Pediatric Basic and Advanced Life Support: 2010 International Consensus on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations

Medical emergency teams (METs) or rapid response teams (RRTs) have been shown to be effective in preventing respiratory and cardiac arrests in selected pediatric inpatient settings.

Family presence during resuscitations has been shown to be beneficial for the grieving process and in general was not found to be disruptive. Thus, family presence is supported if it does not interfere with the resuscitative effort.

The introduction of METs or RRTs was associated with a decrease in pediatric hospital mortality in 1 LOE 3 meta-analysis³  and 3 pediatric LOE 3 studies with historic controls.^4,–,6  The introduction of a MET or RRT was associated with

• a decrease in respiratory but not cardiac arrest in 1 LOE 3⁷  study with historic controls

• a decrease in preventable total number of arrests in 1 LOE 3 study compared with a retrospective chart review⁸ 

• a decrease in total number of arrests in 2 LOE 3^4,8  studies

• a decrease in preventable cardiac arrests in 1 LOE 3⁶  study

• a decrease in cardiac arrest and non–pediatric intensive care unit (PICU) mortality in 1 LOE 3⁹  pediatric cohort study using historical controls

Pediatric RRT or MET systems may be beneficial to reduce the risk of respiratory and/or cardiac arrest in hospitalized pediatric patients outside an intensively monitored environment.

Is it the team or the staff education associated with MET or RRT implementation that leads to improved patient outcomes? Is the team effectiveness due to validated team activation criteria or specific team composition? Do the benefits attributed to these teams extend to children in a community hospital setting?

Ten studies (LOE 2¹⁰ ; LOE 3¹¹ ; LOE 4^12,–,19 ) documented that parents wish to be given the option of being present during the resuscitation of their children. One LOE 2,¹⁰  1 LOE 3,¹¹  2 LOE 4,^13,19  and 1 LOE 5²⁰  studies confirmed that most parents would recommend parent presence during resuscitation.

One LOE 2,¹⁰  1 LOE 3,¹¹  6 LOE 4,^{12,14,19,21,–,23}  and 2 LOE 5^20,24  studies of relatives present during the resuscitation of a family member reported that they believed their presence was beneficial to the patient.

One LOE 2,¹⁰  1 LOE 3,¹¹  6 LOE 4,^{12,13,16,–,19}  and 1 LOE 5²⁴  studies reported that most relatives present during the resuscitation of a family member benefited from the experience. One LOE 3,¹¹  4 LOE 4,^12,13,20,21  and 2 LOE 5^24,25  studies reported that being present during the resuscitation helped their adjustment to the family member's death.

One LOE 2¹⁰  and 2 LOE 4^12,13  studies observed that allowing family members to be present during a resuscitation in a hospital setting did them no harm, whereas 1 LOE 4²⁶  study suggested that some relatives present for the resuscitation of a family member experienced short-term emotional difficulty.

One LOE 2,¹⁰  1 LOE 3,²⁷  3 LOE 4,^12,23,28  and 3 LOE 5^20,24,29  studies showed that family presence during resuscitation was not perceived as being stressful to staff or to have negatively affected staff performance. However, 1 survey (LOE 4³⁰ ) found that 39% to 66% of emergency medical services (EMS) providers reported feeling threatened by family members during an out-of-hospital resuscitation and that family presence interfered with their ability to perform resuscitations.

In general, family members should be offered the opportunity to be present during the resuscitation of an infant or child. When deciding whether to allow family members to be present during an out-of-hospital resuscitation, the potential negative impact on EMS provider performance must be considered.

How does the presence of a dedicated support person help family members and, potentially, healthcare providers during the resuscitation of an infant or child? What training is appropriate for staff who may serve as support persons for family members during resuscitation of an infant or child? Why is family presence during resuscitation perceived more negatively by out-of-hospital care providers than by in-hospital staff?

Many healthcare providers find it difficult to rapidly and accurately determine the presence or absence of a pulse. On the basis of available evidence, the Task Force decided to deemphasize but not eliminate the pulse check as part of the healthcare provider assessment. The Task Force members recognized that healthcare providers who work in specialized settings may have enhanced skills in accurate and rapid pulse checks, although this has not been studied.

There are considerable data regarding use of end-tidal carbon dioxide (Petco₂) measurement, capnography and capnometry, during cardiopulmonary resuscitation (CPR) as an indicator of CPR quality and as a predictive measure of outcome. Although capnography/capnometry may reflect the quality of CPR, there is insufficient evidence of its reliability in predicting resuscitation success in infants and children.

Thirteen LOE 5 studies^31,–,43  observed that neither laypersons nor healthcare providers are able to perform an accurate pulse check in healthy adults or infants within 10 seconds. In 2 LOE 5 studies in adults^44,45  and 2 LOE 3 studies in children with nonpulsatile circulation,^46,47  blinded healthcare providers commonly assessed pulse status inaccurately and their assessment often took >10 seconds. In the pediatric studies, healthcare professionals were able to accurately detect a pulse by palpation only 80% of the time. They mistakenly perceived a pulse when it was nonexistent 14% to 24% of the time and failed to detect a pulse when present in 21% to 36% of the assessments. The average time to detect an actual pulse was approximately 15 seconds, whereas the average time to confirm the absence of a pulse was 30 seconds. Because the pulseless patients were receiving extracorporeal membrane oxygenation (ECMO) support, one must be cautious in extrapolating these data to the arrest situation; all pulseless patients did have perfusion and therefore had signs of circulation as evidenced by warm skin temperature with brisk capillary refill. All patients evaluated were in an intensive care unit (ICU) setting without ongoing CPR.

Palpation of a pulse (or its absence) is not reliable as the sole determinant of cardiac arrest and need for chest compressions. If the victim is unresponsive, not breathing normally, and there are no signs of life, lay rescuers should begin CPR. In infants and children with no signs of life, healthcare providers should begin CPR unless they can definitely palpate a pulse within 10 seconds.

Is there an association between the time required to successfully detect a suspected cardiac arrest victim's pulse and resuscitation outcome? Is there a difference in outcome when the decision to start chest compressions is based on the absence of signs of life as opposed to absence of a pulse?

In 1 small LOE 4 pediatric case series⁴⁸  cardiac activity was rapidly visualized by echocardiography without prolonged interruption of chest compressions, and this cardiac activity correlated with the presence or absence of a central pulse. In 1 pediatric LOE 4 case report,⁴⁹  echocardiography was useful for diagnosing pericardial tamponade as the cause of cardiac arrest and was useful in guiding treatment.

In 8 LOE 5 adult case series,^50,–,57  echocardiographic findings correlated well with the presence or absence of cardiac activity in cardiac arrest. These reports also suggested that echocardiography may be useful in identifying patients with potentially reversible causes for the arrest.

There is insufficient evidence to recommend for or against the routine use of echocardiography during pediatric cardiac arrest. Echocardiography may be considered to identify potentially treatable causes of an arrest when appropriately skilled personnel are available, but the benefits must be carefully weighed against the known deleterious consequences of interrupting chest compressions.

Can echocardiography be performed during cardiac arrest in infants and children without significant interruptions in chest compressions? How often does echocardiography during cardiac arrest provide information that can affect treatment and outcome?

Three LOE 5 animal studies,^58,–,60  4 LOE 5 adult,^61,–,64  and 1 LOE 5 pediatric series⁶⁵  showed a strong correlation between Petco₂ and interventions that increase cardiac output during resuscitation from shock or cardiac arrest. Similarly 3 LOE 5 animal models^66,–,68  showed that measures that markedly reduce cardiac output result in a fall in Petco₂.

Two LOE 5 adult out-of-hospital studies^69,70  supported continuous Petco₂ monitoring during CPR as a way of determining return of spontaneous circulation (ROSC), particularly if the readings during CPR are >15 mm Hg (2.0 kPa). In 1 LOE 4⁷¹  and 2 LOE 5 adult case series,^72,73  an abrupt and sustained rise in Petco₂ often preceded identification of ROSC.

Two LOE 4 pediatric cases series,^65,74  8 LOE 5 adult,^70,75,–,81  and 1 LOE 5 animal study⁵⁹  showed that a low Petco₂ (<10 mm Hg [1.33 kPa] to <15 mm Hg [2.0 kPa]) despite 15 to 20 minutes of advanced life support (ALS) is strongly associated with failure to achieve ROSC. On the basis of 2 LOE 5 animal studies^71,82  and 2 adult LOE 5 case series,^70,78  Petco₂ after at least 1 minute of CPR may be more predictive of outcome than the initial value because the initial Petco₂ is often increased in patients with asphyxial cardiac arrest.

The wide variation for initial Petco₂ during resuscitation limits its reliability in predicting outcome of resuscitation and its value as a guide to limiting resuscitation efforts. Two LOE 5 animal studies^71,82  and 2 large LOE 5 adult trials^70,78  suggested that the initial Petco₂ is higher if the etiology of the cardiac arrest is asphyxial rather than if it is a primary cardiac arrest.

Interpretation of the end-tidal CO₂ during resuscitation is affected by the quality of the measurement, the minute ventilation delivered during resuscitation, the presence of lung disease that increases anatomic dead space, and the presence of right-to-left shunting.^83,–,85 

In 1 LOE 5 adult study,⁸⁶  sodium bicarbonate transiently increased end-tidal CO₂, and in 3 LOE 5 adult^87,–,89  and 2 LOE 5 animal^90,91  studies, epinephrine (and other systemic vasoconstrictive agents) transiently decreased Petco₂.

Continuous capnography or capnometry monitoring, if available, may be beneficial by providing feedback on the effectiveness of chest compressions. Whereas a specific target number cannot be identified, if the Petco₂ is consistently <15 mm Hg, it is reasonable to focus efforts on improving the quality of chest compressions and avoiding excessive ventilation.

Although a threshold Petco₂ may predict a poor outcome from resuscitation and might be useful as a guide to termination of CPR, there are insufficient data to establish the threshold and the appropriate duration of ALS needed before such evaluation in children. The Petco₂ must be interpreted with caution for 1 to 2 minutes after administration of epinephrine or other vasoconstrictive medications because these medications may decrease the Petco₂.

Does Petco₂ monitoring during CPR improve quality of chest compressions and/or outcome of pediatric resuscitation? During CPR, can Petco₂ be reliably measured via a laryngeal mask airway (LMA)? Is there a threshold Petco₂ that predicts ROSC or low likelihood of ROSC during resuscitation from pediatric cardiac arrest? Can the initial Petco₂ distinguish asphyxial from cardiac etiology of pediatric cardiac arrest? Is detection of ROSC using Petco₂ monitoring more accurate than palpation of a pulse? Are Petco₂ targets during CPR different for subgroups of infants and children with alterations in pulmonary blood flow or high airway resistance?

Opening and maintaining a patent airway and providing ventilations are fundamental elements of pediatric CPR, especially because cardiac arrest often results from, or is complicated by, asphyxia. There are no new data to change the 2005 ILCOR recommendation to use manual airway maneuvers (with or without an oropharyngeal airway) and bag-mask ventilation (BMV) for children requiring airway control or positive-pressure ventilation for short periods in the out-of-hospital setting. When airway control or BMV is not effective, supraglottic airways may be helpful when used by properly trained personnel.

When performing tracheal intubation, data suggest that the routine use of cricoid pressure may not protect against aspiration and may make intubation more difficult.

Routine confirmation of tracheal tube position with capnography/capnometry is recommended with the caveat that the Petco₂ in infants and children in cardiac arrest may be below detection limits for colorimetric devices.

Following ROSC, toxic oxygen byproducts (reactive oxygen species, free radicals) are produced that may damage cell membranes, proteins, and DNA (reperfusion injury). Although there are no clinical studies in children (outside the newborn period) comparing different concentrations of inspired oxygen during and immediately after resuscitation, animal data and data from newborn resuscitation studies suggest that it is prudent to titrate inspired oxygen after return of a perfusing rhythm to prevent hyperoxemia.

There are no studies comparing ventilation of infants and children in cardiac arrest with different inspired oxygen concentrations. Two LOE 5 meta-analyses of several randomized controlled trials comparing neonatal resuscitation initiated with room air versus 100% oxygen^92,93  showed increased survival when resuscitation was initiated with room air.

Seven LOE 5 animal studies^94,–,100  suggested that ventilation with room air or an Fio₂ of <1.0 during cardiac arrest may be associated with less neurologic deficit than ventilation with an Fio₂ of 1.0, whereas 1 LOE 5 animal study¹⁰¹  showed no difference in outcome. In 5 LOE 5 animal studies^{95,97,–,99,102}  ventilation with 100% oxygen during and following resuscitation contributed to free radical–mediated reperfusion injury to the brain.

There is insufficient evidence to recommend any specific inspired oxygen concentration for ventilation during resuscitation from cardiac arrest in infants and children. Once circulation is restored, it is reasonable to titrate inspired oxygen to limit hyperoxemia.

Does the use of any specific concentration of supplementary oxygen during resuscitation from cardiac arrest in infants and children improve or worsen outcome? What is the appropriate target oxygen saturation for the pediatric patient after achieving ROSC?

There are no studies that compare the safety and efficacy of cuffed versus uncuffed tubes in infants and children who require emergency intubation.

Two LOE 5 randomized controlled studies^103,104  and 1 LOE 5 cohort-controlled study¹⁰⁵  in a pediatric anesthesia setting showed that the use of cuffed tracheal tubes was associated with a higher likelihood of selecting the correct tracheal tube size (and hence a lower reintubation rate) with no increased risk of perioperative or airway complications. Cuff pressures in these 3 studies were maintained at <25 cm H₂O. Two perioperative LOE 5 cohort-controlled pediatric studies^105,106  similarly showed that cuffed tubes were not associated with an increased risk of perioperative airway complications.

One LOE 5 pediatric case series¹⁰⁷  observed that the use of cuffed tracheal tubes was not a risk factor for developing subglottic stenosis in patients having corrective surgery for congenital cardiac defects. In the intensive care setting, 2 LOE 5 prospective cohort-controlled studies^108,109  and 1 LOE 5 retrospective cohort-controlled study¹¹⁰  documented no greater risk of complications for children. >8 years of age who were intubated with cuffed compared with uncuffed tracheal tubes.

One small LOE 5 case-controlled study¹¹¹  showed that cuffed tracheal tubes decreased the incidence of aspiration in the PICU, and 1 LOE 5 case series¹⁰⁵  of children with burns undergoing general anesthesia showed a significantly higher rate of excessive air leak requiring immediate reintubation in patients initially intubated with an uncuffed tracheal tube.

Both cuffed and uncuffed tracheal tubes are acceptable for infants and children undergoing emergency intubation. If cuffed tracheal tubes are used, avoid excessive cuff pressures.

What is the best technique to determine cuff pressure and/or the presence of an air leak when using cuffed tracheal tubes in infants and children? What is the optimal cuff or leak pressure for children of different ages? Does optimal cuff pressure vary based on the type of cuffed tube (eg, Microcuff®) used?

Are the data generated in elective operating room studies applicable to emergency resuscitation scenarios? Are there select populations of pediatric patients whose outcomes are improved by the use of cuffed tracheal tubes during resuscitation?

Evidence from 1 LOE 2 prospective randomized trial of elective intubation in a pediatric operating room¹⁰³  was used to support the existing formula for estimation of appropriate cuffed tracheal tube internal diameter (ID): ID (mm)=(age in years/4) + 3, also known as the Khine formula. Detailed analysis of this paper, however, reveals that the aggressive rounding up of age employed by the authors in their calculations commonly resulted in selection of a tube with an ID 0.5 mm larger than the size derived from the formula.

Evidence from 1 LOE 2 prospective randomized multicenter study,¹⁰⁴  1 LOE 2,¹¹²  and 3 LOE 4 prospective observational studies of elective intubation in the pediatric operating room^113,–,115  supported use of 3-mm ID cuffed tracheal tubes for newborns and infants (3.5 kg to 1 year of age) and 3.5-mm ID cuffed tracheal tubes for patients 1 to 2 years of age.

One LOE 2 prospective randomized multicenter study¹⁰⁴  and 3 LOE 4 prospective observational studies of elective intubation in the pediatric operating room^113,–,115  using Microcuff® tracheal tubes support the use of the following formula for cuffed endotracheal tubes in children: ID (mm)=(age/4) + 3.5. One LOE 2 prospective observational study of elective intubation in the pediatric operating room¹¹²  found that formula acceptable but associated with a marginally greater reintubation rate than with the Khine formula (ID [mm]=[age in years/4] + 3).

If a cuffed tracheal tube is used in infants ≥3.5 kg and <1 year of age, it is reasonable to use a tube with an ID of 3.0 mm. If a cuffed tracheal tube is used in children between 1 and 2 years of age, it is reasonable to use a tube with an ID of 3.5 mm.

After the age of 2, it is reasonable to estimate the cuffed tracheal tube size with the formula ID (mm)=(age in years/4) + 3.5. If the tracheal tube meets resistance during insertion, a tube with an ID 0.5 mm smaller should be used. If there is no leak around the tube with the cuff deflated, reintubation with a tube ID 0.5 mm smaller may be beneficial when the patient is stable.

Are the formulas for estimation of tracheal tube size that are used for elective intubation in the operating room setting applicable during resuscitation? Is there an upper age limit for the validity of the formula to estimate tube size? Are length-based formulas more accurate compared with age- or weight-based formulas for estimating tracheal tube size in infants and children?

One LOE 1 study¹¹⁶  compared paramedic out-of-hospital BMV with intubation for children with cardiac arrest, respiratory arrest, or respiratory failure in an EMS system with short transport intervals and found equivalent rates of survival to hospital discharge and neurologic outcome. One LOE 1 systematic review that included this study¹¹⁷  also reached the same conclusion.

One LOE 2 study of pediatric trauma patients¹¹⁸  observed that out-of-hospital intubation is associated with a higher risk of mortality and postdischarge neurologic impairment compared with in-hospital intubation. These findings persisted even after stratification for severity of trauma and head trauma.

In 1 LOE 2 (nonrandomized) prehospital pediatric study,¹¹⁹  if paramedics provided BMV while awaiting the arrival of a physician to intubate the patient, the risk of cardiac arrest and overall mortality was lower than if the patient was intubated by the paramedics. These findings persisted even after adjusting for Glasgow Coma Scale score.

Four LOE 4 studies^120,–,123  showed a significantly greater rate of failed intubations and complications in children compared with adults in the out-of-hospital and emergency department settings. Conversely 1 LOE 3 out-of-hospital study¹²⁴  and 1 LOE 4 out-of-hospital study¹²⁵  failed to demonstrate any difference in intubation failure rates between adults and children.

BMV is recommended over tracheal intubation in infants and children who require ventilatory support in the out-of-hospital setting when transport time is short.

For the experienced airway specialist, does tracheal intubation improve outcomes in comparison with BMV for pediatric resuscitation? Does the use of neuromuscular blocking drugs improve the outcome of children undergoing intubation during resuscitation? What is the minimal initial training and ongoing experience needed to improve success rate and reduce complications of emergent intubation of infants and children?

No studies have directly compared BMV to the use of supraglottic airway devices during pediatric resuscitation other than for the newly born in the delivery room. Nine LOE 5 case reports^126,–,134  demonstrated the effectiveness of supraglottic airway devices, primarily the LMA, for airway rescue of children with airway abnormalities.

One LOE 5 out-of-hospital adult study¹³⁵  supports the use of LMAs by first responders during CPR, but another LOE 5 out-of-hospital adult cardiac arrest study¹³⁶  of EMS personnel providing assisted ventilation by either bag-mask device or LMA failed to show any significant difference in ventilation (Paco₂). Six LOE 5 studies during anesthesia^137,–,142  demonstrated that complication rates with LMAs increase with decreasing patient age and size.

In 2 LOE 5 manikin studies^143,144  trained nonexpert providers successfully delivered positive-pressure ventilation using the LMA. Tracheal intubations resulted in a significant incidence of tube misplacement (esophageal or right mainstem bronchus), a problem not present with the LMA, but time to effective ventilation was shorter and tidal volumes were greater with BMV.

In 2 LOE 5 studies of anesthetized children^145,146  suitably trained ICU and ward nurses placed LMAs with a high success rate, although time to first breath was shorter in the BMV group. In a small number of cases ventilation was achieved with an LMA when it proved impossible with BMV.

BMV remains the preferred technique for emergency ventilation during the initial steps of pediatric resuscitation. In infants and children for whom BMV is unsuccessful, use of the LMA by appropriately trained providers may be considered for either airway rescue or support of ventilation.

Are the data regarding use of supraglottic airways for elective airway management in the operating room applicable to emergency resuscitation scenarios? With an LMA in place, is it necessary to pause chest compressions to provide effective ventilations? Is the combination of an oropharyngeal airway with BMV more or less effective than supraglottic airways?

There are no data to identify the optimal minute ventilation (tidal volume or respiratory rate) for infants or children with an advanced airway during CPR, regardless of arrest etiology.

Three LOE 5 animal studies^147,–,149  showed that ventilation during CPR after VF or asphyxial arrest resulted in improved ROSC, survival, and/or neurologic outcome compared with no positive-pressure breaths.

Evidence from 4 LOE 5 adult studies^150,–,153  showed that excessive ventilation is common during resuscitation from cardiac arrest. In 1 LOE 5 animal study¹⁵⁰  excessive ventilation during resuscitation from cardiac arrest decreased cerebral perfusion pressure, ROSC, and survival compared with lower ventilation rates. One good LOE 5 animal study¹⁴⁹  found that increasing respiratory rate during conditions of reduced cardiac output improved alveolar ventilation but not oxygenation, and it reduced coronary perfusion pressure.

In 1 LOE 5 prospective, randomized adult study¹⁵⁴  constant-flow insufflation with oxygen compared with conventional mechanical ventilation during CPR did not change outcome (ROSC, survival to admission, and survival to ICU discharge). In another LOE 5 adult study,¹⁵⁵  adults with witnessed VF arrest had improved neurologically intact survival with passive oxygen insufflation compared with BMV, whereas there was no difference in survival if the VF arrest was unwitnessed.

Two LOE 5 animal studies showed that ventilation or continuous positive airway pressure (CPAP) with oxygen compared with no ventilation improved arterial blood gases¹⁵⁶  but did not change neurologically intact survival.¹⁵⁷  One good-quality LOE 5 animal study¹⁵⁸  showed that reducing tidal volume by 50% during CPR resulted in less excessive ventilation without affecting ROSC.

Following placement of a secure airway, avoid excess ventilation of infants and children during resuscitation from cardiac arrest, whether asphyxial or due to VF. A reduction in minute ventilation to less than baseline for age is reasonable to provide sufficient ventilation to maintain adequate ventilation-to-perfusion ratio during CPR while avoiding the harmful effects of excessive ventilation. There are insufficient data to identify the optimal tidal volume or respiratory rate.

What is the optimal minute ventilation to achieve ventilation-perfusion matching during pediatric CPR? Is it preferable to reduce tidal volume or respiratory rate to achieve optimal minute ventilation during pediatric CPR? Does hypoventilation (ie, hypercarbia) during resuscitation affect outcome from pediatric cardiac arrest? Does passive oxygen insufflation or CPAP during cardiac arrest in infants and children provide adequate gas exchange or improve outcome from resuscitation?

No single assessment method accurately and consistently confirms tracheal tube position. Three LOE 4 studies^71,159,160  showed that when a perfusing cardiac rhythm is present in infants (>2 kg) and children, detection of exhaled CO₂ using a colorimetric detector or capnometer has a high sensitivity and specificity for confirming endotracheal tube placement. One of these studies⁷¹  included infants and children in cardiac arrest. In the cardiac arrest population the sensitivity of exhaled CO₂ detection was only 85% (ie, false-negatives occurred), whereas the specificity remained at 100%.

One neonatal LOE 5 study¹⁶¹  of delivery room intubation demonstrated that detection of exhaled CO₂ by capnography was 100% sensitive and specific for detecting esophageal intubation and took less time than clinical assessment to identify esophageal intubation. Two additional neonatal LOE 5 studies^162,163  showed that confirmation of tracheal tube position is faster with capnography than with clinical assessment.

Two pediatric LOE 4 studies^164,165  showed that in the presence of a perfusing rhythm, exhaled CO₂ detection or measurement can confirm tracheal tube position accurately during transport, while 2 LOE 5 animal studies^166,167  showed that tracheal tube displacement can be detected more rapidly by CO₂ detection than by pulse oximetry.

One LOE 2 operating room study¹⁶⁸  showed that the esophageal detector device (EDD) is highly sensitive and specific for correct tracheal tube placement in children >20 kg with a perfusing cardiac rhythm; there have been no studies of EDD use in children during cardiac arrest. An LOE 4 operating room (ie, non-arrest) study¹⁶⁹  showed that the EDD performed well but was less accurate in children <20 kg.

Confirmation of tracheal tube position using exhaled CO₂ detection (colorimetric detector or capnography) should be used for intubated infants and children with a perfusing cardiac rhythm in all settings (eg, out-of-hospital, emergency department, ICU, inpatient, operating room).

In infants and children with a perfusing rhythm, it may be beneficial to monitor continuous capnography or frequent intermittent detection of exhaled CO₂ during out-of-hospital and intra-/interhospital transport.

The EDD may be considered for confirmation of tracheal tube placement in children weighing >20 kg when a perfusing rhythm is present.

Which technique for CO₂ detection (colorimetric versus capnography) is more accurate during pediatric resuscitation? For infants and children in cardiac arrest, what is the most reliable way to achieve confirmation of tracheal tube position?

There are no data to show that cricoid pressure prevents aspiration during rapid sequence or emergency tracheal intubation in infants or children. Two LOE 5 studies^170,171  showed that cricoid pressure may reduce gastric inflation in children. One LOE 5 study in children¹⁷²  and 1 LOE 5 study in adult cadavers¹⁷³  demonstrated that esophageal reflux is reduced with cricoid pressure.

In 1 LOE 5 adult systematic review¹⁷⁴  laryngeal manipulation enhanced BMV or intubation in some patients while impeding it in others. One LOE 5 study in anesthetized children¹⁷⁵  showed that cricoid pressure can distort the airway with a force of as low as 5 newtons.

If cricoid pressure is used during emergency intubations in infants and children it should be discontinued if it impedes ventilation or interferes with the speed or ease of intubation.

Can cricoid pressure reduce the incidence of aspiration during emergent intubation of infants or children? How much cricoid pressure should be applied, and what is the best technique to reduce gastric inflation during BMV?

The concept of chest compression-only CPR is appealing because it is easier to teach than conventional CPR, and immediate chest compressions may be beneficial for resuscitation from sudden cardiac arrest caused by VF or pulseless VT. Animal studies showed that conventional CPR, including ventilations and chest compressions, is best for resuscitation from asphyxial cardiac arrest. In a large study of out-of-hospital pediatric cardiac arrest,¹⁷⁶  few children with asphyxial arrest received compression-only CPR and their survival was no better than in children who received no CPR.

To be effective, chest compressions must be deep, but it is difficult to determine the optimal depth in infants and children; should recommended depth be expressed as a fraction of the depth of the chest or an absolute measurement? How can this be made practical and teachable?

Evidence from 1 LOE 2 large out-of-hospital pediatric prospective observational investigation¹⁷⁶  showed that children with cardiac arrest of noncardiac etiology (asphyxial arrest) had a higher 30-day survival with more favorable neurologic outcome if they received standard bystander CPR (chest compressions with rescue breathing) compared with chest compression-only CPR. Standard CPR and chest compression-only CPR were similarly effective and better than no bystander CPR for pediatric cardiac arrest from cardiac causes. Of note, the same study showed that more than 50% of children with out-of-hospital cardiac arrest did not receive any bystander CPR. Compression-only CPR was as ineffective as no CPR in the small number of infants and children with asphyxial arrest who did not receive ventilations.

Two LOE 5 animal studies^148,177  demonstrated improved survival rates and favorable neurologic outcome with standard CPR compared with no CPR. One LOE 5 animal study¹⁷⁸  showed that blood gases deteriorated with compression-only CPR compared with standard CPR in asphyxial arrests.

Data from 1 LOE 5 animal study¹⁷⁷  indicated that compression-only CPR is better than no CPR for asphyxial arrest but not as effective as standard CPR, and 7 LOE 5 clinical observational studies in adults^179,–,184  showed that compression-only CPR can result in successful resuscitation from an asphyxial arrest. Moreover, in 10 LOE 5 animal studies^185,–,194  and 7 LOE 5 adult clinical observational studies^{179,–,184,195}  compression-only bystander CPR was generally as effective as standard 1-rescuer bystander CPR for arrests from presumed cardiac causes.

Rescuers should provide conventional CPR (rescue breathing and chest compressions) for in-hospital and out-of-hospital pediatric cardiac arrests. Lay rescuers who cannot provide rescue breathing should at least perform chest compressions for infants and children in cardiac arrest.

Does teaching compression-only CPR to lay rescuers increase the likelihood that CPR will be performed during out-of-hospital pediatric cardiac arrest?

There are no outcome studies comparing 1- versus 2-hand chest compressions for children in cardiac arrest. Evidence from 1 LOE 5 randomized crossover child manikin study¹⁹⁶  showed that higher chest-compression pressures are generated by healthcare professionals using the 2-hand technique. Two LOE 5 studies^197,198  report no increase in rescuer fatigue comparing 1-hand with 2-hand chest compressions delivered by healthcare providers to a child-sized manikin.

Either a 1- or 2-hand technique can be used for performing chest compressions in children.

Does the use of 1-hand compared with 2-hand chest compressions during pediatric cardiac arrest affect quality of CPR or outcome?

There are no studies that compare the 2-thumb chest compression technique with and without a “circumferential squeeze” in infants.

There are insufficient data for or against the need for a circumferential squeeze of the chest when performing the 2-thumb technique of external chest compression for infants.

Does the addition of a circumferential squeeze to 2-thumb compression in infants provide more effective chest compressions or improve resuscitation outcome?

Evidence from anthropometric measurements in 3 good-quality LOE 5 case series^199,–,201  showed that in children the chest can be compressed to one third of the anterior-posterior chest diameter without causing damage to intrathoracic organs. One LOE 5 mathematical model based on neonatal chest computed tomography scans²⁰²  suggests that one third anterior-posterior chest compression depth is more effective than one fourth compression depth and safer than one half anterior-posterior compression depth.

A good-quality LOE 5¹⁵²  adult study found that chest compressions are often inadequate, and a good-quality LOE 4 pediatric study²⁰⁰  showed that during resuscitation of patients >8 years of age, compressions are often too shallow, especially following rescuer changeover. Evidence from 1 pediatric LOE 4 systematic review of the literature²⁰³  showed that rib fractures are rarely associated with chest compressions.

In infants, rescuers should be taught to compress the chest by at least one third the anterior-posterior dimension or approximately 1½ inches (4 cm). In children, rescuers should be taught to compress the chest by at least one third the anterior-posterior dimension or approximately 2 inches (5 cm).

Can lay rescuers or healthcare providers reliably perform compressions to the recommended depth during pediatric cardiac arrest? Is there harm from compressions that are “too deep” in infants?

The ILCOR Neonatal Task Force continues to recommend a compression-ventilation ratio of 3:1 for resuscitation of the newly born in the delivery room, with a pause for ventilation whether or not the infant has an advanced airway. The Pediatric Task Force reaffirmed its recommendation for a 15:2 compression-ventilation ratio for 2-rescuer infant CPR, with a pause for ventilation in infants without an advanced airway, and continuous compressions without a pause for ventilation for infants with an advanced airway.

No previous recommendations were made for hospitalized newborns cared for in areas other than the delivery area or with primary cardiac rather than asphyxial arrest etiology. For example, consider the case of a 3-week-old infant who has a cardiac arrest following cardiac surgery. In the neonatal intensive care unit such an infant would be resuscitated according to the protocol for the newly born, but if the same infant is in the PICU, resuscitation would be performed according to the infant guidelines. A resolution to this dilemma is suggested on the basis of the arrest etiology and ease of training.

There are insufficient data to identify an optimal compression-ventilation ratio for CPR in infants and children. In 4 LOE 5 manikin studies^204,–,207  examining the feasibility of compression-ventilation ratios of 15:2 and 5:1, lone rescuers could not deliver the desired number of chest compressions per minute at a ratio of 5:1. In 5 LOE 5 studies^208,–,212  using a variety of manikin sizes comparing compression-ventilation ratios of 15:2 with 30:2, a ratio of 30:2 yielded more chest compressions with no, or minimal, increase in rescuer fatigue. One LOE 5 study²¹³  of volunteers recruited in an airport to perform 1-rescuer layperson CPR on an adult-sized manikin observed less “no flow time” with the use of a 30:2 ratio compared with a 15:2 ratio.

One LOE 5 observational human study²¹⁴  comparing resuscitations by firefighters before and after the change from a recommended 15:2 to 30:2 compression-ventilation ratio reported more chest compressions per minute with the 30:2 ratio, but the rate of ROSC was unchanged. Three LOE 5 animal studies^192,215,216  showed that coronary perfusion pressure, a major determinant of success in resuscitation, rapidly declines when chest compressions are interrupted; once compressions are resumed, several chest compressions are needed to restore coronary perfusion pressure to preinterruption levels. Thus, frequent interruptions of chest compressions prolong the duration of low coronary perfusion pressure and flow and reduce the mean coronary perfusion pressure. Three LOE 5 manikin studies^213,217,218  and 3 LOE 5^151,152,219  in- and out-of-hospital adult human studies documented long interruptions in chest compressions during simulated or actual resuscitations. Three LOE 5 adult studies^220,–,222  demonstrated that these interruptions reduced ROSC.

In 5 LOE 5 animal studies^{191,192,194,215–216}  chest compressions without ventilations were sufficient to resuscitate animals with VF-induced cardiac arrest. Conversely in 2 LOE 5 animal studies^223,224  decreasing the frequency of ventilation was detrimental in the first 5 to 10 minutes of resuscitation of VF-induced cardiac arrest.

One LOE 5 mathematical model²²⁵  suggested that the compression-ventilation ratio in children should be lower (more ventilations to compressions) than in adults and should decrease with decreasing weight. Two LOE 5 studies of asphyxial arrest in pigs^148,177  showed that ventilations added to chest compressions improved outcome compared with compressions alone. Thus, ventilations are more important during resuscitation from asphyxia-induced arrest than during resuscitation from VF. But even in asphyxial arrest, fewer ventilations are needed to maintain an adequate ventilation-perfusion ratio in the presence of the low cardiac output (and consequently low pulmonary blood flow) produced by chest compressions.

For ease of teaching and retention, a compression-ventilation ratio of 30:2 is recommended for the lone rescuer performing CPR in infants and children, as is used for adults. For healthcare providers performing 2-rescuer CPR in infants and children, a compression-ventilation ratio of 15:2 is recommended. When a tracheal tube is in place, compressions should not be interrupted for ventilations.

What is the optimal compression-ventilation ratio to improve outcome for neonates, infants, and children in cardiac arrest?

There are insufficient data to identify an optimal compression-ventilation ratio for all infants in the first month of life. One LOE 5 animal study¹⁹²  showed that coronary perfusion pressure declined with interruptions in chest compressions; after each interruption, several chest compressions were required to restore coronary perfusion pressure to preinterruption levels. One LOE 5 adult human study²²¹  and 2 LOE 5 animal studies^215,222  showed that interruptions in chest compression reduced the likelihood of ROSC in VF cardiac arrest.

One LOE 5 1-rescuer manikin study²⁰⁷  showed that more effective ventilation was achieved with a 3:1 ratio than with a 5:1, 10:2, or 15:2 ratio. One LOE 5 mathematical study of cardiovascular physiology²²⁶  suggested that blood flow rates in neonates are best at compression rates of >120/min.

There are insufficient data to recommend an optimal compression-ventilation ratio during CPR for all infants in the first month of life (beyond the delivery room). The limited data available suggest that if the etiology of the arrest is cardiac, a 15:2 ratio (2 rescuers) may be more effective than a 3:1 ratio.

Do healthcare providers perform better CPR if they learn 1 rather than 2 compression-ventilation ratios based on etiology of the arrest (cardiac or asphyxial)?

There is insufficient evidence to determine if an intubated neonate has a better outcome from cardiac arrest using a 3:1 compression-ventilation ratio and interposed ventilations compared with continuous chest compressions without pause for ventilations (asynchronous compressions and ventilations).

Two LOE 5 adult^220,222  and 2 LOE 5 animal^191,192  studies demonstrated that interruptions in chest compressions reduced coronary perfusion pressure, a key determinant of successful resuscitation in adults, and decreased ROSC. There are no equivalent studies evaluating the impact of interrupted chest compressions in asphyxiated neonates or neonatal animal models.

In 1 LOE 5 piglet study²²⁷  of VF arrest, myocardial blood flow increased using simultaneous chest compressions and high–airway pressure ventilations in a 1:1 ratio as compared with conventional CPR at a 5:1 ratio. Another LOE 5 VF piglet study²²⁸  demonstrated equivalent cardiac output but worsened gas exchange using a 1:1 compression-ventilation ratio (ie, simultaneous compressions and ventilations) with high airway pressures compared with conventional CPR at a 5:1 ratio.

One LOE 5¹⁴⁸  study in nonintubated asphyxiated piglets resuscitated with a 5:1 compression-ventilation ratio showed that ventilations are important for successful resuscitation. One LOE 5 study in intubated asphyxiated piglets¹⁷⁸  showed that the addition of ventilations resulted in lower arterial CO₂ tension (Paco₂) without compromising hemodynamics when compared with compressions alone. One LOE 5 manikin study²²⁹  found that healthcare providers were unable to achieve the recommended rate of ventilations during infant CPR at a 3:1 compression-ventilation ratio, with 20% delivering a net rate of 40 breaths per minute after 5 minutes of resuscitation. There are no studies that evaluate the impact of continuous compressions on minute ventilation, gas exchange, or the outcome of resuscitation during CPR for intubated neonates.

For ease of training, providers should use the compression-ventilation ratio and resuscitation approach that is most commonly used in their practice environment for intubated term or near-term newborns within the first month of life. Intubated newborns (ie, those with an advanced airway) who require CPR in non-neonatal settings (eg, prehospital, emergency department, PICU, etc) or those with a cardiac etiology of cardiac arrest, regardless of location, should receive CPR according to infant guidelines (continuous chest compressions without pause for ventilations).

In intubated infants in cardiac arrest, can effective ventilations be performed during continuous chest compressions with asynchronous ventilations? Do pauses for ventilations during CPR affect the outcome from cardiac arrest in intubated infants?

There is no new evidence to change the 2005 ILCOR recommendations on vascular access, including the early use of intraosseous (IO) access and deemphasis of the tracheal route of drug delivery. Epidemiological data, largely from the National Registry of CPR (NRCPR), reported an association between vasopressin, calcium, or sodium bicarbonate administration and an increased likelihood of death. These data, however, cannot be interpreted as a cause-and-effect relationship. The association may be due to more frequent use of these drugs in children who fail to respond to standard basic and advanced life support interventions. These and other data in adults question the benefit of intravenous (IV) medications during resuscitation and reemphasize the importance of high-quality CPR.

There are no studies comparing IO with IV access in children with cardiac arrest. In 1 LOE 5 study of children in shock²³⁰  IO access was frequently more successful and achieved more rapidly than IV access. Eight LOE 4 case series^231,–,238  showed that providers with many levels of training could rapidly establish IO access with minimal complications for children with cardiac arrest.

IO cannulation is an acceptable route of vascular access in infants and children with cardiac arrest. It should be considered early in the care of critically ill children whenever venous access is not readily attainable.

Does the use of IO compared with IV vascular access improve outcome of pediatric cardiac arrest? Does the use of newer IO devices (eg, bone injection guns and drills) compared with conventional IO needles affect outcome in pediatric cardiac arrest?

One LOE 3 study of children with in-hospital cardiac arrest²³⁹  demonstrated similar ROSC and survival rates, whereas 2 LOE 5 studies of adults in cardiac arrest^240,241  demonstrated reduced ROSC and survival to hospital discharge rates when tracheal instead of IV epinephrine was given. One LOE 5 case series of neonatal asphyxial bradycardia²⁴²  demonstrated similar rates of ROSC whether IV or tracheal epinephrine was administered, whereas another LOE 5 study²⁴³  demonstrated a lower rate of ROSC in neonates given tracheal as opposed to IV epinephrine. Many of the human studies used tracheal epinephrine doses of <0.1 mg/kg.

In some animal studies^244,–,249  lower doses of tracheal epinephrine (0.01 to 0.05 mg/kg) produced transient deleterious β-adrenergic vascular effects resulting in lower coronary artery perfusion. One LOE 5 study²⁵⁰  of animals in VF cardiac arrest demonstrated a higher rate of ROSC in those treated with tracheal vasopressin compared with IV placebo.

Four LOE 5 studies of animals in cardiac arrest^251,–,254  demonstrated similar ROSC and survival rates when either tracheal or IV routes were used to deliver epinephrine. These studies also demonstrated that to reach an equivalent biological effect, the tracheal dose must be up to 10 times the IV dose.

The preferred routes of drug delivery for infants and children in cardiac arrest are IV and IO. If epinephrine is administered via a tracheal tube to infants and children (not including the newly born) in cardiac arrest, the recommended dose is 0.1 mg/kg.

What is the optimal dose of tracheal epinephrine during pediatric cardiac arrest?

The Pediatric Task Force evaluated several issues related to defibrillation, including safe and effective energy dosing, stacked versus single shocks, use of automated external defibrillators (AEDs) in infants <1 year of age and paddle/pad type, size, and position. There were a few new human and animal studies on these topics, and the level of evidence (LOE) was generally 3 to 5. No new data are available to support a change in drug treatment of recurrent or refractory VF/pulseless VT. There were several human and animal publications on defibrillation energy dose, but the data are contradictory and the optimal safe and effective energy dose remains unknown.

The new recommendation of an initial dose of 2 to 4 J/kg is based on cohort studies showing low success in termination of VF in children with 2 J/kg. However, these studies do not provide data on success or safety of higher energy doses. The reaffirmation of the recommendation for a single initial shock rather than stacked shocks (first made in 2005) is extrapolated from the ever-increasing adult data showing that long pauses in chest compressions required for stacked shocks are associated with worse resuscitation outcomes and that the initial shock success rate is relatively high with biphasic defibrillation.

No changes are recommended in pad/paddle size or position. Although the safety of AEDs in infants <1 year is unknown, case reports have documented successful defibrillation using AEDs in infants. A manual defibrillator or an AED with pediatric attenuation capabilities is preferred for use in infants and small children.

One LOE 5 study in adults²⁵⁵  demonstrated that shock success increased from 31% to 82% when pad size was increased from 8×8 cm to 12×12 cm. Three pediatric LOE 4,^256,–,258  3 adult LOE 5,^255,259,260  and 3 LOE 5 animal^261,–,263  studies demonstrated that transthoracic impedance decreases with increasing pad size. Decreased transthoracic impedance increases transthoracic current and, thus, presumably, transmyocardial current.

One pediatric LOE 4 study²⁶⁴  observed no difference in the rate of ROSC between antero-lateral and anterior-posterior electrode positions for shock delivery. One pediatric LOE 2 study,²⁵⁶  2 adult LOE 5 studies,^265,266  and 1 LOE 5 animal study²⁶³  demonstrated that transthoracic impedance is not dependent on pad position. Transthoracic impedance was increased in 1 adult LOE 5²⁶⁷  study by placing the pads too close together and in 1 LOE 5²⁶⁰  study when the pads were placed over the female breast. Additionally, 1 adult LOE 5²⁶⁸  study showed that placing the apical pad in a horizontal position lowers transthoracic impedance.

There is insufficient evidence to alter the current recommendations to use the largest size paddles/pads that fit on the infant or child's chest without touching each other or to recommend one paddle/pad position or type over another.

There are limited studies comparing self-adhesive defibrillation pads (SADPs) with paddles in pediatric cardiac arrest. One pediatric LOE 4²⁶⁴  study demonstrated equivalent ROSC rates when paddles or SADPs were used. One LOE 5²⁶⁹  adult out-of-hospital cardiac arrest study suggested improved survival to hospital admission when SADPs rather than paddles were used.

One adult LOE 5²⁷⁰  study showed a lower rate of rhythm conversion, and 1 small adult LOE 5²⁷¹  study showed at least equivalent success with the use of SADPs in comparison with paddles in patients undergoing cardioversion for atrial fibrillation. Two adult LOE 5^272,273  studies showed equivalent transthoracic impedance with SADPs or paddles. One adult LOE 5²⁶⁶  and 2 LOE 5 animal^274,275  studies showed that SADPs had a higher transthoracic impedance than paddles.

One LOE 4²⁷⁶  study described difficulty with fitting self-adhesive pads onto the thorax of a premature infant without the pads touching. One LOE 5²⁷⁷  study demonstrated the improved accuracy of cardiac rhythm monitoring following defibrillation using SADPs compared with the combination of paddles and gel pads.

Using standard resuscitation protocols in simulated clinical environments, 1 LOE 5²⁷⁸  study found no significant difference in the time required to deliver shocks using either SADPs or paddles, and 1 LOE 5²⁷⁹  study found no significant difference in time without compressions when SADPs or paddles were used.

Either self-adhesive defibrillation pads or paddles may be used in infants and children in cardiac arrest.

Is the use of hands-on defibrillation safe for rescuers and does it improve outcome for infants and children in cardiac arrest (eg, by presumably reducing interruptions in chest compressions)?

There are no randomized controlled studies examining a single versus sequential (stacked) shock strategy in children with VF/pulseless VT. Evidence from 7 LOE 5 studies in adults with VF^{221,280,–,285}  supported a single-shock strategy over stacked or sequential shocks because the relative efficacy of a single biphasic shock is high and the delivery of a single shock reduces duration of interruptions in chest compressions.

A single-shock strategy followed by immediate CPR (beginning with chest compressions) is recommended for children with out-of-hospital or in-hospital VF/pulseless VT.

Are there circumstances during which the use of stacked or multiple shocks can improve outcome from pediatric cardiac arrest?

Two LOE 4^264,286  studies reported no relationship between defibrillation dose and survival to hospital discharge or neurologic outcome from VF/pulseless VT. Evidence from 3 LOE 4 studies in children in out-of-hospital and in-hospital settings^264,287,288  observed that an initial dose of 2 J/kg was effective in terminating VF 18% to 50% of the time. Two LOE 4 studies^286,289  reported that children often received more than 2 J/kg during out-of hospital cardiac arrest, with many (69%) requiring ≥3 shocks of escalating energy doses. One in-hospital cardiac arrest LOE 4 study²⁶⁴  reported that the need for multiple shocks with biphasic energy doses of 2.5 to 3.2 J/kg was associated with lack of ROSC.

Evidence from 2 LOE 5 animal studies^290,291  observed that 0% to 8% of episodes of long-duration VF were terminated by a 2 J/kg monophasic shock and up to 32% were terminated by biphasic shocks. Animals in these studies received both fixed and escalated doses, and most required 2 or more shocks to terminate VF. In 1 LOE 5 animal study²⁶³  the defibrillation threshold for short-duration VF was 2.4 J/kg, whereas in another²⁹¹  it was 3.3 J/kg.

In 4 LOE 5 animal studies^{290,292,–,294}  of AED shocks delivered using a pediatric attenuator, 50 J and 50→76→86 J (2.5 to 4 J/ kg) escalating doses were effective at terminating long-duration VF but required multiple shocks. In 1 LOE 5 animal study²⁹⁵  10 J/kg shocks were more effective at terminating long-duration VF (6 minutes) with 1 shock than 4 J/kg shocks.

In 2 LOE 4 pediatric studies^264,286  and 4 LOE 5 animal studies,^{290,292,–,294}  energy doses of 2 to 10 J/kg for short- or long-duration VF resulted in equivalent rates of survival. Myocardial damage, as assessed by hemodynamic or biochemical measurements, was less when a pediatric attenuator was used with an adult energy dose compared with a full adult AED dose, but the degree of myocardial damage was not associated with any difference in 4- or 72-hour survival. An LOE 5 animal study²⁹⁵  found no difference in hemodynamic parameters or biochemical measurements of myocardial damage comparing biphasic 150 J (4 J/kg) with monophasic 360 J/kg (10 J/kg) shocks.

In 2 LOE 5 animal studies^290,291  biphasic waveforms were more effective than monophasic waveforms for treatment of VF/pulseless VT. There are no human data that directly compare monophasic to biphasic waveforms for pediatric defibrillation.

An initial dose of 2 to 4 J/kg is reasonable for pediatric defibrillation. Higher subsequent energy doses may be safe and effective.

What is the minimum effective and maximum safe defibrillation energy dose for pediatric VF/pulseless VT? What is the optimal parameter (eg, weight or length) on which to base defibrillation energy doses for infants and children? Should the energy dose for defibrillation be escalated for shock-refractory VF?

Does the use of biphasic waveforms when compared to monophasic waveforms improve outcome from pediatric cardiac arrest?

In 2 LOE 5 prospective out-of-hospital adult trials IV amiodarone improved ROSC and survival to hospital admission but not hospital discharge when compared with placebo²⁹⁶  or lidocaine²⁹⁷  for treatment of shock-refractory VF/pulseless VT. Evidence from 2 LOE 5 case series in children^298,299  supported the effectiveness of amiodarone for the treatment and acute conversion of life-threatening (nonarrest) ventricular arrhythmias. There are no pediatric data investigating the efficacy of lidocaine for shock refractory VF/ pulseless VT.

Amiodarone may be used for the treatment of shock-refractory or recurrent VF/pulseless VT in infants and children; if amiodarone is not available, lidocaine may be considered.

Does the use of amiodarone compared with lidocaine improve outcome from shock-refractory or recurrent VF/pulseless VT in infants and children? Is lidocaine effective for the treatment of VF/pulseless VT in children?

One LOE 4³⁰⁰  and 2 LOE 5^288,301  studies showed that infants in cardiac arrest (in- and out-of-hospital) may have shockable rhythms. Evidence from 3 LOE 5^302,–,304  studies showed that many AED devices can safely and accurately distinguish between a shockable and nonshockable rhythm in infants and children.

The optimal energy dose for defibrillation in infants has not been established, but indirect data from 5 LOE 5 animal studies^{287,294,305,–,307}  showed that the young myocardium may be able to tolerate high-energy doses. In 3 LOE 5 animal studies a pediatric attenuator used with an adult-dose biphasic AED shock was as effective and less harmful than monophasic weight-based doses²⁹⁰  or biphasic adult doses.^292,293 

Two LOE 4 case reports^308,309  described survival of infants with out-of-hospital cardiac arrest when AED use was coupled with bystander CPR and defibrillation using an AED. Two pediatric LOE 5 case reports^310,311  noted successful defibrillation with minimal myocardial damage and good neurologic outcome using an AED with adult energy doses.

For treatment of out-of-hospital VF/pulseless VT in infants, the recommended method of shock delivery by device is listed in order of preference below. If there is any delay in the availability of the preferred device, the device that is available should be used. The AED algorithm should have demonstrated high specificity and sensitivity for detecting shockable rhythms in infants. The order of preference is as follows:

1. Manual defibrillator

2. AED with dose attenuator

3. AED without dose attenuator

Is there a lower limit of infant size or weight below which an AED should not be used?

The evidence on emergency treatment of arrhythmias was reviewed and the only change was the addition of procainamide as possible therapy for refractory supraventricular tachycardia (SVT).

There is insufficient evidence to support or refute the efficacy of electric therapy over drug therapy or the superiority of any drug for the emergency treatment of unstable VT in the pediatric age group. In 2 LOE 5 adult case series,^312,313  early electric cardioversion was effective for treatment of unstable VT.

In 4 small LOE 4 pediatric case series^{298,299,314,315}  amiodarone was effective in the management of VT. One prospective randomized multicenter safety and efficacy LOE 2 trial evaluating amiodarone for the treatment of pediatric tachyarrhythmias³¹⁶  found that 71% of children treated with amiodarone experienced cardiovascular side effects. Both efficacy and adverse events were dose-related.

It is reasonable to use synchronized electric cardioversion as the preferred first therapy for pediatric VT with hypotension or evidence of poor perfusion. If drug therapy is used to treat unstable VT, amiodarone may be a reasonable choice, with careful hemodynamic monitoring performed during its slow delivery.

What is the optimal dose of energy for synchronized cardioversion during treatment of unstable VT in pediatric patients?

Twenty-two LOE 4 studies in infants and children^317,–,338  demonstrated the effectiveness of adenosine for the treatment of hemodynamically stable or unstable SVT. One LOE 4 study³³⁹  and 4 larger LOE 5 studies involving teenagers and adults^340,–,343  also demonstrated the efficacy of adenosine, although frequent but transient side effects were reported.

One LOE 2 study³⁴⁴  showed highly successful (approximately 90%) treatment of SVT in infants and children using adenosine or verapamil and superiority of these drugs to digitalis (61% to 71%). One LOE 5 randomized prospective adult study³⁴⁵  and 1 LOE 5 meta-analysis, primarily involving adults but including some children,³⁴⁶  demonstrated the effectiveness of verapamil and adenosine in treating SVT and highlighted the cost-effectiveness of verapamil over adenosine.

One LOE 4 randomized, prospective study³¹⁶  and 15 LOE 4 small case series and observational studies in infants and children^{298,299,314,315,347,–,357}  showed that amiodarone was effective in the treatment of supraventricular tachyarrhythmias. Generalization to pediatric SVT treatment with amiodarone may be limited, however, since most of these studies in children involved postoperative junctional tachycardia.

Rare but significant side effects have been reported in association with rapid administration of amiodarone. Bradycardia and hypotension were reported in 1 prospective LOE 4 study,³¹⁶  cardiovascular collapse was reported in 2 LOE 5 case reports,^358,359  and polymorphic VT was reported in 1 small LOE 4 case series.³⁶⁰  Other LOE 5 case reports describe late side effects of pulmonary toxicity³⁵⁹  and hypothyroidism.³⁶² 

In 1 LOE 2 pediatric comparison control study³⁶³  procainamide had a significantly higher success rate and an equal incidence of adverse effects when compared with amiodarone for treating refractory SVT. In 5 LOE 4 observational studies^364,–,368  and 5 LOE 5 case reports^369,–,373  procainamide effectively suppressed or slowed the rate in children with SVT. A wide variety of arrhythmias were studied, including ectopic atrial tachycardia, atrial flutter, and orthodromic reciprocating tachycardia.

In LOE 5 studies in children,³⁷⁴  adults,^375,376  and animals,³⁷⁷  hypotension from procainamide infusion resulted from vasodilation and not decreased myocardial contractility. Initial observational LOE 4 reports^378,–,380  and 1 LOE 4 case series³⁸¹  described successful treatment of pediatric SVT with verapamil. However, multiple small LOE 4 case series^344,382  and LOE 5 case reports^383,384  documented severe hypotension, bradycardia, and heart block causing hemodynamic collapse and death following IV administration of verapamil for SVT in infants. Two small LOE 4 pediatric case series^385,386  described esmolol and dexmedetomidine in the treatment of SVT.

For infants and children with SVT with a palpable pulse, adenosine should be considered the preferred medication. Verapamil may be considered as alternative therapy in older children but should not be routinely used in infants. Procainamide or amiodarone given by a slow IV infusion with careful hemodynamic monitoring may be considered for refractory SVT.

Does the use of alternate medications (eg, esmolol, dexmedetomidine) in the treatment of SVT in infants and children improve outcome? What is the role of vagal maneuvers in the treatment of SVT?

The Task Force reviewed evidence related to several key questions about the management of shock in children. There is ongoing uncertainty about the indications for using colloid versus crystalloid in shock resuscitation. One large adult trial suggested that normal saline (isotonic crystalloid) is equivalent to albumin, although subgroup analysis suggested harm associated with the use of colloid in patients with traumatic brain injury. There were insufficient data to change the 2005 recommendations.

The optimal timing for intubation of children in shock remains unclear, although reports of children and adults with septic shock suggested potential beneficial effects of early intubation (before signs of respiratory failure develop) combined with a protocol-driven management approach. When children in septic shock were treated with a protocol that included therapy directed to normalizing central venous oxygen saturation, patient outcome appeared to improve.

Performing rapid sequence induction and tracheal intubation of a child with shock can cause acute cardiovascular collapse. Etomidate typically causes less hemodynamic compromise than other induction drugs and is therefore often used in this setting. However, data suggest that the use of this drug in children and adults with septic shock is associated with increased mortality that may be secondary to etomidate's inhibitory effects on corticosteroid synthesis. Administering stress-dose corticosteroids in septic shock remains controversial, with recent adult trials failing to show a beneficial effect.

There are no pediatric studies of the timing or extent of volume resuscitation in hemorrhagic shock with hypotension. Nine LOE 5 adult^387,–,395  studies reported conflicting results with regard to the effect of timing and extent of volume resuscitation on outcome of hemorrhagic shock with hypotension.

There is insufficient evidence as to the best timing or quantity for volume resuscitation in infants and children with hemorrhagic shock following trauma.

What is the appropriate clinical indicator for volume resuscitation during treatment of hemorrhagic shock in infants and children?

There are no studies investigating the role of intubation and assisted ventilation before the onset of respiratory failure in infants and children with shock. Two LOE 5 animal studies in septic shock^396,397  and 1 LOE 5 animal study in pericardial tamponade³⁹⁸  showed improved hemodynamics and select organ perfusion with intubation before the onset of respiratory failure. One report of 2 adult patients³⁹⁹  (LOE 5) described cardiac arrest following intubation of 1 adult patient with tamponade due to penetrating trauma and improvement in hemodynamics during spontaneous breathing in 1 mechanically ventilated adult patient with post–cardiac surgery tamponade.

One LOE 5 study of septic shock in adults⁴⁰⁰  suggested a reduced mortality with early ventilation compared with historic controls who only received ventilation for respiratory failure. One LOE 5 study of animals in septic shock⁴⁰¹  showed that early assisted ventilation does not reduce oxygen extraction or prevent the development of lactic acidosis.

There is insufficient evidence to support or refute the use of endotracheal intubation of infants and children in shock before the onset of respiratory failure.

Does the timing of respiratory support in infants and children with shock affect outcome?

Evidence from 3 randomized blinded LOE 1 controlled trials in children with dengue shock syndrome^402,–,404  and 1 LOE 1 open randomized trial in children with septic shock⁴⁰⁵  suggested no clinically important differences in survival from therapy with colloid versus therapy with isotonic crystalloid solutions for shock.

In 1 large LOE 5 randomized controlled trial of fluid therapy in adult ICU patients⁴⁰⁶  and in 6 good-quality LOE 5 meta-analyses, predominantly of adults,^407,–,412  no mortality differences were noted when colloid was compared with hypertonic and isotonic crystalloid solutions, and no differences were noted between types of colloid solutions.

Three LOE 5 studies comparing the use of crystalloids and colloids for adults in shock suggested that crystalloid may have an associated survival benefit over colloid in subgroups of patients with shock, including general trauma,⁴⁰⁹  traumatic brain injury,⁴¹³  and burns.⁴¹⁴  One randomized controlled LOE 5 study of children with severe malaria suggested better survival with colloid than with crystalloid infusion.⁴¹⁵ 

Isotonic crystalloids are recommended as the initial resuscitation fluid for infants and children with any type of shock. There is insufficient evidence to identify the superiority of any specific isotonic crystalloid over others.

Does the use of any specific crystalloid solution (Ringer's lactate, normal saline, hypertonic saline) improve outcome for pediatric shock? Are there subgroups of children in shock whose outcome is improved with the use of colloid compared with crystalloid?

One LOE 4 observational study⁴¹⁶  suggested that the course of pediatric septic shock physiology is dynamic and that serial assessments are required to titrate the type and dose of inotropes or vasopressor therapy to achieve optimal hemodynamic results. Evidence from 4 LOE 1 pediatric randomized controlled studies,^417,–,420  3 LOE 5 adult randomized controlled studies,^421,–,423  and 1 LOE 5 adult systematic review⁴²⁴  showed that no inotrope or vasopressor is superior in reducing mortality from pediatric or adult distributive shock.

Two LOE 1 pediatric randomized controlled studies^417,418  showed that children with “cold” (ie, low cardiac index) septic shock improved hemodynamically with brief (4-hour) administration of milrinone (bolus and infusion). One LOE 1 pediatric randomized controlled study⁴²⁰  of vasodilatory shock compared the addition of vasopressin versus placebo to standard vasoactive agents and showed no change in duration of vasopressor infusion but observed a trend toward increased mortality.

Eleven small LOE 4 pediatric case series^425,–,435  showed improved hemodynamics but not survival when vasopressin or terlipressin was administered to children with refractory, vasodilatory, septic shock.

There is insufficient evidence to recommend a specific inotrope or vasopressor to improve mortality in pediatric distributive shock. Selection of an inotrope or vasopressor to improve hemodynamics should be tailored to each patient's physiology and adjusted as clinical status changes.

Does the use of any specific vasoactive agent improve outcome for infants and children with distributive shock?

One LOE 4 pediatric case series⁴³⁶  showed that critically ill children requiring inotropic support have wide variability in hemodynamic responses to different infusion rates of dobutamine. One LOE 2 blinded crossover study⁴³⁷  found dopamine and dobutamine had equal hemodynamic effects in infants and children requiring post–cardiac surgical inotropic support but that dopamine at an infusion rate of >7 mcg/kg per minute increased pulmonary vascular resistance.

Six LOE 3 studies^438,–,443  showed that both dopamine and dobutamine infusions improve hemodynamics in children with cardiogenic shock.

Evidence from 1 LOE 1 pediatric placebo-controlled trial⁴⁴⁴  showed that milrinone is effective in preventing low cardiac output syndrome in infants and children following biventricular cardiac repair. One LOE 4 study⁴⁴⁵  showed that milrinone improved cardiac index in neonates with low cardiac output following cardiac surgery.

One small LOE 1 study⁴⁴⁶  showed that children had better hemodynamic parameters and shorter ICU stays if they received milrinone compared with low-dose epinephrine plus nitroglycerin for inotropic support following repair of tetralogy of Fallot.

In 2 LOE 4 small case series,^447,448  when children with heart failure secondary to myocardial dysfunction were given levosimendan, they demonstrated improved ejection fraction, required a shorter duration of catecholamine infusions,⁴⁴⁷  and showed a trend toward improved hemodynamics and reduced arterial lactate levels.⁴⁴⁸ 

In subgroup analysis from 1 LOE 5 randomized controlled trial in adults,⁴⁴⁹  patients with cardiogenic shock treated with norepinephrine versus dopamine had an improved survival at 28 days. When all causes of shock were included, patients treated with norepinephrine also had fewer arrhythmias than those treated with dopamine (12% versus 24%).

The catecholamine dose for inotropic support in cardiogenic shock must be individually titrated because there is a wide variability in clinical response. It is reasonable to use epinephrine, levosimendan, dopamine, or dobutamine for inotropic support in infants and children with cardiogenic shock. Milrinone may be beneficial for the prevention and treatment of low cardiac output following cardiac surgery.

There are insufficient data to support or refute the use of norepinephrine in pediatric cardiogenic shock.

Does the use of any specific vasoactive agent improve outcome for infants and children with cardiogenic shock who have not undergone cardiac surgery?

One LOE 4 study of children with septic shock⁴⁵⁰  showed that adrenal suppression occurred after the administration of a single dose of etomidate and persisted for at least 24 hours. Evidence from 2 LOE 4^451,452  studies and 1 LOE 5⁴⁵³  study showed that etomidate can be used to facilitate tracheal intubation in infants and children with minimal hemodynamic effect, but very few of these reports included patients with hypotensive septic shock. One LOE 4 study⁴⁵⁰  suggested an association with mortality when etomidate is used to facilitate the intubation of children with septic shock.

One adult LOE 5 study⁴⁵⁴  observed an increased mortality associated with the use of etomidate for intubation of patients in septic shock, even with steroid supplementation. Conversely, 1 underpowered adult LOE 5 study⁴⁵⁵  in septic patients did not show an increase in mortality.

One multicenter adult LOE 5 comparative trial of etomidate versus ketamine for intubation⁴⁵⁶  found no difference in organ failure over the first 72 hours and no mortality difference, but this study included only a small number of patients with shock. Adrenal insufficiency was more common in etomidate-treated patients.

Etomidate should not be routinely used when intubating an infant or child with septic shock. If etomidate is used in infants and children with septic shock, the increased risk of adrenal insufficiency should be recognized.

If etomidate is used, does steroid administration improve outcome for infants and children with septic shock?

In 6 LOE 5 randomized controlled trials in adults with septic shock^{454,457,–,461}  the addition of low-dose hydrocortisone decreased the time to shock reversal. Three LOE 5 randomized controlled trials in adults with vasopressor-dependent septic shock^457,462,463  showed that survival was improved when low-dose hydrocortisone was administered, while 1 small adult LOE 5 randomized controlled trial⁴⁶⁴  showed a trend toward increased survival.

One fair-quality, small LOE 1 study in children with septic shock⁴⁶⁵  found that low-dose hydrocortisone administration resulted in no survival benefit. One fair-quality LOE 1 study of administration of low-dose hydrocortisone to children with septic shock⁴⁶⁶  demonstrated earlier shock reversal. Data from 1 LOE 4 hospital discharge database⁴⁶⁷  noted the association between the use of steroids in children with severe sepsis and decreased survival.

In 1 LOE 5 study in adults with septic shock⁴⁵⁷  survival improved significantly with the use of low-dose hydrocortisone and fludrocortisone compared with placebo. Conversely 4 LOE 5 adult trials in septic shock^{454,459,–,461}  showed no survival benefit with low-dose corticosteroid therapy. In 1 large LOE 5 randomized controlled trial of adults in septic shock,⁴⁵⁴  corticosteroid administration was associated with an increased risk of secondary infection.

There is insufficient evidence to support or refute the routine use of stress-dose or low-dose hydrocortisone and/or other corticosteroids in infants and children with septic shock. Stress-dose corticosteroids may be considered in children with septic shock unresponsive to fluids and requiring vasoactive support.

What is the appropriate “stress dose” of hydrocortisone for hypotensive septic shock? Should the dose of hydrocortisone be titrated to the degree of shock? Should an adrenocorticotrophin (ACTH) stimulation test be performed to determine if an infant or child in septic shock has adrenal insufficiency?

In 1 LOE 1 randomized controlled trial in children with severe sepsis or fluid-refractory septic shock,⁴⁶⁸  protocol-driven therapy that included targeting a superior vena caval oxygen saturation >70%, coupled with treating clinical signs of shock (prolonged capillary refill, reduced urine output, and reduced blood pressure), improved patient survival to hospital discharge in comparison to treatment guided by assessment of clinical signs alone.

Two LOE 5 studies of adults with septic shock, one a randomized controlled trial⁴⁶⁹  and the other a cohort study,⁴⁷⁰  documented improved survival to hospital discharge following implementation of protocol-driven early goal-directed therapy, including titration to a central venous oxygen saturation (Svco₂) ≥70%. In 1 large multicenter LOE 5 adult study⁴⁷¹  evaluating the “Surviving Sepsis” bundle, early goal-directed therapy to achieve an Svco₂ ≥70% was not associated with an improvement in survival, but venous oxygen saturations were measured in <25% of participants.

There are insufficient data on the utility of other diagnostic tests (eg, pH, lactate) to help guide the management of infants and children with shock.

A protocol-driven therapy, which includes titration to a superior vena caval oxygen saturation ≥70%, may be beneficial for infants and children (without cyanotic congenital heart disease) with fluid-refractory septic shock. No treatment recommendations can be made to target Svco₂ saturation in the management of fluid-refractory septic shock in pediatric patients with cyanotic congenital heart disease or for other forms of pediatric shock.

What is the optimal diagnostic test (ie, lactate, Svco₂) to guide management of pediatric shock? Does continuous versus intermittent Svco₂ monitoring affect outcome?

The Task Force reviewed and updated evidence to support medications used during cardiac arrest and bradycardia, but no new recommendations were made. It was again emphasized that calcium and sodium bicarbonate should not be routinely used in pediatric cardiac arrest (ie, should not be used without specific indications).

Eight LOE 5 studies^472,–,479  concluded that length-based methods are more accurate than age-based or observer (parent or provider) estimate-based methods in the prediction of body weight. Four LOE 5 studies^{472,474,480,481}  suggested that the addition of a category of body habitus to length may improve prediction of body weight.

Six LOE 5 studies^482,–,487  attempted to find a formula based on drug pharmacokinetics and physiology that would allow the calculation of a pediatric dose from the adult dose.

In nonobese pediatric patients, initial resuscitation drug doses should be based on actual body weight (which closely approximates ideal body weight). If necessary, body weight can be estimated from body length.

In obese patients the initial doses of resuscitation drugs should be based on ideal body weight that can be estimated from length. Administration of drug doses based on actual body weight in obese patients may result in drug toxicity.

Subsequent doses of resuscitation drugs in both nonobese and obese patients should take into account observed clinical effects and toxicities. It is reasonable to titrate the dose to the desired therapeutic effect, but it should not exceed the adult dose.

What is the most accurate method for calculating resuscitation drug doses for children? Does the accuracy of the estimated weight used to calculate drug dose affect patient outcome? Do specific resuscitation drugs require different adjustments for estimated weight, maturity and/or body composition?

Are formulas for scaling drug doses with formulas from adult doses superior to existing weight-based methods?

No studies have compared epinephrine versus placebo administration for pulseless cardiac arrest in infants and children. One LOE 5 randomized controlled adult study⁴⁸⁸  of standard drug therapy compared with no drug therapy during out-of-hospital cardiac arrest showed improved survival to hospital admission with any drug delivery but no difference in survival to hospital discharge.

Evidence from 1 LOE 1 prospective, randomized, controlled trial,⁴⁸⁹  2 LOE 2 prospective trials,^490,491  and 2 LOE 2 case series with concurrent controls^492,493  showed no increase in survival to hospital discharge or improved neurologic outcome when epinephrine doses of >10 mcg/kg IV were used in out-of-hospital or in-hospital pediatric cardiac arrest. In 1 LOE 1 prospective trial⁴⁸⁹  of pediatric in-hospital cardiac arrest comparing high-dose (100 mcg/kg) with standard-dose epinephrine administered if cardiac arrest persisted after 1 standard dose of epinephrine, 24-hour survival was reduced in the high-dose epinephrine group.

Evidence extrapolated from adult prehospital or in-hospital studies, including 9 LOE 1 randomized trials,^494,–,502  3 LOE 2 trials,^503,–,505  and 3 LOE 3 studies,^506,–,508  showed no improvement in survival to hospital discharge or neurologic outcome when doses >1 mg of epinephrine were given.

In infants and children with out-of-hospital or in-hospital cardiac arrest, the appropriate dose of IV epinephrine is 10 mcg/kg per dose (0.01 mg/kg) for the first and for subsequent doses. The maximum single dose is 1 mg.

Does epinephrine administration improve outcome from cardiac arrest in infants and children? Are there specific patients or arrest types (eg, prolonged arrest, asphyxial arrest, VF arrest) for which epinephrine is more effective?

There are no randomized controlled studies in infants and children examining the use of sodium bicarbonate as part of the management of pediatric cardiac arrest. One LOE 2 multicenter retrospective in-hospital pediatric study⁵⁰⁹  found that sodium bicarbonate administered during cardiac arrest was associated with decreased survival, even after controlling for age, gender, and first documented cardiac rhythm.

Two LOE 5 randomized controlled studies have examined the value of sodium bicarbonate in the management of arrest in other populations: 1 adult out-of-hospital cardiac arrest study⁵¹⁰  and 1 study in neonates with respiratory arrest in the delivery room.⁵¹¹  Both failed to show an improvement in overall survival.

Routine administration of sodium bicarbonate is not recommended in the management of pediatric cardiac arrest.

Are there circumstances under which sodium bicarbonate administration improves outcome from pediatric cardiac arrest?

In 1 pediatric LOE 3 study⁵¹²  vasopressin was associated with lower ROSC and a trend toward lower 24-hour and discharge survival. In 3 pediatric LOE 4^513,–,515  and 2 adult LOE 5^516,517  case series/reports (9 patients) vasopressin⁵¹³  or its long-acting analogue, terlipressin,^514,515  administration was associated with ROSC in patients with refractory cardiac arrest (ie, standard therapy failed).

Extrapolated evidence from 6 LOE 5 adult studies^518,–,523  and 1 LOE 1 adult meta-analysis⁵²⁴  showed that vasopressin used either by itself or in combination with epinephrine during cardiac arrest does not improve ROSC, hospital discharge, or neurologic outcome. Evidence from 1 LOE 5 animal study⁵²⁵  of an infant asphyxial arrest model showed no difference in ROSC when terlipressin was administered alone or in combination with epinephrine as compared with epinephrine alone.

There is insufficient evidence for or against the administration of vasopressin or its long-acting analogue, terlipressin, in pediatric cardiac arrest.

Are there patient subgroups who might benefit from vasopressin (with or without other vasopressors) for pediatric cardiac arrest? Does the use of “early” versus “late” (ie, rescue) vasopressin affect outcome in pediatric cardiac arrest? Is vasopressin effective when administered via a tracheal tube?

Evidence from 3 LOE 2^509,526,527  studies in children and 5 LOE 5 adult studies^528,–,532  failed to document an improvement in survival to hospital admission, hospital discharge, or favorable neurologic outcome when calcium was administered during cardiopulmonary arrest in the absence of documented hypocalcemia, calcium channel blocker overdose, hypermagnesemia, or hyperkalemia. Four LOE 5 animal studies^533,–,536  showed no improvement in ROSC when calcium, compared with epinephrine or placebo, was administered during cardiopulmonary arrest.

Two studies investigating calcium for in-hospital pediatric cardiac arrest suggested a potential for harm. One LOE 2 study examining data from the NRCPR⁵²⁶  observed an adjusted odds ratio of survival to hospital discharge of 0.6 in children who received calcium, and 1 LOE 3 multicenter study⁵⁰⁹  showed an odds ratio for increased hospital mortality of 2.24 associated with the use of calcium. One LOE 2 study of cardiac arrest in the PICU setting⁵²⁷  suggested a potential for harm with the administration of calcium during cardiac arrest; the administration of 1 or more boluses was an independent predictor of hospital mortality.

Routine use of calcium for infants and children with cardiopulmonary arrest is not recommended in the absence of hypocalcemia, calcium channel blocker overdose, hypermagnesemia, or hyperkalemia.

Are there indications for calcium administration that may be associated with improved outcome from pediatric cardiac arrest? Does the increased mortality risk associated with calcium administration reflect harm from calcium or does it simply identify patients who failed to respond to other ALS interventions and therefore were at a higher risk of death?

Evidence from 1 LOE 3 study of in-hospital pediatric cardiac arrest⁵³⁷  observed an improved odds of survival to discharge for those patients who received atropine based on multivariate analysis, whereas the use of epinephrine was associated with decreased odds of survival. Another large LOE 3 study⁵³⁸  demonstrated no association between atropine administration and survival.

In 1 LOE 5 adult case series,⁵³⁹  6 of 8 patients in cardiac arrest who did not respond to epinephrine did respond to atropine with a change to a perfusing rhythm; 3 survived to hospital discharge. An LOE 5 retrospective adult review⁵⁴⁰  observed that a small number of asystolic patients who failed to respond to epinephrine did respond to atropine, but none survived to hospital discharge.

Four LOE 5 adult studies^541,–,544  showed a benefit of atropine in vagally mediated bradycardia. One small LOE 4 pediatric case series⁵⁴⁵  showed that atropine is more effective than epinephrine in increasing heart rate and blood pressure in children with post–cardiac surgical hypotension and bradycardia (Bezold-Jarisch reflex mediated bradycardia).

Four LOE 5 adult^{542,546,–,548}  and 4 LOE 5 animal^549,–,552  studies showed no benefit from atropine used to treat bradycardia or cardiac arrest. One LOE 5 animal study⁵⁵³  did show a benefit of atropine when used with epinephrine in cardiac arrest.

Epinephrine may be used for infants and children with bradycardia and poor perfusion that is unresponsive to ventilation and oxygenation. It is reasonable to administer atropine for bradycardia caused by increased vagal tone or cholinergic drug toxicity. There is insufficient evidence to support or refute the routine use of atropine for pediatric cardiac arrest.

What is the optimal dose of epinephrine for pediatric bradycardia? Is there a role for titrated doses? Does the use of epinephrine versus atropine improve outcome from pediatric bradycardia? Are there circumstances under which atropine administration improves outcome from pediatric cardiac arrest?

There is increasing evidence that extracorporeal cardiac life support (ECLS) can act as a bridge to maintain oxygenation and circulation in selected infants and children with cardiac arrest if they are transplant candidates or have a self-limited or treatable illness. When ECLS is initiated for the treatment of cardiac arrest, it is referred to as ECPR (extracorporeal CPR). ECPR can only be employed if the cardiac arrest occurs in a monitored environment with protocols and personnel for rapid initiation.

One LOE 2⁵⁵⁴  and 26 LOE 4 studies^555,–,580  reported favorable early outcome after ECPR in children with primary cardiac disease who were located in an ICU or other highly supervised environment using ECPR protocols at the time of the arrest.

One LOE 2⁵⁵⁴  and 2 LOE 4^555,564  studies indicated poor outcome from ECPR in children with noncardiac diseases.

In 1 LOE 4 study⁵⁵⁶  survival following ECPR in children was associated with shorter time interval between arrest and ECPR team activation and shorter CPR duration. Two LOE 4 studies^560,581  found insignificant improvements in outcome after ECPR in children following protocol changes leading to shorter durations of CPR. One LOE 2⁵⁵⁴  and 3 LOE 4^555,559,565  studies found no relationship between CPR duration and outcome after ECPR in children.

Three small LOE 4 studies,^582,–,584  including a total of 21 children, showed favorable outcome with ECPR following out-of-hospital cardiac arrest associated with environmentally induced severe hypothermia (temperature <30°C).

ECPR may be beneficial for infants and children with cardiac arrest if they have heart disease amenable to recovery or transplantation and the arrest occurs in a highly supervised environment such as an ICU with existing clinical protocols and available expertise and equipment to rapidly initiate ECPR. There is insufficient evidence for any specific threshold for CPR duration beyond which survival with ECPR is unlikely. ECPR may be considered in cases of environmentally induced severe hypothermia (temperature <30°C) for pediatric patients with out-of-hospital cardiac arrest if the appropriate expertise, equipment, and clinical protocols are in place.

What are the long-term neurologic outcomes of pediatric patients treated with ECPR? Is there an upper limit for the duration of standard CPR beyond which using ECPR will be of no benefit?

The Task Force reviewed evidence regarding hypothermia for pediatric patients who remain comatose following resuscitation from cardiac arrest. There is clear benefit for adult patients who remain comatose after VF arrest, but there is little evidence regarding effectiveness for infants (ie, beyond the neonatal period) and young children who most commonly have asphyxial arrest.

Some patients with sudden death without an obvious cause have a genetic abnormality of myocardial ion channels (ie, a channelopathy), which presumably leads to a fatal arrhythmia. Because this is an inherited abnormality, family members might be affected, but special tests are required for the detection of this inherited genetic defect.

There are no randomized pediatric studies on induced therapeutic hypothermia following cardiac arrest.

Two prospective randomized LOE 5 studies of adults with VF arrest^585,586  and 2 prospective randomized LOE 5 studies of newborns with birth asphyxia^587,588  showed that therapeutic hypothermia (32° to 34°C) up to 72 hours after resuscitation has an acceptable safety profile and may be associated with better long-term neurologic outcome.

One LOE 2 observational study⁵⁸⁹  neither supports nor refutes the use of therapeutic hypothermia after resuscitation from pediatric cardiac arrest. However, patients in this study were not randomized, and the cooled patients were much sicker and younger than those not cooled.

Therapeutic hypothermia (to 32°C to 34°C) may be beneficial for adolescents who remain comatose following resuscitation from sudden witnessed out-of-hospital VF cardiac arrest. Therapeutic hypothermia (to 32°C to 34°C) may be considered for infants and children who remain comatose following resuscitation from cardiac arrest.

Does therapeutic hypothermia improve outcome following pediatric cardiac arrest? Is there a difference in effectiveness for VF arrest versus asphyxial arrest? What is the optimal protocol for cooling after pediatric cardiac arrest (timing, duration, goal temperature, rate of rewarming)?

There are no studies evaluating the role of vasoactive medications after ROSC in children. Evidence from 2 LOE 3 studies in children,^590,591  2 LOE 5 studies in adults,^592,593  and 2 LOE 5 animal studies^594,595  documented that myocardial dysfunction and vascular instability are common following resuscitation from cardiac arrest.

Evidence from 6 LOE 5 animal studies^594,–,599  documented hemodynamic improvement when vasoactive medications (dobutamine, milrinone, levosimendan) were given in the post–cardiac arrest period. Evidence from 1 large LOE 5 pediatric⁴⁴⁴  and 4 LOE 5 adult^600,–,603  studies of patients with low cardiac