The benefits of youth are lost on the young cardiac arrest patient

Children and young adults tend to have reduced mortality and disability after acquired brain injuries such as trauma or stroke and across other disease processes seen in critical care medicine. However, after out-of-hospital cardiac arrest (OHCA), outcomes are remarkably similar across age groups. The consistent lack of witnessed arrests and a high incidence of asphyxial or respiratory etiology arrests among pediatric and young adult patients with OHCA account for a substantial portion of the difference in outcomes. Additionally, in younger children, differences in pre-hospital response and the activation of developmental apoptosis may explain more severe outcomes after OHCA. These require us to consider whether present practices are in line with the science. The present recommendations for compression-only cardiopulmonary resuscitation in young adults, normothermia as opposed to hypothermia (33°C) after asphyxial arrests, and paramedic training are considered within this review in light of existing evidence. Modifications in present standards of care may help restore the benefits of youth after brain injury to the young survivor of OHCA.

Outcomes by age in acute neurologic injuries tend to favor youth. When stratified by age, younger age groups tend to have improved rates of mortality and favorable neurologic recovery in traumatic brain injury (TBI) 1-3 and stroke 4-6 , including both intracerebral 7,8 and subarachnoid 9 hemorrhages. Children and young adults have fewer comorbidities and more reserve by virtue of their youth. It is well known that vascular disease increases with each decade of life 10 such that one would expect far better cerebral and coronary blood flow assuming that the same quality of cardiopulmonary resuscitation (CPR) was administered. Age-predicted functional and electroencephalogram recovery from intracarotid amobarbital 11 is consistent with the fact that younger brains are either more resistant to the insult or faster to recover. After TBI, younger patients compared with older ones with a similar disability score at admission to rehabilitation exhibited increased recovery after 1 and 5 years 1 .
Indeed, this pattern of less mortality and disability with younger age is observed throughout critical care and medicine. Yet after out-of-hospital cardiac arrest (OHCA), outcomes are remarkably similar in both the short and long term when comparing pediatric and adult patients (Table 1). This remains the case in nations where pre-hospital termination of resuscitation efforts is not permitted ( Table 2). Survival to hospital discharge and good neurologic function at discharge and 1 month are all below 10% in most studies. It is only when one considers long-term outcomes that children appear to have a slight advantage in outcome.

Explanations for the lack of neurologic protection in younger out-of-hospital cardiac arrest
The consistent lack of witnessed arrests among pediatric OHCA, particularly in infants and preschool children, accounts for a substantial portion of the mortality (Table 3). In the case of infants, most deaths occur during sleep and are attributed to sudden infant death syndrome (most of the "Other etiology" in infants on Table 3) or suffocation 12-15 . In preschool-and early school-aged children, asphyxia from drowning, choking, or suffocation 15 again occurs often out of immediate sight of an adult. Indeed, having OHCA at home is an independent predictor of worse outcome in some OHCA studies 15 . In older children (for example, adolescents) and young adults, OHCA is often unwitnessed as it is the result of suicide or substance abuse with overdose 12,13,15-17 . A recent report from the Resuscitation Outcomes Consortium (ROC) documents the fact that increasing overdose deaths tend to occur in young adults and tend to be unwitnessed 16 .
Since it is uncertain when the precise time of CA occurs, the noflow times can be substantial. In our personal experience (since this is rarely reported), many of these children are noted to be cyanotic and lifeless by their parents when CPR is first begun, and epidemiologic reports confirm asystole as the most frequent presenting rhythm 12,13 . Young hearts fare much better than young brains as evidenced by return of spontaneous circulation (ROSC) rates of 30% or better 12,13 (despite the lack of a witness in most cases), which are comparable to that of witnessed adult non-shockable OHCA 18 . Infants were the only group in the large Japanese data set who had a lower incidence of ROSC 19 . This may reflect a somewhat enhanced capacity of the heart to tolerate ischemia 20 compared with the brain.
The fact that many arrests are indeed unwitnessed resulted in the observation that bystander CPR rates, though not dissimilar to those of other age groups, did not improve outcomes 12,13 . Presumably, CPR started by bystanders on younger patients is occurring later after the onset of no flow since a larger proportion of pediatric OHCA is unwitnessed (that is, the patient is found in no flow of uncertain duration). Bystander CPR initiated later in no flow likely lacks the benefits imparted by prompt bystander CPR in a randomized study of older adults 21 .
The second important factor explaining why pediatric and young adult OHCA patients do not fare better is the extremely high incidence of asphyxial or respiratory etiology arrests as opposed to cardiac etiology. This is particularly true after infancy (when many congenital defects account for cardiac death) through young adulthood (Table 3). Outcome also appears to vary on the basis of etiology, although this association is controversial. We previously found worsened neurologic outcomes and survival associated with non-cardiac etiology OHCA 22 , which was likewise noted in the Korean Hypothermia Network (KORHN) registry 23 , whereas the opposite association was noted by the ROC 24 . Cardiac etiology OHCA tends to present with ventricular fibrillation and ventricular tachycardia, whereas non-cardiac etiology CA, such as asphyxia, often presents with pulseless electrical activity or asystole as a first rhythm 25,26 . Shockable rhythms are well known to yield the best outcomes 23,24,27,28 , even in children 29 . In the recently reported pediatric hypothermia study (THAPCA-OHCA) 30 , asphyxial OHCA accounted for more than 70% of all OHCA, and shockable rhythms were present in fewer than 10% of cases. This is compared with incidences of approximately 24% shockable rhythm in adult OHCA 27,28 . Young adults, who tend to make up the majority of overdose-related OHCA, generally present with non-shockable rhythms 16 . Although it remains uncertain why asphyxial CA confers more severe outcomes, this observation has been consistently made in experimental comparisons of different forms of CA where insult time is matched 31,32 . Yet treatment of OHCA patients after ROSC is the same regardless of rhythm or etiology 33 ; the only difference is that coronary angiography is recommended in suspected cardiac etiology CA due to acute coronary occlusion 16 .
The low incidence of pediatric OHCA relative to adult OHCA also means that paramedics are unlikely to have much repetition with resuscitation annually. With most things in medicine, "practice makes perfect"; therefore, low incidence means little practice. This is confounded by the added complexity resulting from pediatric life support 34,35 , which demands variable dosing and equipment sizing based on age and uses different algorithms than adult basic and advanced life support 36,37 . It is no surprise that in anonymous surveys paramedics report far less comfort with their technical skills in pediatric resuscitation and far less experience 38 . Compliance with American Heart Association (AHA) guidelines for rate and depth is significantly improved in "adult size" adolescents compared with children who are 1 to 11 years old 39 achieving compliance rates comparable to that seen in adults within the same registry 40,41 .
In the past decade, the concept of developmental apoptosis has been identified and characterized 42 . It appears that the cellular and molecular machinery used for normal pruning of neurons prior to Data were extracted from multiple epidemiologic reports as cited. The denominator was emergency medical services-treated out-of-hospital cardiac arrest. In many cases, our age bands do not strictly conform to other studies; therefore, data may be repeated and actual age ranges are provided for each data element. The dotted line represents the separation of children and young adults (0-39 years old) from older adults (>40 years old). One study reported data in graphic form only; therefore, sample sizes are estimates (denoted by ≈). For each category, we calculated a mean across studies by pooling outcomes and denominators but excluding those that did not report samples size (denoted by *). NR, not reported. Data were extracted from a subset of the epidemiologic reports in Table 1 (as cited). The denominator was emergency medical services-treated out-of-hospital cardiac arrest. In many cases, our age bands do not strictly conform to other studies; therefore, data may be repeated and actual age ranges are provided for each data element. The dotted line represents the separation of children and young adults (0-39 years old) from older adults (>40 years old). One study reported data in graphic form only; therefore, sample sizes are estimates (denoted by  Mean = 58 (928) birth and in infancy can be activated by sustained exposure to some sedatives or anesthetics (or both) with use consistent with what might be encountered with operating room or pediatric intensive care unit management or both 43 . Impairment of neurogenesis may also be operating 44 . The result of these developmental phenomena is that in children apoptosis may be accelerated, exceeding what is normally seen as a result of global ischemia-reperfusion injury and thus precipitating a more severe injury. This mechanism could contribute to poorer-than-anticipated outcomes after cardiac arrest in infants, although the upper age limit for potential contribution of this mechanism remains to be determined 45,46 . However, the quantitative contribution of this molecular disadvantage in infants to outcome after ischemic insults to the brain remains to be defined.

Implications for management
The significant differences between the older adult and pediatric/ young adult populations are important issues to consider for management. Since 2008, the AHA has recommended chest compression-only CPR (COCPR) for adults who have an OHCA that is witnessed by a bystander 47 . This was based on pre-clinical and human observational data in adults that showed equivalent or improved outcomes when COCPR was used 48-53 . Pragmatically, COCPR is easier to teach and instruct (for 911 dispatchers) and results in fewer interruptions of CPR for rescue breaths as well as perhaps lower intrathoracic pressure (increasing coronary perfusion pressure). However, it must be noted that COCPR is ineffective in the pediatric population 29 . In this study, COCPR had similar outcomes to standard CPR only among pediatric patients with arrests of cardiac origin, a distinct minority (Table 3).
The AHA's call to action is intended "for bystanders [and] does not apply to unwitnessed cardiac arrest, cardiac arrest in children, or cardiac arrest presumed to be of non-cardiac origin", nor does it apply to paramedics who can provide full CPR 47 . In practice, however, this distinction is not conveyed to the public or appreciated by 911 operators; we routinely care for asphyxial OHCA patients who have received COCPR. For children who had OHCA from non-cardiac causes, full CPR with rescue breathing yielded higher odds of ROSC, 1-month survival, and neurologically intact survival 29 . When Ogawa and colleagues examined this issue in the same national database (Japan), they found a similar trend to better outcomes with full CPR (versus COCPR) in the 20-to 39-yearold cohort, which was fully explained by better outcomes in the non-cardiac etiology subset 51 . Given the high prevalence of drug overdoses for arrest etiology in adolescents and young adults 15,54 , the now-dominant practice of instructing COCPR needs to be questioned if not modified to teach standard CPR in the younger population. Given that it is often easiest to teach a single approach and that outcomes from arrests of cardiac origin were not different in standard CPR versus COCPR, it would seem preferable to teach standard CPR for use across the full pediatric/young adult setting.
Differences in the characteristics of OHCA among patient populations are important when considering therapeutic hypothermia following OHCA. Two trials involving adult patients showed that  58 . Given the existing data, one must question whether pediatric and young adult OHCA survivors, particularly those known to have had non-cardiac etiology OHCA, would not benefit from a lower temperature goal. When temperature strategies are compared 30,57 , it would appear that the potential benefits of cooling asphyxial arrest survivors outweigh the risks. Hypothermia has been suggested in pre-clinical studies to have specific protective effects against developmental apoptosis and this may also contribute to its efficacy in newborn asphyxia 60 .
With regard to paramedic performance in the setting of pediatric arrests, recent studies have demonstrated the challenge for paramedic providers in the setting of OHCA in children given how rarely they are encountered in routine paramedic practice 60 . However, simulation approaches applied systematically across multiple paramedic systems are being evaluated, as is the analysis of the specific errors that are most frequently observed in this setting 61 .

Conclusions
Given the unique facets of CA in children and young adults and the emerging importance of the opioid epidemic, we believe that a careful re-examination of our current care in regard to resuscitation training and temperature management is vital and that new investigation is crucial to optimizing outcomes in what should be a resilient population.

Competing interests
The authors declare that they have no competing interests.

Grant information
The author(s) declared that no grants were involved in supporting this work.

Open Peer Review Current Referee Status:
Editorial Note on the Review Process are commissioned from members of the prestigious and are edited as a F1000 Faculty Reviews F1000 Faculty service to readers. In order to make these reviews as comprehensive and accessible as possible, the referees provide input before publication and only the final, revised version is published. The referees who approved the final version are listed with their names and affiliations but without their reports on earlier versions (any comments will already have been addressed in the published version).

The referees who approved this article are:
Version 1