AnesthesiaWeb - Lit Reviews- Anesthesia-related cardiac arrest in children: Initial findings of the pediatric perioperative cardiac arrest (POCA) registry.

August 2000

Anesthesia-related cardiac arrest in children: Initial findings of the pediatric perioperative cardiac arrest (POCA) registry.
Morray JP, Geiduschek JM, Ramamoorthy C, Haberkern CM, Hackel A, Caplan RA, Domino KB, Posner K, Cheney FW. Anesthesiology2000; 93: 6-14

Return to the Current Literature Review Front Page

Commentary by Charles Cot�, M.D.

[see abstract below]

Morray and colleagues set out to define current contributions to cardiac arrests that occur in the perioperative period. [1] As outlined in their introduction, a number of previous studies have described children as being at increased risk for cardiac arrest compared with adults. Factors deemed to be important were young age, ASA physical status, emergency surgery, and underlying medical conditions. Each of these factors seems logical and the current study reaffirms those risk factors but also defines a change in underlying causes. This change in contributory causes, particularly those related to the respiratory system, may reflect in part the venue from which the majority of the data was collected and the improved monitoring currently available. I will attempt to describe potential strengths and weaknesses of the study

Strengths:

Data were collected prospectively
Data were collected from a large number of institutions (63) - most involved teaching programs
A strict definition was used (chest compressions or death)
A standardized data collection form was utilized with areas for individual comments.
Each report requested the reporter's assessment of contributory factors: anesthesia, surgery, disease, etc.
Outcome was evaluated on a standardized scale.
All cases were submitted anonymously.
All cases were reviewed by 3 members of the cardiac arrest registry and unanimity among the 3 reviewers was required for classification.
Only anesthesia related events were considered in the final analysis.

Findings:

52 percent of 289 submitted cases were deemed related to anesthesia.
There was a high inter-evaluator agreement on causation for the 150 cases that formed the final database.
There was an overall cardiac arrest rate of approximately 1.4 � 0.45 per 10,000 anesthetics.
Infants < one year accounted for 55% of cases with approximately one third in ASA 1 or 2 and two thirds in ASA 3-5 patients.
99% occurred while the patients were monitored with pulse oximetry and 97% with EKG, 95% with blood pressure, and 86% with capnography.
37% occurred during induction and 45 % during maintenance of anesthesia
Medication errors, particularly cardiac depression related to inhalation agents was the most common cause of arrest, particularly halothane.
Two were associated with sevoflurane.
Most anesthetic agent associated arrests occurred during induction.
Some children had underlying cardiomyopathy.
3/37 of these patients died and two survived with permanent neurologic injury.
Controlled ventilation and difficult intravenous access were contributory factors.
5 children arrested with the intravascular injection of local anesthetics.
10/18 cardiovascular related arrested occurred in children with underlying congenital heart disease.
Only 20% of arrests were related to the respiratory system with most related to laryngospasm or airway obstruction.

Weakness:

Reporting was voluntary, so we do not know the true incidence of cardiac arrest.
The veracity of the information was dependent upon the reporter, not on an independent review of the hospital records and interview of the patient care team.
There was a changing pattern of the use of inhalation agents with sevoflurane which was only coming into common use in the US toward the end of the study period.
As described by the authors we do not know the numerator or denominator for sevoflurane vs. halothane anesthetics.
Most cases were collected from teaching institution and therefore the data may not be reflective of the entire community providing care for children.

This initial analysis demonstrates a changing cause for cardiac arrests in children undergoing anesthesia. In the past, most cardiac arrests were felt to be related to difficulty in managing the airway whereas this study demonstrates drug-related causes (anesthetic agent and local anesthetics) as the cause of most arrests. Similar to other studies of adverse anesthetic events [2,3,4-8] most adverse events occurred in the very young patient where there is often difficulty in obtaining intravenous access, technical difficulties in maintaining a patent airway, a fragile balance between anesthetic agent induced myocardial depression vs adequate anesthesia to start an IV yet prevent laryngospasm, and relative hypovolemia due to excessive fasting. In addition at this young age some underlying congenital malformations or disease processes may not as yet have become manifest. In fact a very important and often unrecognized cause of cardiac arrest is underlying myocarditis in a child with a history of a recent upper respiratory infection. This contributory factor is likely never to be reduced since there is no way of making this diagnosis if the infant is asymptomatic. [9,10]

Perhaps the most important observation is that a disaster can strike even in the best of hospitals at the hands of anesthesiologists who care for children on a daily basis. Cardiovascular depression induced by anesthetic agents and intravenous injection of local anesthetics are in most cases preventable accidents, and this study points out the need for improved vigilance and perhaps better training in the use of these agents. This study states that "reviewers regarded cardiovascular depression by halothane to be a frequent cause of arrest in this group." Certainly this is true but the real question is, is it the drug or is it the people using the drug or is it some other factor?? If one examines the original MAC studies for both halothane and sevoflurane it is interesting to note that the incidence of hypotension with both agents was identical and even higher, in terms of percentages, in infants anesthetized with sevoflurane compared to infants anesthetized with halothane. [11,12] Although sevoflurane has been shown to cause less myocardial depression than halothane, this is primarily reflected in sophisticated measures of cardiac function. [13,14] The best study in infants [13] concluded that "sevoflurane decreased cardiac output less than halothane did in infants. The minor decrease in contractility was compensated by a greater decrease in SVS (systemic vascular resistance) and mainly by no change in heart rate compared with halothane in infants. Therefore, a wide epidemiologic study is needed to determine if sevoflurane appears safer than halothane for inducing anesthesia in infants." If atropine is administered to increase the heart rate, it is likely that there would be virtually minimal cardiac depression with halothane.

A more important factor may be a systems issue, i.e. the fact that only about 2.5 MAC of sevoflurane can be delivered compared with approximately 5 MAC with halothane vaporizers. In the current study the median concentration of halothane used was 2% (>2.5 MAC) but 14 patients received 3% halothane or greater! Likewise there was a high association with cardiac arrest and control of respirations. I venture to say that many of these arrests would have been avoided if the child was allowed to breathe spontaneously (controlled ventilation = death!), if only low concentrations were used until an intravenous line was inserted, if IM atropine had been administered in infants with potential difficult intravenous access or if atropine had been administered soon after establishing intravenous access.

In conclusion, these authors have performed a great service in better defining issues surrounding perioperative cardiac arrest associated with anesthesia but a lot more work must be performed to determine if the incidence of such events is different in an "all children's environment" compared with the "occasional children's environment", particularly for infants less than one year of age. We must determine if in fact the vaporizer is an issue and if it is the drug rather than the administrator of the drug that is the source of most problems. I am concerned that halothane, which is a wonderful anesthetic and very cost efficient, will be discarded without adequate scientific determination of cause and effect.