Practice in a specialized area such as pediatric anesthesia should be the least reliant on myths, yet myths stubbornly persist, even in the face of hard data. I'm going to try to debunk four such myths by laying out the facts and figures you need to make evidence-based practice decisions.

1 No fluids for at least two hours pre-op
The American Society of Anesthesiology published pre-operative fasting guidelines in 1999 that set forth the following minimal fasting periods:

• two hours for clear liquids,
• four hours for breast milk,
• six hours for non-human milk,
• six hours for infant formula,
• six hours for a light meal and
• eight hours for solids.1

We follow these protocols, the reasoning goes, because appropriate fasting will decrease the gastric fluid volume, which will decrease the incidence of pulmonary aspiration. Since a 1974 study, we've believed that the amount of acid aspirate needed to damage the lungs is 0.4 ml/kg.2 But consider that the study authors never said they supported 0.4 ml/kg - they only mentioned the number in the introduction, saying "our preliminary work in the rhesus monkey suggests that 0.4 ml/kg is the maximum acid aspirate that does not produce significant changes in the lungs." Is this really enough basis for practice?

A more recent investigation involving a primate model found that the maximal acid aspirate that will not cause damage to the lungs is 0.8 ml/kg.3 And a review of NPO pediatric perioperative studies reveals that the residual gastric fluid volume in children after two hours NPO of clear liquids before surgery ranges from 0.24?0.31 ml/kg to 0.66?0.79 ml/kg.4-5 Together, these data seem to indicate that 15 percent to 50 percent of patients would be at risk, and that the incidence of pulmonary aspiration would be quite high. All recent studies indicate, however, that the incidence of pulmonary aspiration under general anesthesia is quite low in infants and children.

When you consider the number of factors that have to fall into place for lung injury to occur, you start to realize how unlikely a complication it is: Not only must there be intake of liquids or solids, but there also must be active vomiting or passive regurgitation, followed by aspiration and then, potentially, lung injury. Let's examine this path.

• Fluid intake. Water is absorbed very quickly; 60 minutes after intake, minimal water remains (see "Food and Liquid Remaining in Stomach"). Perioperative clinical studies indicate equally minimal gastric fluid volume at two hours after water intake. The data of perioperative studies are less clear when formula, breast milk or rice cereal are discussed, but results of an ultrasonographic study by Sethi et al. demonstrates that the traditional NPO times might be unnecessarily long.6
• Solid intake. The illustration demonstrates that gastric emptying of solids is more involved and is dependent on a number of factors. Overall, one can conclude based on available data that it takes at least eight hours to evacuate a full heavy dinner (1,692 Kcal) and between four and six hours for light breakfast.
• Active vomiting or passive regurgitation. There is little data in the literature with regard to this issue. Plourde and Hardy have addressed the issue of passive regurgitation in a cat model, finding that the gastric volume required to produce regurgitation in cats under general anesthesia is 20.8?7.8 ml/kg and the range is 8 ml/kg to 41 ml/kg - far higher than the 0.8 ml/kg of gastric aspirate that is needed to cause lung damage.7 So if a patient were to have a gastric volume of 0.8 ml/kg, it's unlikely he'd passively regurgitate the contents, let alone enough of it.
• Aspiration. Next, a patient would need to actually aspirate the stomach content that was vomited or regurgitated. There are no data in the literature with regard to the likelihood of this event's occuring.
• Lung injury. Finally, even if the patient aspirates, that does not necessarily mean that significant morbidity will develop.8

There are no data to indicate that the NPO period for light or full meals should be challenged. But based on all the data, it seems to me that the ASA recommendations on fluids can be questioned. Specifically, clear liquids may be reduced to one hour; breast milk to three; non-human milk to three; and infant formula to four.

2 Automatically cancel for upper respiratory illness
Upper respiratory illness is the most common unpreventable reason pediatric surgeries are cancelled. Before discussing whether we should cancel or proceed with a case, we must be aware of the data available in the literature:

• Most of the available literature suggests that the incidence of minor complications (such as mild oxygen desaturation) and more potentially serious complications (such as bronchospasm, laryngospasm and respiratory failure), may increase in children who either have URIs or who have recently recovered from one.9-13
• One prospective, case-controlled study involving over 20,000 anesthetics, found that the incidence of respiratory complications increased twofold to sevenfold with the presence of a URI, and increased eleven-fold if endotracheal intubation was performed in children with URI. And the younger the patient, the higher the risk.14
• Another recent study of more than 1,000 children who were scheduled to undergo general anesthesia and elective surgery found that children with active or recent URI had significantly more episodes of breath holding, desaturation episodes (O2 under 90 percent), and a greater incidence of overall adverse respiratory episodes. Risk factors for development of respiratory complications included endotracheal intubation, history of prematurity, surgical procedure involving the airway, reactive airway disease and nasal congestion.15

If we don't cancel all cases involving URI, should we cancel all cases of a serious URI? Plus, how do you define a serious URI? There are no good lab tests you can use to decide whether to postpone surgery. It's best, then, to carefully assess children who are already at risk due to

• asthma,
• bronchopulmonary dysphasia,
• history of prematurity,
• age younger than 1,
• sickle cell disease or
• scheduled procedure involving the airway.

If two or more of the following are present, research suggests you should delay the case:

• temperature over 101?F,
• laryngitis,
• malaise,
• sore throat,
• sneezing,
• non-productive cough,
• congestion and
• rhinorrhea.16-17

So why not cancel all cases involving URI? Practical reasons. The average child gets six or seven URIs per year, each lasting seven days to 10 days. Further, evidence suggests airway reactivity is increased for at least seven weeks after a URI. Translation: In a year, that leaves only about nine weeks during which the average child is not suffering or recovering from a URI. Even if you cancel and reschedule for two weeks down the road, a child is likely to be developing another URI in the meantime.

We may need to be less conservative with children who are about to undergo procedures such as PE tube placement or T&A. These procedures are therapeutic and may directly impact the patient's respiratory well-being.

3 Premature infants can only be operated on at 60 weeks post-conceptual age
General anesthetics and sedative hypnotics can depress ventilatory drive and cause central apnea in infants less than 56 weeks post-conceptual age; former premature infants are especially at risk for anemia, bronchopulmonary dysplasia and post-operative apnea.18-20 Post-op apnea after general anesthesia is particularly important, as a significant number of patients are today being treated as outpatients, and apnea is a risk for about 24 hours post-op.

Apnea has been defined as anything from a lack of turbulent airflow under the nostrils to no breathing movements for at least 30 seconds. 18-20 (See "How to Handle the Risks of Obstructive Sleep Apnea" on page 46.) Accordingly, studies show the incidence of apnea in the PACU ranges between zero and 32 percent.

At what post-conceptual age is the child no longer at risk to develop this phenomenon? The answer varies based on the data examined and it ranges from 48 weeks to 60 weeks post-conceptual age. Children who are at risk to develop post-operative apnea must be admitted to the hospital for 24 hours after anesthesia for cardiorespiratory monitoring (see "Determining When to Operate on Preemies"). In addition, the 60-weeks mark is somewhat conservative, and some anesthesiologists would replace it with 48 or 52 weeks.

The question of when to perform semi-elective surgery, such as inguinal herniorrhaphy, in premature infants is controversial. Although inguinal herniorrhaphy may be considered an elective procedure, there is an increased risk of incarceration in these patients, making this not a purely elective procedure. We must frequently proceed in these cases and accept the risk of post-operative apnea. Purely elective surgery, however, is best postponed until the child reaches at least 52 weeks to 60 weeks post-conceptual age.

4 Parents should be present during anesthesia induction
Proponents of parental presence during induction of anesthesia tout benefits such as reduced need for pre-operative sedatives such as midazolam, avoidance of separation anxiety, increased compliance from the child and increased parental satisfaction. Critics say that parents' presence crowds the OR, disrupts the surgical routine, prolongs induction and stresses anesthesia should a complication occur. In addition, parents often have adverse reactions upon seeing their child lose consciousness.

Although early studies suggested reduced anxiety if parents were present during induction, all recent randomized controlled trials indicate that routine parental presence is not beneficial.21-24 It's important to note, however, that the design of a randomized controlled study does not reflect the practice of all anesthesiologists. Research interests should instead emphasize what parents actually do during induction of anesthesia, rather than simply on their presence. Some parental behaviors, such as criticism and commands, reassurance with empathic statements, apologizing and bargaining are associated with greater distress. In contrast, parental behaviors such as distraction methods (humor, toys, party blowers) and commands to engage in active coping are associated with decreased anxiety.25 Simply letting a parent into the OR without significant preparation may be counterproductive - you must implement some process control.

Finally, parental presence during induction of anesthesia is associated with important legal implications. In one case in which a mother was invited to accompany her son into an emergency treatment room, the mother fainted and suffered a head injury.26 In its verdict, the Illinois Supreme Court stated that a hospital that lets a non-patient accompany a patient during treatment doesn't have a duty to protect the non-patient from fainting. But if medical personnel invite the non-patient, then the hospital has a legal responsibility toward the non-patient. You might want to require that parents sign a written informed consent acknowledging the risk of being present during induction of anesthesia, and perhaps rethink bringing parents into the OR for induction.

Yes, but why?
We in healthcare talk all the time about "evidence-based practice" and "best practices" - yet even when hard data presents itself, we're reluctant to change the way we work. Instead, we too often rely on "the way we were taught" and anecdotal experiences, especially in sensitive areas, such as anesthesia and pediatrics. Why is that the case? I hope you're now asking that very question.