient pediatric resuscitation.  Crit Care Med  1986; 14:469.

Gilman et al., 1990. Gilman AG, Rall TW, Nies AS, et al: Goodman and Gilman's the pharmacological basis of therapeutics,  8th ed.. Elmsford, NY, Pergamon Press, 1990.

Girardi and Barie, 1995. Girardi LN, Barie PS: Improved survival after intraoperative cardiac arrest in noncardiac surgical patients.  Arch Surg  1995; 130:15.

Glaeser et al., 1993. Glaeser PW, Hellmich TR, Szewczuga D, et al: Five-year experience in prehospital intraosseous infusions in children and adults.  Ann Emerg Med  1993; 22:1119.

Goetting and Paradis, 1991a. Goetting MG, Paradis MA: Right atrial-jugular venous pressure gradients during CPR in children.  Ann Emerg Med  1991; 20:27.

Goetting et al., 1991b. Goetting MG, Paradis NA, Appleton TJ, et al: Aortic-carotid artery pressure differences and cephalic perfusion pressure during cardiopulmonary resuscitation in humans.  Crit Care Med  1991; 19:1012.

Goetting and Paradis, 1989. Goetting MG, Paradis NA: High dose epinephrine in refractor pediatric cardiac arrest.  Crit Care Med  1989; 17:1258.

Goldberg, 1974. Goldberg AH: Cardiopulmonary arrest.  N Engl J Med  1974; 290:381.

Gonzalez et al., 1998. Gonzalez ER, Kannewurf BS, Ornato JP: Intravenous amiodarone for ventricular arrhythmias: Overview and clinical use.  Resuscitation  1998; 39:33.

Gonzalez et al., 1989. Gonzalez ER, Ornato JP, Garnett AR, et al: Dose-dependent vasopressor response to epinephrine during CPR in human beings.  Ann Emerg Med  1989; 18:920.

Gonzalez et al., 1988. Gonzalez ER, Ornato JP, Levine RL: Vasopressor effect of epinephrine with and without dopamine during cardiopulmonary resuscitation.  Drug Intell Clin Pharm  1988; 22:868.

Grayling et al., 2002. Grayling M, Wilson IH, Thomas B: The use of the laryngeal mask airway and Combitube in cardiopulmonary resuscitation; a national survey.  Resuscitation  2002; 52:183.

Greenblatt et al., 1976. Greenblatt DJ, Gross PL, Bolognini V: Pharmacotherapy of cardiopulmonary arrest.  Am J Hosp Pharm  1976; 33:379.

Greissman et al., 1995. Greissman A, Silver P, Nimkoff L, et al: Transvenous right ventricular pacing during cardiopulmonary resuscitation of pediatric patients with acute cardiomyopathy.  Pediatr Emerg Care  1995; 11:17.

Grmec et al., 2003. Grmec S, Lah K, Tusek-Bunc K: Difference in end-tidal CO2 between asphyxia cardiac arrest and ventricular fibrillation/pulseless ventricular tachycardia cardiac arrest in the prehospital setting.  Crit Care  2003; 7:R139.

Grundler et al., 1986. Grundler W, Weil MH, Rackow EC: Arteriovenous carbon dioxide and pH gradients during cardiac arrest.  Circulation  1986; 77:234.

Gudipati et al., 1986. Gudipati CV, Weil MH, Deshmukh HG, et al: Right atrial-jugular venous pressure gradients during experimental CPR.  Chest  1986; 89:443s.

Guerci et al., 1985. Guerci AD, Shi A, Levin H, et al: Transmission of intrathoracic pressure to the intracranial space during cardiopulmonary resuscitation in dogs.  Circ Res  1985; 56:20.

Gutgesell et al., 1976. Gutgesell HP, Tacker WA, Geddes LA, et al: Energy dose for ventricular defibrillation of children.  Pediatrics  1976; 58:898.

Guy et al., 1993. Guy J, Haley K, Zuspan SJ: Use of intraosseous infusion in the pediatric trauma patient.  J Pediatr Surg  1993; 28:158.

Hachimi-Idrissi et al., 2001. Hachimi-Idrissi S, Corne L, Huyghens L: The effect of mild hypothermia and induced hypertension on long term survival rate and neurological outcome after asphyxial cardiac arrest in rats.  Resuscitation  2001; 49:73.

Hackl et al., 1990. Hackl W, Simon P, Mauritz W, et al: Echocardiographic assessment of mitral valve function during mechanical cardiopulmonary resuscitation in pigs.  Anesth Analg  1990; 70:350.

Halperin et al., 1986b. Halperin HR, Guerci AD, Chandra N, et al: Vest inflation without simultaneous ventilation during cardiac arrest in dogs: Improved survival from prolonged cardiopulmonary resuscitation.  Circulation  1986; 74:1407.

Halperin et al., 1987. Halperin HR, Tsitlik JE, Beyar R, et al: Intrathoracic pressure fluctuations move blood during CPR: Comparison of hemodynamic data with predictions from a mathematical model.  Ann Biomed Eng  1987; 15:385.

Halperin et al., 1993. Halperin HR, Tsitlik JE, Gelfand M, et al: A preliminary study of cardiopulmonary resuscitation by circumferential compression of the chest with use of a pneumatic vest.  N Engl J Med  1993; 329:762.

Halperin et al., 1986a. Halperin HR, Tsitlik JE, Guerci AD, et al: Determinants of blood flow to vital organs during cardiopulmonary resuscitation in dogs.  Circulation  1986; 73:539.

Harada et al., 1991. Harada Y, Fuseno H, Ohtomo T, et al: Self-administered hyperventilation cardiopulmonary resuscitation for 100 s of cardiac arrest during Holter monitoring.  Chest  1991; 99:1310.

Harris et al., 1993. Harris L, Kimura Y, Shaikh NA: Phospholipase inhibition and the electrophysiology of acute ischemia: Studies with amiodarone.  J Mol Cell Cardiol  1993; 25:1075.

Harris et al., 1967. Harris LC, Kirimli B, Safar P: Ventilation-cardiac compression rates and ratios in cardiopulmonary resuscitation.  Anesthesiology  1967; 28:806.

Hartz et al., 1990. Hartz R, LoCicero J, Sanders JH, et al: Clinical experience with portable cardiopulmonary bypass in cardiac arrest patients.  Ann Thorac Surg  1990; 50:437.

Hayes et al., 2003. Hayes MM, Berg RA, Otto CW: Monitoring during cardiac arrest: Are we there yet?.  Curr Opin Crit Care  2003; 9:211.

Haynes et al., 1981. Haynes RE, Chinn TL, Copass MK, et al: Comparison of bretylium tosylate and lidocaine in management of out-of-hospital ventricular fibrillation: A randomized clinical trial.  Am J Cardiol  1981; 48:353.

Heining et al., 1984. Heining MPD, Band DM, Linton RAF: Choice of calcium salt: A comparison of the effects of calcium chloride and gluconate on plasma ionized calcium.  Anaesthesia  1984; 39:1079.

Helmy et al., 1988. Helmy I, Herre JM, Gee G, et al: Use of intravenous amiodarone for emergency treatment of life-threatening ventricular arrhythmias.  J Am Coll Cardiol  1988; 12:1015.

Hickey et al., 2000. Hickey RW, Ferimer H, Alexander HL, et al: Delayed, spontaneous hypothermia reduces neuronal damage after asphyxial cardiac arrest in rats.  Crit Care Med  2000; 28:3511.

Hicks et al., 2000. Hicks SD, DeFranco DB, Callaway CW: Hypothermia during reperfusion after asphyxial cardiac arrest improves functional recovery and selectively alters stress-induced protein expression.  J Cereb Blood Flow Metab  2000; 20:520.

Higano et al., 1990. Higano ST, Oh JK, Ewy GA, et al: The mechanism of blood flow during closed chest cardiac massage in humans: Transesophageal echocardiographic observations.  Mayo Clin Proc  1990; 65:2.

Hilleman et al., 1998. Hilleman D, Miller MA, Parker R, et al: Optimal management of amiodarone therapy: Efficacy and side effects.  Pharmacotherapy  1998; 18:138S.

Hoffman et al., 2002. Hoffman TM, Bush DM, Wernovsky G, et al: Postoperative junctional ectopic tachycardia in children: Incidence, risk factors, and treatment.  Ann Thorac Surg  2002; 74:1607.

Hoffman et al., 2003. Hoffman TM, Wernovsky G, Atz AM, et al: Efficacy and safety of milrinone in preventing low cardiac output syndrome in infants and children after corrective surgery for congenital heart disease.  Circulation  2003; 107:996.

Holmes et al., 1980. Holmes HR, Babbs CF, Voorhees WD, et al: Influence of adrenergic drugs upon vital organ perfusion during CPR.  Crit Care Med  1980; 8:137.

Holthoff et al., 1990. Holthoff V, Beil C, Hartmann-Klosterklotter U, et al: Effect of nimodipine on glucose metabolism in stroke patients.  Stroke  1990; 21(suppl IV):IV-95.

Howard et al., 1987. Howard M, Carruba C, Foss F, et al: Interposed abdominal compression-CPR: Its effects on parameters of coronary perfusion in human subjects.  Ann Emerg Med  1987; 16:253.

Howard et al., 1984. Howard M, Carruba C, Guiness M, et al: Interposed abdominal compression CPR: Its effects on coronary perfusion pressure in human subjects.  Ann Emerg Med  1984; 13:989.

Hypothermia after Cardiac Arrest Study Group, 2002. Hypothermia after Cardiac Arrest Study Group : Mild therapeutic hypothermia to improve the neurologic outcome after cardiac arrest.  N Engl J Med  2002; 346:549.

Jackson et al., 1981. Jackson DL, Dole WP, McGloin J, et al: Total cerebral ischemia: Application of a new model system to studies of cerebral microcirculation.  Stroke  1981; 12:66.

Jackson et al., 1984. Jackson RE, Joyce K, Danosi SF, et al: Blood flow in the cerebral cortex during cardiac resuscitation in dogs.  Ann Emerg Med  1984; 13:657.

Jasani et al., 1994. Jasani MS, Nadkarni VM, Finkelstein MS, et al: Effects of different techniques of endotracheal epinephrine administration in pediatric porcine hypoxic-hypercarbic cardiopulmonary arrest.  Crit Care Med  1994; 22:1174.

Jessurun et al., 1998. Jessurun GA, Boersma WG, Crijns HJ: Amiodarone-induced pulmonary toxicity. Predisposing factors, clinical symptoms and treatment.  Drug Saf  1998; 18:339.

Jewitt et al., 1968. Jewitt DE, Kishow Y, Thomas M: Lidocaine in the management of arrhythmias after myocardial infarction.  Circulation  1968; 37:965.

Jones et al., 1978. Jones JL, Lepeschkin E, Jones RE, et al: Response of cultured myocardial cells to countershock-type electrical field stimulation.  Am J Physiol  1978; 235:H214.

Jude et al., 1961. Jude JR, Kouwenhoven WB, Knickerbocker GG: Cardiac arrest: Report of application of external cardiac massage on 118 patients.  JAMA  1961; 178:1063.

Katz and Reuter, 1979. Katz AM, Reuter M: Cellular calcium and cardiac cell death.  Am J Cardiol  1979; 44:188.

Kawashima et al., 2003. Kawashima Y, Takahashi S, Suzuki M, et al: Anesthesia-related mortality and morbidity over a 5-year period in 2,363,038 patients in Japan.  Acta Anaesthesiol Scand  2003; 47:809.

Keenan and Boyan, 1985. Keenan RL, Boyan CP: Cardiac arrest due to anesthesia: A study of incidence and causes.  JAMA  1985; 253:2373.

Keenan et al., 1991. Keenan RL, Shapiro JH, Dawson KD: Frequency of anesthetic cardiac arrests in infants: Effect of pediatric anesthesiologists.  J Clin Anesth  1991; 3:433.

Kerber and Sarnat, 1979. Kerber RE, Sarnat W: Factors influencing the success of ventricular defibrillation in man.  Circulation  1979; 60:226.

Kern et al., 1986. Kern KB, Carter AB, Showen RL, et al: Twenty-four hour survival in a canine model of cardiac arrest comparing three methods of manual cardiopulmonary resuscitation.  J Am Coll Cardiol  1986; 7:859.

Kern et al., 1998. Kern KB, Hilwig RW, Berg RA, et al: Efficacy of chest compression-only BLS CPR in the presence of an occluded airway.  Resuscitation  1998; 39:179.

Kern et al., 1992. Kern KB, Sanders AB, Raife J, et al: A study of chest compression rates during cardiopulmonary resuscitation in humans. The importance of rate-directed chest compressions.  Arch Intern Med  1992; 152:145.

Kernstine et al., 1982. Kernstine KH, Tyson GS, Maier GW, et al: Determinants of direct cardiac compression during external cardiac massage in intact dogs.  Crit Care Med  1982; 10:231.

Kilecki and Curry, 1997. Kilecki PF, Curry SC: Poisoning by sodium channel blocking agents.  Crit Care Clin  1997; 13:829-848.

Kleinman et al., 1999. Kleinman ME, Oh W, Stonestreet BS: Comparison of intravenous and endotracheal epinephrine during cardiopulmonary resuscitation in newborn piglets.  Crit Care Med  1999; 27:2748.

Koch-Weser, 1979. Koch-Weser J: Drug therapy, bretylium.  N Engl J Med  1979; 300:473.

Koehler et al., 1983. Koehler RC, Chandra N, Guerci AD, et al: Augmentation of cerebral perfusion by simultaneous chest compression and lung inflation with abdominal binding after cardiac arrest in dogs.  Circulation  1983; 67:266.

Koehler et al., 1985a. Koehler RC, Michael JR, Guerci AD, et al: Beneficial effect of epinephrine infusion on cerebral and myocardial blood flow during CPR.  Ann Emerg Med  1985; 14:744.

Koehler and Michael, 1985b. Koehler RC, Michael JR: Cardiopulmonary resuscitation, brain blood flow, and neurologic recovery.  Crit Care Clin  1985; 1:205.

Kottmeier and Gravenstein, 1968. Kottmeier CA, Gravenstein JS: The parasympathomimetic activity of atropine and atropine methylbromide.  Anesthesiology  1968; 29:1125.

Kouwenhoven et al., 1960. Kouwenhoven WB, Jude JR, Knickerbocker GG: Closed-chest cardiac massage.  JAMA  1960; 173:1064.

Krischer et al., 1989. Krischer JP, Fine EF, Weisfeldt ML, et al: Comparison of prehospital conventional and simultaneous compression-ventilation cardiopulmonary resuscitation.  Crit Care Med  1989; 17:1263.

Kudenchuk et al., 1999. 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.

Kuhn et al., 1991. Kuhn C, Juchems R, Frese W: Evidence for the “cardiac pump theory” in cardiopulmonary resuscitation in man by transesophageal echocardiography.  Resuscitation  1991; 22:275.

Kuntzman et al., 1970. Kuntzman R, Tsai I, Chang R: Disposition of bretylium in man and rat: A sensitive chemical method for its estimation in plasma and urine.  Clin Pharmacol Ther  1970; 11:829.

Langhelle et al., 2002. Langhelle A, Stromme T, Sunde K, et al: Inspiratory impedance threshold valve during CPR.  Resuscitation  2002; 52:39.

Lasbennes et al., 1983. Lasbennes F, Sercombe R, Seylaz J: Monoamine oxidase activity in brain microvessels determined using natural and artificial substances: Relevance to the blood-brain barrier.  J Cereb Blood Flow Metab  1983; 3:521.

Lee and Bullock, 1991. Lee CJ, Bullock LJ: Determining the pulse for infant CPR: Time for a change?.  Milit Med  1991; 156:190.

Lee et al., 1994. Lee G, Antognini JF, Gronert GA: Complete recovery after prolonged resuscitation and cardiopulmonary bypass for hyperkalemic cardiac arrest.  Anesth Analg  1994; 79:172.

Lee et al., 1981. Lee HR, Wilder RJ, Downs P, et al: MAST augmentation of external cardiac compression: Role of changing intrapleural pressure.  Ann Emerg Med  1981; 10:560.

Lee et al., 1984. Lee SK, Vaagenes P, Safar P, et al: Effect of cardiac arrest time on cortical cerebral blood flow generated by subsequent standard external cardiopulmonary resuscitation in rabbits.  Ann Emerg Med  1984; 13:385.

Leigh and Belton, 1949. Leigh MD, Belton MK: Anesthetic deaths in infants and children.  J Mich Med Soc  1949; 48:871.

Levine et al., 1987. Levine R, Gorayeb M, Safar P, et al: Cardiopulmonary bypass after cardiac arrest and prolonged closed-chest CPR in dogs.  Ann Emerg Med  1987; 16:620.

Levine et al., 1997. Levine RL, Wayne MA, Miller CC: End-tidal carbon dioxide and outcome of out-of-hospital cardiac arrest.  N Engl J Med  1997; 337:301.

Lichstein et al., 1973. Lichstein E, Chadda KD, Gupta PK: Atrioventricular block with lidocaine therapy.  Am J Cardiol  1973; 31:277.

Lilja et al., 1981. Lilja GP, Long RS, Ruiz E: Augmentation of systolic blood pressure during external cardiac compression by use of the MAST suit.  Ann Emerg Med  1981; 10:182.

Lindner et al., 1993. Lindner KH, Brinkmann A, Pfenninger EG, et al: Effect of vasopressin on hemodynamic variables, organ blood flow, and acid-base status in a pig model of cardiopulmonary resuscitation.  Anesth Analg  1993; 77:427.

Lindner et al., 1997. Lindner KH, Dirks B, Strohmenger HU, et al: Randomized comparison of epinephrine and vasopressin in patients with out-of-hospital ventricular fibrillation.  Lancet  1997; 349:535.

Lindner et al., 1996. Lindner KH, Prengel AW, Brinkmann A, et al: Vasopressin administration in refractory cardiac arrest.  Ann Intern Med  1996; 124:1061.

Lindner et al., 1995. Lindner KH, Prengel AW, Pfenninger EG, et al: Vasopressin improves vital organ blood flow during closed-chest cardiopulmonary resuscitation in pigs.  Circulation  1995; 91:215.

Livesay et al., 1978. Livesay JJ, Follette DM, Fey KH, et al: Optimizing myocardial supply/demand balance with alpha-adrenergic drugs during cardiopulmonary resuscitation.  J Thorac Cardiovasc Surg  1978; 76:244.

Longstreth et al., 1986. Longstreth WT, Diehr P, Cobb LA, et al: Neurologic outcome and blood glucose levels during out-of-hospital cardiopulmonary resuscitation.  Neurology  1986; 36:1186.

Longstreth and Inui, 1984. Longstreth WT, Inui TS: High blood glucose level on hospital admission and poor neurological recovery after cardiac arrest.  Ann Neurol  1984; 15:59.

Losek et al., 1999. Losek JD, Endom E, Dietrich A, et al: Adenosine and pediatric supraventricular tachycardia in the emergency department: Multicenter study and review.  Ann Emerg Med  1999; 33:185.

Luce et al., 1984. Luce JM, Rizk NA, Niskanen RA: Regional blood flow during cardiopulmonary resuscitation in dogs.  Crit Care Med  1984; 12:874.

Luce et al., 1983. Luce JM, Ross BK, O'Quin RJ, et al: Regional blood flow during cardiopulmonary resuscitation in dogs using simultaneous and nonsimultaneous compression and ventilation.  Circulation  1983; 67:258.

Luedtke et al., 1997. Luedtke SA, Kuhn RJ, McCaffrey FM: Pharmacologic management of supraventricular tachycardias in children. Part 2: Atrial flutter, atrial fibrillation, and junctional and atrial ectopic tachycardia.  Ann Pharmacother  1997; 31:1347.

Lurie et al., 1994. Lurie KG, Shultz JJ, Callaham ML, et al: Evaluation of active compression-decompression CPR in victims of out-of-hospital cardiac arrest.  JAMA  1994; 271:1405.

Lurie et al., 2002. Lurie KG, Zielinski T, Voelckel W, et al: Augmentation of ventricular preload during treatment of cardiovascular collapse and cardiac arrest.  Crit Care Med  2002; 30:S162-S165.

MacKenzie et al., 1976. MacKenzie ET, McCulloch J, O'Keane M, et al: Cerebral circulation and norepinephrine: Relevance of the blood-brain barrier.  Am J Physiol  1976; 231:483.

MacKenzie et al., 1964. MacKenzie GJ, Taylor SH, McDonald AH, et al: Haemodynamic effects of external cardiac compression.  Lancet  1964; 1:1342.

Mahoney and Mirick, 1983. Mahoney BD, Mirick MJ: Efficacy of pneumatic trousers in refractory prehospital cardiopulmonary arrest.  Ann Emerg Med  1983; 12:8.

Maier et al., 1984. Maier GW, Tyson GS, Olsen CO, et al: The physiology of external cardiac massage: High-impulse cardiopulmonary resuscitation.  Circulation  1984; 70:86.

Mandapati et al., 2000. Mandapati R, Byrum CJ, Kavey RE, et al: Procainamide for rate control of postsurgical junctional tachycardia.  Pediatr Cardiol  2000; 21:123.

Mann et al., 2002. Mann K, Berg RA, Nadkarni V: Beneficial effects of vasopressin in prolonged pediatric cardiac arrest: A case series.  Resuscitation  2002; 52:149.

Markis and Koch-Weser, 1971. Markis JE, Koch-Weser J: Characterizations and mechanisms of inotropic and chronotropic actions of bretylium tosylate.  J Pharmacol Exp Ther  1971; 178:94.

Marquez et al., 1986. Marquez J, Martin D, Virja MA, et al: Cardiovascular depression secondary to ionic hypocalcemia during hepatic transplantation in humans.  Anesthesiology  1986; 65:457.

Martin et al., 1990. Martin D, Werman HA, Brown CG: Four case studies: high dose epinephrine in cardiac arrest.  Ann Emerg Med  1990; 19:322.

Martin et al., 1986. Martin GB, Carden DL, Nowak RM, et al: Aortic and right atrial pressures during standard and simultaneous compression and ventilation CPR in human beings.  Ann Emerg Med  1986; 15:125.

Martin et al., 1990. Martin GB, Gentile NT, Paradis NA, et al: Effect of epinephrine on end-tidal carbon dioxide monitoring during CPR.  Ann Emerg Med  1990; 19:396.

Martin et al., 1998. Martin GB, Rivers EP, Paradis NA, et al: Emergency department cardiopulmonary bypass in the treatment of human cardiac arrest.  Chest  1998; 113:743.

Martinez et al., 1979. Martinez LR, Holland S, Fitzgerald J, et al: pH homeostasis during cardiopulmonary resuscitation in critically ill patients.  Resuscitation  1979; 7:109.

Marx et al., 1973. Marx GF, Mateo CV, Orkin LR: Computer analysis of postanesthetic deaths.  Anesthesiology  1973; 39:54.

Mateer et al., 1985. Mateer JR, Stueben HA, Thompson BM, et al: Pre-hospital IAC-CPR versus standard CPR: Paramedic resuscitation of cardiac arrests.  Am J Emerg Med  1985; 3:143.

Mattar et al., 1974. Mattar JA, Weil MH, Shubin H, et al: Cardiac arrest in the critically ill: II. Hyperosmolal states following cardiac arrest.  Am J Med  1974; 56:162.

Mattox and Beall, 1976. Mattox KL, Beall AC: Resuscitation of the moribund patient using portable cardiopulmonary bypass.  Ann Thorac Surg  1976; 22:436.

Mauer et al., 1996. Mauer D, Schneider T, Dick W, et al: Active compression-decompression resuscitation: A prospective, randomized study in a two-tiered EMS system with physicians in the field.  Resuscitation  1996; 33:125.

Mazer et al., 2003. Mazer SP, Weisfeldt M, Bai D, et al: Reverse CPR: A pilot study of CPR in the prone position.  Resuscitation  2003; 57:279.

Michael et al., 1984. Michael JR, Guerci AD, Koehler RC, et al: Mechanisms by which epinephrine augments cerebral and myocardial blood perfusion during cardiopulmonary resuscitation in dogs.  Circulation  1984; 69:822.

Miranda and Newton, 2001. Miranda CC, Newton MC: Successful defibrillation in the prone position.  Br J Anaesth  2001; 87:937.

Mogayzel et al., 1995. Mogayzel C, Quan L, Graves JR, et al: Out-of-hospital ventricular fibrillation in children and adolescents: Causes and outcomes.  Ann Emerg Med  1995; 25:484.

Morgan et al., 1993. Morgan CA, Webb RK, Cockings J, et al: The Australian Incident Monitoring Study. Cardiac arrest—an analysis of 2000 incident reports.  Anaesth Intensive Care  1993; 21:626.

Morray et al., 2000. Morray JP, Geiduschek JM, Ramamoorthy C, et al: Anesthesia-related cardiac arrest in children: initial findings of the Pediatric Perioperative Cardiac Arrest (POCA) Registry.  Anesthesiology  2000; 93:6.

Morray, 2002. Morray JP: Anesthesia-related cardiac arrest in children. An update.  Anesthesiol Clin North Am  2002; 20:1.

Morris et al., 1997. Morris DC, Dereczyk BE, Grzybowski M, et al: Vasopressin can increase coronary perfusion pressure during human cardiopulmonary resuscitation.  Acad Emerg Med  1997; 4:878.

Muir et al., 1990. Muir JD, Randalls PR, Smith GB: End tidal carbon dioxide detector for monitoring cardiopulmonary resuscitation.  Br Med J  1990; 301:41.

Mullins et al., 1999. Mullins ME, Pinnick RV, Terhes JM: Life threatening diphenhydramine overdose treated with charcoal hemoperfusion and hemodialysis.  Ann Emerg Med  1999; 33:104-107.

Myers, 1979. Myers R: Lactic acid accumulation as a cause of brain edema and cerebral necrosis resulting from oxygen deprivation.   In: Korbin R, Gilleminault C, ed. Advances in perinatal neurology,  New York: Spectrum; 1979:84.

Nakatani et al., 1999. Nakatani K, Yukioka H, Fujimori M, et al: Utility of colorimetric end-tidal carbon dioxide detector for monitoring during prehospital cardiopulmonary resuscitation.  Am J Emerg Med  1999; 17:203.

Nattel, 1993. Nattel S: Comparative mechanisms of action of antiarrhythmic drugs.  Am J Cardiol  1993; 72:13F.

Newland et al., 2002. Newland MC, Ellis SJ, Lydiatt CA, et al: Anesthetic-related cardiac arrest and its mortality: A report covering 72,959 anesthetics over 10 years from a US teaching hospital.  Anesthesiology  2002; 97:108.

Niemann et al., 1992. Niemann JT, Cairns CB, Sharma J, et al: Treatment of prolonged ventricular fibrillation: immediate countershock versus high-dose epinephrine and CPR preceding countershock.  Circulation  1992; 85:281.

Niemann et al., 2003. Niemann JT, Garner D, Khaleeli E, et al: Milrinone facilitates resuscitation from cardiac arrest and attenuates post-resuscitation myocardial dysfunction.  Circulation  2003; 108:3031.

Niemann et al., 1981a. Niemann JT, Rosborough JP, Hausknecht M, et al: Blood flow without cardiac compression during closed chest CPR.  Crit Care Med  1981; 9:380.

Niemann et al., 1980. Niemann JT, Rosborough JP, Hausknecht M, et al: Cough-CPR: documentation of systemic perfusion in man and in an experimental model: A “window” to the mechanism of blood flow in external CPR.  Crit Care Med  1980; 8:141.

Niemann et al., 1981b. Niemann JT, Rosborough JP, Hausknecht M, et al: Pressure-synchronized cineangiography during experimental cardiopulmonary resuscitation.  Circulation  1981; 64:985.

Niemann et al., 1990. Niemann JT, Rosborough JP, Pelikin PCD: Hemodynamic determinants of subdiaphragmatic venous return during closed-chest CPR in a canine cardiac arrest model.  Ann Emerg Med  1990; 19:1232.

Niemann et al., 1984. Niemann JT, Rosborough JP, Ung S, et al: Hemodynamic effects of continuous abdominal binding during cardiac arrest and resuscitation.  Am J Cardiol  1984; 53:269.

Niskanen, 1997. Niskanen RA: Automated external defibrillators: Experiences with their use and options for their further development.  New Horiz  1997; 5:137.

Nolan et al., 1998. Nolan J, Smith G, Evans R, et al: The United Kingdom pre-hospital study of active compression-decompression resuscitation.  Resuscitation  1998; 37:119.

Nolan et al., 2003. Nolan JP, Morley PT, Vanden Hoek TL, et al: Therapeutic hypothermia after cardiac arrest. An advisory statement by the Advanced Life Support Task Force of the International Liaison Committee on Resuscitation.  Circulation  2003; 108:118-121.

Olsson and Hallen, 1988. Olsson GL, Hallen B: Cardiac arrest during anaesthesia. A computer-aided study in 250,543 anaesthetics.  Acta Anaesthesiol Scand  1988; 32:653.

Orlowski et al., 1989. Orlowski JP, Julius CJ, Petras RE, et al: The safety of intraosseous infusions: Risks of fat and bone marrow emboli to the lungs.  Ann Emerg Med  1989; 18:1062.

Orlowski, 1980. Orlowski JP: Cardiopulmonary resuscitation in children.  Pediatr Clin North Am  1980; 27:495.

Ornato et al., 1988. Ornato JP, Gonzalez ER, Garnett AR, et al: Effect of cardiopulmonary resuscitation compression rate on end-tidal carbon dioxide concentration and arterial pressure in man.  Crit Care Med  1988; 16:241.

Pannier and Leusen, 1968. Pannier JL, Leusen I: Contraction characteristics of papillary muscle during changes in acid-base composition of the bathing fluid.  Arch Int Physiol Biochem  1968; 76:624.

Paradis et al., 1989. Paradis NA, Martin GB, Goetting MG, et al: Simultaneous aortic, jugular bulb, and right atrial pressures during cardiopulmonary resuscitation in humans: Insights into mechanisms.  Circulation  1989; 80:361.

Paradis et al., 1990. Paradis NA, Martin GB, Rivers EP, et al: Coronary perfusion pressure and the return of spontaneous circulation in human cardiopulmonary resuscitation.  JAMA  1990; 263:1106.

Parra et al., 2000. Parra DA, Totapally BR, Zahn E, et al: Outcome of cardiopulmonary resuscitation in a pediatric cardiac intensive care unit.  Crit Care Med  2000; 28:3296-3300.

Pearson and Redding, 1965. Pearson JW, Redding JS: Influence of peripheral vascular tone on resuscitation.  Anesth Analg  1965; 44:746.

Peleska, 1963. Peleska B: Cardiac arrhythmias following condenser discharges and their dependence upon strength of current and phase of cardiac cycle.  Circ Res  1963; 13:21.

Perry et al., 1996. Perry JC, Fenrich AL, Hulse JE, et al: Pediatric use of intravenous amiodarone: efficacy and safety in critically ill patients from a multicenter protocol.  J Am Coll Cardiol  1996; 27:1246-1250.

Perry et al., 1993. Perry JC, Knilans TK, Marlow D, et al: Intravenous amiodarone for life-threatening tachyarrhythmias in children and young adults.  J Am Coll Cardiol  1993; 22:95.

Phillips et al., 1983. Phillips SJ, Ballentine B, Slonine D, et al: Percutaneous initiation of cardiopulmonary bypass.  Ann Thorac Surg  1983; 36:223.

Pionkowski et al., 1983. Pionkowski RS, Thompson BM, Gruchow HW, et al: Resuscitation time in ventricular fibrillation: A prognostic indicator.  Ann Emerg Med  1983; 12:733.

Plaisance et al., 1997. Plaisance P, Adnet F, Vicaut E, et al: Benefit of active compression-decompression cardiopulmonary resuscitation as a prehospital advanced cardiac life support. A randomized multicenter study.  Circulation  1997; 95:955.

Plaisance et al., 2000. Plaisance P, Lurie KG, Payen D: Inspiratory impedance during active compression-decompression cardiopulmonary resuscitation—A randomized evaluation in patients in cardiac arrest.  Circulation  2000; 101:989-994.

Podrid, 1995. Podrid PJ: Amiodarone: Reevaluation of an old drug.  Ann Intern Med  1995; 122:689.

Prado et al., 1988. Prado R, Ginsberg MD, Dietrich WD, et al: Hyperglycemia increases infarct size in collaterally perfused but not end-arterial vascular territories.  J Cereb Blood Flow Metab  1988; 8:186.

Prengel et al., 1996a. Prengel AW, Lindner KH, Keller A, et al: Cardiovascular function during the post-resuscitation phase after cardiac arrest in pigs: A comparison of epinephrine versus vasopressin.  Crit Care Med  1996; 24:2014.

Prengel et al., 1996b. Prengel AW, Lindner KH, Keller A: Cerebral oxygenation during cardiopulmonary resuscitation with epinephrine and vasopressin in pigs.  Stroke  1996; 27:1241.

Prengel et al., 1998. Prengel AW, Lindner KH, Wenzel V, et al: Splanchnic and renal blood flow after cardiopulmonary resuscitation with epinephrine and vasopressin in pigs.  Resuscitation  1998; 38:19.

Pretto et al., 1987. Pretto E, Safar P, Saito R, et al: Cardiopulmonary bypass after prolonged cardiac arrest in dogs.  Ann Emerg Med  1987; 16:611.

Pulsinelli et al., 1983. Pulsinelli WA, Levy DE, Sigsbee B, et al: Increased damage after ischemic stroke in patients with hyperglycemia with or without established diabetes mellitus.  Am J Med  1983; 74:540.

Quan et al., 1992. Quan L, Graves JR, Kinder DR, et al: Transcutaneous cardiac pacing in the treatment of out-of-hospital pediatric cardiac arrests.  Ann Emerg Med  1992; 21:905.

Rackow et al., 1961. Rackow H, Salanitre E, Green LT: Frequency of cardiac arrest associated with anesthesia in infants and children.  Pediatrics  1961; 28:697.

Raessler et al., 1988. Raessler KL, Kern KB, Sanders AB, et al: Aortic and right atrial systolic pressures during cardiopulmonary resuscitation: A potential indicator of the mechanism of blood flow.  Am Heart J  1988; 115:1021.

Rainer, 1957. Rainer EH: Respiratory and cardiac arrest during anaesthesia in children.  Br Med J  1957; 2:1024.

Ralston et al., 1982. Ralston SH, Babbs CF, Niebauer MJ: Cardiopulmonary resuscitation with interposed abdominal compression in dogs.  Anesth Analg  1982; 61:645.

Ralston et al., 1984. Ralston SH, Voorhees WD, Babbs CF: Intrapulmonary epinephrine during prolonged cardiopulmonary resuscitation: Improved regional blood flow and resuscitation in dogs.  Ann Emerg Med  1984; 13:79.

Redding and Pearson, 1963. Redding JS, Pearson JW: Evaluation for drugs for cardiac resuscitation.  Anesthesiology  1963; 24:203.

Redding, 1971. Redding JS: Abdominal compression in cardiopulmonary resuscitation.  Anesth Analg  1971; 50:668.

Reich et al., 1990. Reich H, Angelos M, Safar P, et al: Cardiac resuscitability with cardiopulmonary bypass after increasing ventricular fibrillation times in dogs.  Ann Emerg Med  1990; 19:887.

Reichman et al., 1990. Reichman RT, Joyo CI, Dembitsky WP, et al: Improved patient survival after cardiac arrest using a cardiopulmonary support system.  Ann Thorac Surg  1990; 49:101.

Reidenberg et al., 1975. Reidenberg MM, Drayer DE, Levy M, et al: Polymorphic acetylation procainamide in man.  Clin Pharmacol Ther  1975; 17:722.

Reis et al., 2002. Reis AG, Nadkarni V, Perondi MB, et al: A prospective investigation into the epidemiology of in-hospital pediatric cardiopulmonary resuscitation using the international Utstein reporting style.  Pediatrics  2002; 109:200.

Resnekov, 1981. Resnekov L: Calcium antagonist drugs: Myocardial preservation and reduced vulnerability to ventricular fibrillation during CPR.  Crit Care Med  1981; 9:360.

Rich et al., 1981. Rich S, Wix HL, Shapiro EP: Clinical assessment of heart size and valve motion during cardiopulmonary resuscitation by two-dimensional echocardiography.  Am Heart J  1981; 102:368.

Rivers et al., 1992. Rivers EP, Martin GB, Smithine H, et al: The clinical implications of continuous central venous oxygen saturation during human CPR.  Ann Emerg Med  1992; 21:1094.

Roberts et al., 1994. Roberts SA, Viana MA, Nazari J, et al: Invasive and noninvasive methods to predict the long-term efficacy of amiodarone: A compilation of clinical observations using meta-analysis.  PACE Pacing Clin Electrophysiol  1994; 17:1590.

Rogers et al., 1979. Rogers MC, Nugent SK, Stidham GL: Effects of closed-chest cardiac massage on intracranial pressure.  Crit Care Med  1979; 7:454.

Romhilt et al., 1972. Romhilt DW, Bloodfield SS, Lipicky RJ: Evaluation of bretylium tosylate for the treatment of premature ventricular contractions.  Circulation  1972; 45:800.

Rudikoff et al., 1980. Rudikoff MT, Maughan WL, Effron M, et al: Mechanism of blood flow during cardiopulmonary resuscitation.  Circulation  1980; 61:345.

Rudolph and Yuan, 1966. Rudolph AM, Yuan S: Response of the pulmonary vasculature to hypoxia and hydrogen ion concentration changes.  J Clin Invest  1966; 45:399.

Rumball and MacDonald, 1997. Rumball CJ, MacDonald D: The PTL, Combitube, laryngeal mask, and oral airway: A randomized prehospital comparative study of ventilatory device effectiveness and cost-effectiveness in 470 cases of cardiorespiratory arrest.  Prehosp Emerg Care  1997; 1:1.

Sack et al., 1992. Sack JB, Kesselbrenner MB, Bregman D: Survival from in-hospital cardiac arrest with interposed abdominal counterpulsation during cardiopulmonary resuscitation.  JAMA  1992; 267:379.

Safar et al., 1961. Safar P, Brown TC, Holtey WJ, et al: Ventilation and circulation with closed-chest cardiac massage in man.  JAMA  1961; 176:92.

Salem et al., 1975. Salem MR, Bennett EJ, Schweiss JF, et al: Cardiac arrest related to anesthesia. Contributing factors in infants and children.  JAMA  1975; 233:238.

Samarkandi et al., 1994. Samarkandi AH, Seraj MA, el Dawlatly A, et al: The role of laryngeal mask airway in cardiopulmonary resuscitation.  Resuscitation  1994; 28:103.

Samson et al., 2003. Samson RA, Berg RA, Bingham R, et al: Pediatric Advanced Life Support Task Force, International Liaison Committee on Resuscitation: Use of automated external defibrillators for children: An update: An advisory statement from the Pediatric Advanced Life Support Task Force, International Liaison Committee on Resuscitation.  Circulation  2003; 107:3250.

Sanders et al., 1985a. Sanders AB, Atlas M, Ewy GA, et al: Expired PCO2 as an index of coronary perfusion pressure.  Am J Emerg Med  1985; 3:147.

Sanders et al., 1982. Sanders AB, Ewy GA, Alferness CA, et al: Failure of one method of simultaneous chest compression, ventilation, and abdominal binding during CPR.  Crit Care Med  1982; 10:50913.

Sanders et al., 1984b. Sanders AB, Ewy GA, Taft TV: Prognostic and therapeutic importance of the aortic diastolic pressure in resuscitation from cardiac arrest.  Crit Care Med  1984; 12:871.

Sanders et al., 1984a. Sanders AB, Kern KB, Ewy GA, et al: Improved outcome from cardiac arrest with open-chest massage.  Ann Emerg Med  1984; 13:672.

Sanders et al., 1985b. Sanders AB, Kern KB, Ewy GA: Time limitations for open-chest cardiopulmonary resuscitation from cardiac arrest.  Crit Care Med  1985; 13:897.

Sanders et al., 1990. Sanders AB, Kern KB, Fonken S, et al: The role of bicarbonate and fluid loading in improving resuscitation from prolonged cardiac arrest with rapid manual chest compression CPR.  Ann Emerg Med  1990; 19:1.

Sanders et al., 1989. 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.

Scheidegger et al., 1977. Scheidegger D, Drop LJ, Laver MB: Interaction between vasoactive drugs and plasma ionized calcium.  Intensive Care Med  1977; 3:200.

Scheinman et al., 1975. Scheinman MM, Thorburn D, Abbott JA: Use of atropine in patients with acute myocardial infarction and sinus bradycardia.  Circulation  1975; 52:627.

Schierhout and Roberts, 1998. Schierhout G, Roberts I: Fluid resuscitation with colloid or crystalloid solutions in critically ill patients: A systematic review of randomized trials.  BMJ  1998; 316:961.

Schleien et al., 1992. Schleien CL, Caceres MJ, Kuluz JW, et al: Light and electron microscopic blood-brain barrier (BBB) changes following cardiopulmonary resuscitation in young piglets.  Anesthesiology  1992; 77(suppl IIIA):712.

Schleien et al., 1986. 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; 73:809.

Schleien et al., 1989. Schleien CL, Koehler RC, Gervais H, et al: Organ blood flow and somatosensory evoked potentials before and after cardiopulmonary resuscitation with epinephrine and phenylephrine.  Circulation  1989; 79:1332.

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

Schleien et al., 1992. Schleien CL, Kuluz JW, Shaffner DH, et al: Cardiopulmonary resuscitation.   In: Rogers MC, ed. Textbook of pediatric intensive care,  2nd ed.. Baltimore: Williams & Wilkins; 1992:52.

Schumacher et al., 1968. Schumacher RR, Lieberson AD, Childress RH, et al: Hemodynamic effects of lidocaine in patients with heart disease.  Am J Cardiol  1968; 24:191.

Schwab et al., 1995. Schwab TM, Callaham ML, Madsen CD, et al: A randomized clinical trial of active compression-decompression CPR versus standard CPR in out-of-hospital cardiac arrest in two cities.  JAMA  1995; 273:1261.

Sessler et al., 1987. Sessler D, Mills P, Gregory G, et al: Effects of bicarbonate on arterial and brain intracellular pH in neonatal rabbits recovering from hypoxic lactic acidosis.  J Pediatr  1987; 111:817.

Shaffner et al., 1999. Shaffner DH, Eleff SM, Brambrink AM, et al: Effect of arrest time and cerebral perfusion pressure during cardiopulmonary resuscitation on cerebral blood flow, metabolism, ATP recovery, and pH in dogs.  Crit Care Med  1999; 27:1335.

Shaffner et al., 1998. Shaffner DH, Eleff SM, Koehler RC, et al: Effect of the no-flow interval and hypothermia on cerebral blood flow and metabolism during cardiopulmonary resuscitation in dogs.  Stroke  1998; 29:2607.

Shaffner et al., 1990. Shaffner DH, Schleien CL, Koehler RC, et al: Cerebral and coronary perfusion with vest cardiopulmonary resuscitation in piglets.  Crit Care Med  1990; 18:S243.

Sharff et al., 1984. Sharff JA, Pantley G, Noel E: Effect of time on regional organ perfusion during two methods of cardiopulmonary resuscitation.  Ann Emerg Med  1984; 13:649.

Sieber and Traystman, 1992. Sieber FE, Traystman RJ: Special issues: Glucose and the brain.  Crit Care Med  1992; 20:104.

Siemkowicz and Hansen, 1978. Siemkowicz E, Hansen AJ: Clinical restitution following cerebral ischemia in hypo-, normo- and hyperglycemic rats.  Acta Neurol Scand  1978; 58:1.

Siesjo, 1984. Siesjo BK: Cerebral circulation and metabolism.  Neurosurgery  1984; 60:883.

Singh et al., 1989. Singh BN, Venkatesh N, Nademanee K, et al: The historical development, cellular electrophysiology and pharmacology of amiodarone.  Prog Cardiovasc Dis  1989; 31:249.

Singh, 1996. Singh BN: Antiarrhythmic actions of amiodarone: A profile of a paradoxical agent.  Am J Cardiol  1996; 78:41.

Sladen, 1984. Sladen A: Closed-chest massage.  JAMA  1984; 251:3137.

Snyder et al., 1991. Snyder AV, Salloum LJ, Barone JE, et al: Predicting short-term outcome of cardiopulmonary resuscitation using central venous oxygen tension measurements.  Crit Care Med  1991; 19:111.

Sprung et al., 2003. Sprung J, Warner ME, Contreras MG, et al: Predictors of survival following cardiac arrest in patients undergoing noncardiac surgery: A study of 518,294 patients at a tertiary referral center.  Anesthesiology  2003; 99:259.

Stacpoole et al., 1987. Stacpoole PAW, Gonzalez MG, Vlasak J, et al: Dichloroacetate derivatives. Metabolic effects and pharmacodynamics in normal rats.  Life Sci  1987; 41:2167.

Stacpoole et al., 1988. Stacpoole PAW, Lorenz AC, Thomas RG, et al: Dichloroacetate in the treatment of lactic acidosis.  Ann Intern Med  1988; 108:58.

Stacpoole et al., 1992. Stacpoole PAW, Wright EC, Baumgartner TG: A controlled clinical trial of dichloroacetate for treatment of lactic acidosis in adults.  N Engl J Med  1992; 327:1564.

Stacpoole, 1989. Stacpoole PAW: The pharmacology of dichloroacetate.  Metab Clin Exp  1989; 38:1124.

Steinhart et al., 1983. Steinhart CR, Permutt S, Gurtner GH, et al: Beta-adrenergic activity and cardiovascular response to severe respiratory acidosis.  Am J Physiol  1983; 244:H46.

Stiell et al., 1992. Stiell IG, Hebert PC, Weitzman BN, et al: High-dose epinephrine in adult cardiac arrest.  N Engl J Med  1992; 327:1045.

Stiell et al., 1996. Stiell IG, Hebert PC, Wells GA, et al: The Ontario trial of active compression-decompression cardiopulmonary resuscitation for in-hospital and prehospital cardiac arrest.  JAMA  1996; 275:1417.

Stone et al., 1998. Stone BJ, Chantler PJ, Baskett PJ: The incidence of regurgitation during cardiopulmonary resuscitation: A comparison between the bag valve mask and laryngeal mask airway.  Resuscitation  1998; 38:3.

Strohmenger et al., 1996. Strohmenger HU, Lindner KH, Keller A, et al: Effects of graded doses of vasopressin on median fibrillation frequency in a porcine model of cardiopulmonary resuscitation: Results of a prospective, randomized, controlled trial.  Crit Care Med  1996; 24:1360.

Stueven et al., 1985. Stueven HA, Aprahamian C, Tonsfeldt DJ, et al: Lack of effectiveness of calcium chloride in refractory asystole.  Ann Emerg Med  1985; 14:630.

Stueven et al., 1985. Stueven HA, Thompson BM, Aprahamian C, et al: The effectiveness of calcium chloride in refractory electromechanical dissociation.  Ann Emerg Med  1985; 14:626.

Sun et al., 1987. Sun JH, Filley GF, Hord K, et al: Carbicarb: an effective substitute for NaHCO3 for the treatment of acidosis.  Surgery  1987; 102:835-839.

Sun et al., 1992. Sun W, Huang F, Kung K, et al: Successful cardiopulmonary resuscitation of two patients in the prone position reversed precordial compression.  Anesthesiology  1992; 77:202.

Suominen et al., 2000. Suominen P, Olkkola KT, Voipio V, et al: Utstein style reporting of in-hospital paediatric cardiopulmonary resuscitation.  Resuscitation  2000; 45:17-25.

Swenson et al., 1988. Swenson RD, Weaver WD, Niskanen RA, et al: Hemodynamics in humans during conventional and experimental methods of cardiopulmonary resuscitation.  Circulation  1988; 78:630.

Tacker et al., 1981. Tacker Jr WA, Babbs CF, Abendschein DR, et al: Transchest defibrillation under conditions of hypothermia.  Crit Care Med  1981; 9:390.

Tang et al., 1991. Tang W, Weil MH, Gazmuri RJ, et al: Pulmonary ventilation/perfusion defects induced by epinephrine during cardiopulmonary resuscitation.  Circulation  1991; 84:2101.

Tang et al., 2002. Tang W, Weil MH, Jorgenson D, et al: Fixed-energy biphasic waveform defibrillation in a pediatric model of cardiac arrest and resuscitation.  Crit Care Med  2002; 30:2736-2741.

Tang et al., 1997. Tang W, Weil MH, Schock RB: Phased chest and abdominal compression-decompression. A new option for cardiopulmonary resuscitation.  Circulation  1997; 95:1335.

Taylor et al., 1977. Taylor GJ, Tucker WM, Greene HL, et al: Importance of prolonged compression during cardiopulmonary resuscitation in man.  N Engl J Med  1977; 296:1515.

Tisherman et al., 1991. Tisherman SA, Safar P, Abramson NS, et al: Feasibility of emergency cardiopulmonary bypass for resuscitation from CPR-resistant cardiac arrest: A preliminary report.  Ann Emerg Med  1991; 20:491.

Tobias et al., 1994. Tobias JD, Mencio GA, Atwood R, et al: Intraoperative cardiopulmonary resuscitation in the prone position.  J Pediatr Surg  1994; 29:1537.

Todres and Rogers, 1975. Todres ID, Rogers MC: Methods of external cardiac massage in the newborn infant.  J Pediatr  1975; 86:781.

Urban et al., 1988. Urban P, Scheidegger D, Buchmann B, et al: Cardiac arrest and blood ionized calcium levels.  Ann Intern Med  1988; 109:110.

Urban et al., 1986. Urban P, Scheidegger D, Buchmann B, et al: The hemodynamic effects of heparin and their relation to ionized calcium levels.  J Thorac Cardiovasc Surg  1986; 91:303.

Vacanti et al., 1970. Vacanti CJ, VanHouten RJ, Hill RC: A statistic analysis of the relationship of physical status to postoperative mortality in 68,388 cases.  Anesth Analg  1970; 49:564.

Voelckel et al., 2002. Voelckel WG, Lurie KG, McKnite S, et al: Effects of epinephrine and vasopressin in a piglet model of prolonged ventricular fibrillation and cardiopulmonary resuscitation.  Crit Care Med  2002; 30:957-962.

Voll and Auer, 1991. Voll CL, Auer RN: Insulin attenuates ischemic brain damage independent of its hypoglycemic effect.  J Cereb Blood Flow Metab  1991; 11:1006.

Voll and Auer, 1988. Voll CL, Auer RN: The effect of postischemic blood glucose levels on ischemic brain damage in the rat.  Ann Neurol  1988; 24:638.

Von Seggern et al., 1986. Von Seggern K, Egar M, Fuhrman BP: Cardiopulmonary resuscitation in a pediatric ICU.  Crit Care Med  1986; 14:275-277.

Voorhees et al., 1983. Voorhees WD, Niebauer MJ, Babbs CF: Improved oxygen delivery during cardiopulmonary resuscitation with interposed abdominal compressions.  Ann Emerg Med  1983; 12:128.

Waisman and Waisman, 1997. Waisman M, Waisman D: Bone marrow infusion in adults.  J Trauma  1997; 42:288.

Walker et al., 1984. Walker JW, Bruestle JC, White BC, et al: Perfusion of the cerebral cortex by use of abdominal counterpulsation during cardiopulmonary resuscitation.  Am J Emerg Med  1984; 2:391.

Walsh et al., 1997. Walsh EP, Saul JP, Sholler GF, et al: Evaluation of a staged treatment protocol for rapid automatic junctional tachycardia after operation for congenital heart disease.  J Am Coll Cardiol  1997; 29:1046.

Ward et al., 1989. Ward KR, Sullivan RJ, Zelenak RR, et al: A comparison of interposed abdominal compression CPR and standard CPR by monitoring end-tidal PCO2.  Ann Emerg Med  1989; 18:831.

Wargovich et al., 1988. Wargovich TJ, MacDonald RG, Hill JA, et al: Myocardial metabolic and hemodynamic effects of dichloroacetate in coronary artery disease.  Am J Cardiol  1988; 61:65.

Wayne et al., 1995. Wayne MA, Levine RL, Miller CC: Use of end-tidal carbon dioxide to predict outcome in prehospital cardiac arrest.  Ann Emerg Med  1995; 25:762.

Weale and Rothwell-Jackson, 1962. Weale FE, Rothwell-Jackson RL: The efficiency of cardiac massage.  Lancet  1962; 1:990.

Weaver et al., 1982. Weaver WD, Cobb LA, Copass MK, et al: Ventricular defibrillation: A comparative trial using 175-J and 320-J shocks.  N Engl J Med  1982; 307:1101.

Weaver et al., 1988. Weaver WD, Hill D, Fahrenbruch CE, et al: Use of the automatic external defibrillator in the management of out-of-hospital cardiac arrest.  N Engl J Med  1988; 319:661.

Weil et al., 1986. Weil MH, Rackow EC, Trevino R, et al: Differences in acid-base state between venous and arterial blood during cardiopulmonary resuscitation.  N Engl J Med  1986; 315:153.

Weiser et al., 1962. Weiser FM, Adler LN, Kuhn LA: Hemodynamic effects of closed and open chest cardiac resuscitation in normal dogs and those with acute myocardial infarction.  Am J Cardiol  1962; 10:555.

Wenzel et al., 2004. Wenzel V, Krismer AC, Arntz HR, et al: European Resuscitation Council Vasopressor during Cardiopulmonary Resuscitation Study Group: A comparison of vasopressin and epinephrine for out-of-hospital cardiopulmonary resuscitation.  N Engl J Med  2004; 350:105.

Wenzel and Lindner, 2002. Wenzel V, Lindner KH: Arginine vasopressin during cardiopulmonary resuscitation: Laboratory evidence, clinical experience and recommendations, and a view to the future.  Crit Care Med  2002; 30:S157.

Werner et al., 1981. Werner JA, Greene HL, Janko CL, et al: Visualization of cardiac valve motion in man during external chest compression using two-dimensional echocardiography: Implications regarding the mechanism of blood flow.  Circulation  1981; 63:1417.

White et al., 1983. White BC, Winegar CD, Wilson RF, et al: Possible role of calcium blockers in cerebral resuscitation: A review of the literature and synthesis for future studies.  Crit Care Med  1983; 11:202.

Wilder et al., 1963. Wilder RJ, Weir D, Rush BF, et al: Methods of coordinating ventilation and closed chest cardiac massage in the dog.  Surgery  1963; 53:186.

Winquist et al., 1982. Winquist RJ, Webb RC, Bohr OF: Relaxation to transmural nerve stimulation and exogenously added norepinephrine in porcine vessels. A study utilizing cerebrovascular intrinsic tone.  Circ Res  1982; 51:769.

Wolcke et al., 2003. Wolcke BB, Mauer DK, Schoefmann MF, et al: Comparison of standard cardiopulmonary resuscitation versus the combination of active compression-decompression cardiopulmonary resuscitation and an inspiratory impedance threshold device for out-of-hospital cardiac arrest.  Circulation  2003; 108:2201.

Woo et al., 1988. Woo E, Chan YW, Yu YL, et al: Admission glucose level in relation to mortality and morbidity outcome in 252 stroke patients.  Stroke  1988; 19:185.

Wood et al., 1963. Wood WB, Manley Jr ES, Woodbury RA: The effects of CO2 induced respiratory acidosis on the depressor and pressor components of the dog's blood pressure to epinephrine.  J Pharmacol Exp Ther  1963; 139:238.

Worthley, 1976. Worthley LS: Sodium bicarbonate in cardiac arrest.  Lancet  1976; 2:903.

Xiao et al., 1998. Xiao F, Safar P, Radovsky A: Mild protective and resuscitative hypothermia for asphyxial cardiac arrest in rats.  Am J Emerg Med  1998; 16:17.

Yakaitis et al., 1979. Yakaitis RW, Otto CW, Blitt CD: Relative importance of alpha and beta adrenergic receptors during resuscitation.  Crit Care Med  1979; 7:293.

Yu et al., 2002. Yu T, Weil MH, Tang W, et al: Adverse outcomes of interrupted precordial compression during automated defibrillation.  Circulation  2002; 106:368.

Zaritsky et al., 1995. Zaritsky A, Nadkarni V, Hazinski MF, et al: Recommended guidelines for uniform reporting of pediatric advanced life support: The pediatric Utstein Style. A statement for healthcare professionals from a task force of the American Academy of Pediatrics, the American Heart Association, and the European Resuscitation Council Writing Group.  Circulation  1995; 92:2006.

Zipes et al., 1975. Zipes DP, Fischer J, King RM, et al: Termination of ventricular fibrillation in dogs by depolarizing a critical amount of myocardium.  Am J Cardiol  1975; 36:37.